NPS units include an enviable slice of North American horse evolution. Quick background: in North America, terrestrial formations are often assigned to a North American Land Mammal Age (NALMA), provided there are suitable mammal remains. The earliest known equids showed up in the Wasatchian NALMA, during the early Eocene, and the horse lineage persisted in North America through the end of the Pleistocene (the Rancholabrean NALMA), whereupon it went kaput, with modern North American horses being imports. The NPS horse record begins in the Wasatchian at Big Bend National Park and Fossil Butte National Monument, and goes right up to the final bow (Rancholabrean Equus spp., as found at a number of NPS units). The record is a bit spotty in the pre-Chadronian Eocene and at the very end of the Miocene, but otherwise every NALMA is represented by either multiple sites or one major site.
The taxonomic diversity is pretty good, too. One caveat: this will probably surprise no one, but the equids, a group of popular and charismatic large mammals with historical importance to the paleontological study of evolution, have some taxonomic issues. Keeping that in mind, I would not be surprised if just about every North American horse genus is represented at least somewhere in the NPS. From my records, three equid species or ichnospecies are based on specimens known to have come from NPS units, and a further thirty are based on specimens that may have come from lands now within Badlands NP, John Day Fossil Beds NM, or Niobrara National Scenic River. The uncertainty with the latter figure isn't terribly surprising when you consider that they were almost all named between 1850 and 1904, a period of time when the evolution of horses was a major research concern and naming species was much more important than trifles like locality information.
Badlands NP, Hagerman Fossil Beds NM, and John Day Fossil Beds NM are certainly the most famous NPS units for producing horses, but Niobrara NSR is a notable "dark horse", if you'll forgive the turn of phrase, and Big Bend NP and Death Valley NP also include multiple productive formations over a long time frame. Thumbnail descriptions follow:
Badlands National Park
Badlands NP, within one of the classic North American Cenozoic fossil grounds, has produced abundant horse fossils in the Chadron and Brule formations, dating to the late Eocene and early Oligocene (Chadronian, Orellan, and Whitneyan). They belong to various species of the genera Mesohippus and Miohippus, as discussed in Benton et al. (2015). Scientific description of Badlands horses goes back to 1850 and Leidy's "Palaeotherium" bairdii, now known as Mesohippus bairdi.
|This is Miohippus intermedius from the Poleslide Member of the Brule Formation of Badlands NP, on display at the American Museum of Natural History.
Big Bend National Park
Big Bend NP shows up near the beginning of horse history with Eohippus fossils in the Wasatchian-age Hannold Hill Formation (a purported Black Peaks Formation record is more likely Hannold Hill float; Lehman et al. 2018). (I used Eohippus rather than Hyracotherium; those issues in horse taxonomy start all the way at the beginning and don't let up.) Horses also show up in the younger Eocene Chisos Formation, the Miocene "Banta Shut-In" Formation, and Pliocene–Pleistocene alluvium. However, although the record is long, it's not especially broad at this time.
Death Valley National Park
Horses at Death Valley NP are represented in late Eocene (Duchesnean-age Titus Canyon Formation), Pliocene, and Pleistocene rocks and sediments. The Eocene and Pleistocene records are of body fossils, but the Pliocene record is of equid tracks, which form the basis of the ichnogenus and species Hippipedia gyripeza Sarjeant and Reynolds 1999.
|Pliocene horse tracks observed in float at Death Valley NP. Sorry, no scale.
Hagerman Fossil Beds National Monument
Hagerman Fossil Beds NM is best known for the "Hagerman horse," originally named Plesippus shoshonensis (Gidley 1930) but now considered a synonym of Equus simplicidens. The nickname, though, has stuck, due to the abundant horse fossils found at Hagerman in the Pliocene Glenns Ferry Formation (Blancan). E. simplicidens is considered the oldest species of the genus Equus.
|An example of Equus simplicidens from Hagerman, on display at the Natural History Museum of Los Angeles County.
John Day Fossil Beds National Monument
The longest more-or-less continuous records of horses in the NPS is found at John Day Fossil Beds NM, where they have been reported from all four of the main bedrock sedimentary formations (Clarno, John Day, Mascall, and Rattlesnake, in ascending order, taking us from the early Eocene through the late Miocene; probably at least a couple of these will eventually become groups subdivided into formations). Horses are particularly well known from the John Day Formation, Cope, Leidy, and Marsh each naming a species or four, mostly in Anchitherium. By now, except for Marsh's Miohippus annectens, those which are still regarded as valid have since been moved to other genera, primarily Miohippus.
Niobrara National Scenic River
The Niobrara River exposes the Miocene-age Valentine Formation (Barstovian) and Ash Hollow Formation (Clarendonian), which contain a bewildering taxonomic abundance of horses. For example, a well-sampled site in the Valentine can produce eight or more distinct genera, as noted in Skinner and Johnson (1984). Like Badlands NP and John Day Fossil Beds NM, this is another area with a long history in the literature, with Leidy naming six species in 1858.
|Neohipparion from the Valentine Formation of Cherry County, Nebraska, on display at the University of Nebraska State Museum; I don't know exactly where it's from, but it's certainly appropriate for Niobrara NSR.
Benton, R. C., D. O. Terry, Jr., E. Evanoff, and H. G. McDonald. 2015. The White River Badlands: geology and paleontology. Indiana University Press, Bloomington and Indianapolis, Indiana.
Gidley, J. W. 1930. A new Pliocene horse from Idaho. Journal of Mammalogy 11:300–303.
Leidy, J. 1850. [Descriptions of Rhinoceros nebrascensis, Agriochoerus antiquus, Palaeotherium proutii, and P. bairdii.] Proceedings of the Academy of Natural Sciences of Philadelphia 5:121–122.
Leidy, J. 1858. Notice of remains of extinct Vertebrata, from the valley of the Niobrara River, collected during the exploring expedition of 1857, in Nebraska, under the command of Lieut. G. K. Warren, U.S. Top. Eng., by Dr. F. V. Hayden, Geologist to the expedition. Proceedings of the Academy of Natural Sciences of Philadelphia 10:20–29.
Lehman, T. M., S. L. Wick, H. L. Beatty, W. H. Straight, and J. R. Wagner. 2018. Stratigraphy and depositional history of the Tornillo Group (Upper Cretaceous–Eocene) of West Texas. Geosphere 14(5):2206–2244. doi:10.1130/GES01641.1.
Sarjeant, W. A. S., and R. E. Reynolds. 1999. Camelid and horse footprints from the Miocene of California and Nevada. San Bernardino Museum Association Quarterly 46(2):3–20.
Skinner, M. F. and F. W. Johnson. 1984. Tertiary stratigraphy and the Frick Collection of fossil vertebrates from north-central Nebraska. American Museum of Natural History Bulletin 178:215–368.