Discussion

Data presented indicate that the Panaca caprine, Sinocapra willdownsi, is not a member of any living genus within the Caprini and most closely resembles the extinct caprine, S. minor of China. The holotype of Sinocapra willdownsi is shown to be an adult female of this relatively small genus within the Caprini that had small, straight, non-diverging, horn cores lacking torsion, and therefore, likely had simple horn sheaths. The metacarpal and metatarsal measurements of Sinocapra willdownsi indicate that the size of the metapodials is about average among the living caprines, and appears most similar in dimensions to Capra ibex, Pseudois nayaur, and Ammotragus lervia (Table 4, Figure 15).

The argument that the type and referred specimens of Sinocapra minor represent male individuals, is based primarily on the width of the pedicle base of the horn cores. These horn cores imply that S. minor had fairly robust but comparatively short, straight, non-diverging, non-torsioned horn sheaths (unlike those observed in the living caprines; see Figure 5). The horn cores of S. minor and S. willdownsi are similar in construction and unlike those of any living caprine. The data and this comparison also imply that Sinocapra exhibited little if any sexual dimorphism.

At this time little is understood about the description and characterization of S. minor from China. Comparisons between the two species of Sinocapra are complicated in that the only bones recovered in common between them are the horn cores, with the Chinese species determined here to be a male and the North American form a female. The slight size differences between them may be due to a minor sexual dimorphism within the genus, an unusual phenomenon within the Caprini yet common among the ‘Rupicaprini.’

Sinocapra willdownsi lived at least in the Intermountain Region of western North America between approximately 4.95 to 4.50 Ma. Sinocapra minor from the Mazegou Formation, Yushe Basin, China, represents a younger caprine that dates to the late Pliocene Yushean land mammal age (Teilhard and Piveteau 1930; Teilhard and Trassaert 1938; Chen 1991; Flynn et al. 1991; Tedford et al. 1991). The locality YS78 falls in magnetic chron C2An3n, about 3.5 Ma. Despite the few specimens of S. minor the different morphology, age difference of at least one million years, and isolated geography (China versus North American) permit separate species designation. This is not ideal but challenges the recovery of more caprines of Pliocene age in both northeastern Asia and northwestern North America. The differences between the two forms become greater if all specimens are considered the same sex. This would imply that S. willdownsi was smaller, and that the younger Yushe caprine developed into a larger taxon by the late Pliocene.

Habitat

Köhler (1993) determined that many morphological characters observed on various skeletal elements within the bovidae are of more ecological or paleocological use than of phylogenetic value. Köhler (1993) divided the bovids into occupants of three habitats: (A) humid-wooded, (B) dry, more open, and (C) mountainous terrain. Inhabitants of B and C are more similar to each other than they are to A. In the Köhler classification, the morphological characters observed on the metacarpal and metatarsal of Sinocapra willdownsi suggest that the Panaca caprine does not belong to an existing genus of Caprinae, and that it likely was not an inhabitant of open, mountainous terrain such as is inhabited by Ovis, Oreamnos, and most other caprines (not including Nemorhaedus or Capricornis). The phalanges indicate that, like many other caprines, Sinocapra willdownsi was more of a mountain climber than a deep forest, flat-land inhabitant. The length-width of the metatarsal and the width of the metacarpal indicate that the Panaca caprine had medium stature. Its feet were not as wide as those of the best cliff-climbers (Oreamnos) or as long (stilted) and wide as the wild sheep (Ovis ammon, O. canadensis, O. catclawensis). The distal articulations of the metacarpal and metatarsal suggest this same conclusion. The length of the extensor tendon insertions on the proximal end of the metatarsal is similar to those of a mountainous inhabitant, although the rugosity of these attachment areas is more similar to those found in forested habitats (Köhler 1993). The width of the distal end of the metacarpal and the presence of a distinct and somewhat narrow sulcus and sagittal groove on both metapodials is more diagnostic to a woodland habitat than to wide-open terrain (Köhler (1993)). Sinocapra willdownsi would seem to fit into the Köhler classification (1993, p. 74) of "Ubiquitouus Type A/B" – a group of bovids situated between the clearly wooded and the clearly open habitat.

Systematics and Evolution

Chen (1991) considered Sinocapra minor more closely related to the ‘goat’ (Capra) lineage within Caprini rather than to the ‘ovine’ (sheep, Ovis) linage. However, the Yushe Sinocapra horn cores clearly lack torsion and a transverse ridge (keel), as does S. willdownsi, typical characters within members of the ‘goat’ lineage. Therefore, we are of the opinion that Sinocapra exhibits more ovine characters and most likely lies within or adjacent to Ovis ancestry.

Ovis makes its first appearance in the Chinese fossil record at approximately 2.42 Ma (Ovis shantungensis; Matsumoto 1926), and first appears in Europe during an unspecified time within the middle Pleistocene (Crégut-Bonnoure 1992). The earliest Ovis in North America is from El Golfo, Sonora, Mexico (Shaw 1981), and possibly dates to 1.2 Ma (Irvingtonian NALMA). This mandibular specimen and its age assignment are in need of detailed appraisal. Ovis sp. is also reported from Porcupine Cave, Colorado, dating to approximately 0.6 Ma (Mead and Taylor 2004).

Given the known fossil record and biostratigraphic position of Sinocapra, Ovis, and Capra, it appears possible that Sinocapra was the progenitor of Ovis and possibly Capra, or, at least a sister taxon. Sinocapra clearly occurs at least one million years earlier in North America than either Ovis or Capra occur in Europe (Crégut-Bonnoure 1992). Ovis shantungensis and S. minor apparently co-existed in China (Teilhard and Trassaert 1938). No evidence indicates that the caprines evolved in North America, so the occurrence of Sinocapra at Panaca indicates an earlier immigration event, possibly at about the time of the Mahui-Gaozhuang hiatus after 6.0 Ma (Flynn et al. 1991; Tedford et al. 1991). If this is correct, then an early form of Sinocapra should have existed in China. It is also possible that an unknown and undescribed progenitor form of early caprine evolved in Asia, migrated to North America during the Hemphillian (as did Neotragocerus), and Sinocapra actually evolved on this continent. A subsequent immigration event during the Pliocene could have brought Sinocapra to Asia from North America.

Whether Sinocapra evolved in Asia from a yet-to-be-realized ancestor in the late Miocene and immigrated to North America, or evolved in North America, later to immigrate to Asia, it seems evident that S. willdownsi is the oldest representative of the genus and is North American. Sinocapra willdownsi appears to be the earliest member of the Caprini in North America and one of the earliest bovids in the New World; Ovis, Oreamnos, and the ovibovines all arrived sometime during the middle Pleistocene (Irvingtonian).

Various researchers presume Neotragocerus to be a ‘rupicaprine.’ We question this assumption and suggest that Neotragocerus is not a member of that tribe; it clearly is not a member of the Caprini. These hypotheses indicate the need for additional detailed study of the bovids of northeastern Asia and North America. It seems clear that much of the history of the caprine bovids of North America remains to be discovered, a history that was richer and more diverse than previously understood.