Parahippus brevidens (Marsh, 1874) Gidley, 1907

Stoneburg, Brittney E., Mcdonald, Andrew T., Dooley Jr., Alton C., Scott, Eric & Hohman, Charlotte J. H., 2021, New remains of middle Miocene equids from the Cajon Valley Formation, San Bernardino National Forest, San Bernardino County, California, USA, PaleoBios 38, pp. 1-10 : 6-7

PARAHIPPUS BREVIDENS GIDLEY, 1907

FIG. 5

Referred specimens —WSC 8914 upper left M1; WSC 8918 lower right p3; WSC 9968 highly worn lower molar. Occurrence —Cajon Pass, San Bernardino National Forest, San Bernardino County, California. Exact locality data are on file at the U.S. Forest Service and WSC. Subdivision Tcv5, Cajon Valley Formation ( Woodburne and Golz 1972, Morton and Miller 2003); middle Miocene, late Hemingfordian–middle Barstovian NALMA; Tcv5 (unit 5) spans approximately 16.5–14 Ma ( Liu 1990).

Description —A brachydont horse larger than Ar. mourningi but smaller than S. sumani is represented by several specimens that include a well-preserved upper left M1 (WSC 8914), a lower right p3 (WSC 8918), and a worn lower molar or premolar (WSC 9968) ( Fig. 5). WSC 8914 exhibits two characters of the genus Parahippus (following Osborn 1918: p. 74): 1) “protocone and hypocone relatively large as compared with the simple lophoid protoconule and metaconule”; and 2) “hypostyle prominent, subtriangular” ( Figs. 5, 6). According to the UCMP online database (https://ucmp.berkeley.edu/collections/databases/), material in the UCMP collection from the Cajon Valley Formation has been previously identified as Parahippus by Michael Woodburne. However, none of these teeth have been formally described as Parahippus . All Cajon Valley Formation specimens listed in the UCMP online database are, as of this publication, currently under the listing “Punchbowl Formation.”

WSC 8914 bears closest resemblance to Pa. brevidens . It exhibits complex plications on both sides of the metaloph (“metaloph ptychoid on both sides” [ Osborn 1918: p. 90]) ( Figs. 5, 6) in contrast with “ Protohippus ” avus ( Marsh 1874) (later assigned to the genus Parahippus [ Gidley 1907] or Desmatippus [ MacFadden 1998]) which lacks these plications entirely, and D. crenidens ( Scott 1893) , which exhibits them only on the mesial side ( Osborn 1918). The protoconule does not align with the protocone, resulting in the curvature of the mesial edge ( Osborn 1918). The metaconule is isolated, and resembles an “ear-shaped lobe” with interior cement with wear ( Osborn 1918). The tooth lacks a median ridge between the metacone and the paracone, and the metaloph connects with the hypocone, but not the hypostyle ( Osborn 1918).

Marsh (1874) named Protohippus avus and Anchippus brevidens , both from the Miocene of Oregon. However, Gidley (1907) transferred the type species of Anchippus , An. texanus ( Leidy 1868) , into Parahippus , thus subsuming the genus Anchippus and creating the new combination Parahippus brevidens . He also placed the species “ Protohippus ” avus into Parahippus , which later authors ( Osborn 1918, Bode 1933) continued. Downs (1956) argued that Pa. brevidens was synonymous with Pa. avus . However, he based this on nomenclatural arguments, and did not justify the synonymy with morphological characters.

Although we refer our specimens to Pa. brevidens , we acknowledge that the taxonomy of both Parahippus and Anchippus is in need of revision (see discussion by Al- bright 1999) but is beyond the scope of the present work.