Non-technical Summary
The John Day region of Oregon preserves fossils from the Eocene through the Miocene, including many of the now-extinct hooved mammals known as oreodonts. Although there were previously as many as 34 different species of oreodonts recognized here, we use more recent analyses and techniques and redescribe these oreodonts as only nine separate species. We then use the extensive fossil collections at the John Day Fossil Beds National Monument to look at when exactly these nine species existed in the area and how many co-existed at one time and examine theories for their eventual local extinctions. The time with the highest number of co-existing oreodont species was 26 million years ago, when five species can be found in the fossil record.
Introduction
In a world of changing climate and declining diversity, it is increasingly important to understand how different environments and ecosystems have accommodated past upheaval (Louys, Reference Louys and Louys2012; McGuire and Davis, Reference McGuire and Davis2014; Kemp et al., Reference Kemp, Boville, Carneiro, Jacisin, Law, Ledesma, Meza, Shields-Estrada and Xu2023). Detailed records of change from the fossil record are therefore paramount to understanding past changing diversity. The John Day region in eastern Oregon preserves an exceptionally well-fossilized record of the change in vertebrate animal populations from the Eocene through the Miocene (Fremd et al., Reference Fremd, Bestland and Retallack1994; Albright et al., 2008; Mohr et al., Reference Mohr, Famoso, Samuels, Laib and Schmitz2025; Samuels et al., In Review, This Issue). One prominent difference between the modern and fossil ecosystems is the change in large herbivore diversity, particularly artiodactyls. Today, only four artiodactyl species are found in the John Day region: Odocoileus hemionus (Rafinesque, Reference Rafinesque1817) (mule deer), Antilocapra americana (Ord, Reference Ord1815) (pronghorn antelope), Cervus canadensis (Erxleben, Reference Erxleben1777) (elk), and Ovis canadensis (Shaw, Reference Shaw1804) (bighorn sheep) (National Park Service, 2025). Yet the paleontological record presents a far different picture of artiodactyl diversity. The Eocene–Miocene John Day Formation records somewhere between 32 and 36 different artiodactyl species from 14 different families, in addition to other large herbivores such as rhinos and horses (Fremd et al., Reference Fremd, Bestland and Retallack1994; Lander, Reference Lander1998; Albright et al., 2008; Fremd, Reference Fremd2010; Famoso, Reference Famoso2017; Retallack and Samuels, Reference Retallack and Samuels2020; Samuels et al., In Review, This Issue). Although recent research has lowered these estimates in certain groups such as hypertragulids (Famoso and Jewell, Reference Famoso and Jewell2024; Emery-Wetherell et al., Reference Emery-Wetherell, Famoso and Samuelsthis issue), the difference remains stark, exceeding even that of the 12 species found across North America today (Burgin et al., Reference Burgin, Widness, Liphardt, Parker, Becker, Rochon, Huckaby and Zijlstra2025). One group (oreodonts, superfamily Merycoidodontoidea) was particularly diverse, with 34 holotypes named in the John Day region. Although fewer species are recognized today, both the primarily clawed family Agriochoeridae and the hooved family Merycoidodontidae are recognized and described from the John Day region (Lander, Reference Lander1998; Albright et al., 2008; Stevens et al., Reference Stevens, Prothero, Cleveland, Welsh, Marriott, Htun, Balassa, Olson, Watmore and Wheeler2023).
For a comparatively diverse modern analog, one must look further afield. Tropical forests in Africa show comparable diversity, with as many as 27 species of artiodactyl overlapping in range (IUCN, 2016). Yet the climate recorded in the Oligocene of Oregon was a subhumid and temperate environment rather than tropical (Bestland et al., Reference Bestland, Retallack and Swisher1997; Dillhoff et al., Reference Dillhoff, Dillhoff, Dunn, Myers and Strömberg2009). The extreme diversity of oreodonts and other artiodactyls suggests that Eocene–Miocene forests in Oregon held a greater diversity of herbivores than equivalent modern environments are capable of supporting. However, the rationale for this diversity may lie not only in overinflated taxonomy (Emery-Wetherell et al., Reference Emery-Wetherell, Famoso and Samuelsthis issue) but also in important, small-scale temporal differences.
The Turtle Cove Member alone of the John Day Formation represents over five million years of nearly continuous deposition and fossil accumulation (Mohr et al., Reference Mohr, Famoso, Samuels, Laib and Schmitz2025). Preliminary biostratigraphy showed considerable overlap of oreodont species during this time, but this has not been updated to reflect modern taxonomic standards (Albright et al., 2008). In this paper, we use a thorough character analysis (Emery-Wetherell et al., Reference Emery-Wetherell, Famoso and Samuelsthis issue) as our base for a revised systematic paleontology of John Day oreodonts and use the collections at the John Day Fossil Beds National Monument to provide a precise and updated biostratigraphy.
Geological setting
We analyzed specimens from the Clarno Formation (53.6–42.7 Ma), John Day Formation (39.67–17.99 Ma), and Mascall Formation (16.44–13.56 Ma) (Bestland et al., Reference Bestland, Hammond, Blackwell, Kays, Retallack and Stimac1999; Maguire et al., Reference Maguire, Samuels and Schmitz2018; Mohr et al., Reference Mohr, Famoso, Samuels, Laib and Schmitz2025). These formations are predominantly volcanic ash and volcaniclastic material, single eruptions of which were quite broad and can be found in sites easily more than 30 kilometers apart (Bestland, Reference Bestland2002). Vertebrate fossils across the John Day region are found almost exclusively in paleosols derived from volcaniclastic claystone deposits, which are often very similar in lithology across great distances (Bestland et al., Reference Bestland, Retallack and Swisher1997; Retallack, Reference Retallack2004; Dillhoff et al., Reference Dillhoff, Dillhoff, Dunn, Myers and Strömberg2009; Fremd, Reference Fremd2010; Retallack and Samuels, Reference Retallack and Samuels2020). The fossil material we studied from these formations comes from multiple geographic regions with the same lithology. This includes the well-known fossil localities in the John Day River Basin, both on and off John Day Fossil Beds National Monument. We also examined and included abundant material from Logan Butte to the south (lower Turtle Cove Member), Sorefoot Creek to the west (lower Turtle Cove Member), and Lonerock to the north (Kimberly Member) (Hanson, Reference Hanson2000; Fremd and Whistler, Reference Fremd and Whistler2009; Fremd, Reference Fremd2010). Specimens in the collections at the John Day Fossil Beds often had precise locality information, allowing us to examine biostratigraphy in greater detail in places such as the lower and upper Turtle Cove members of the John Day Formation. Exact locality and stratigraphic information are associated with each specimen listed in Supplementary Data Set 1. For more information on the local geology, see Supplementary Data Set 2.
Methods
Terminology
We follow Bärmann and Rössner (Reference Bärmann and Rössner2011) for dental terminology and Emery-Wetherell et al. (Reference Emery-Wetherell, Famoso and Samuelsthis issue) for other oreodont-specific cranial and dental anatomy.
Systematics and taxonomic scope
Diagnostic descriptions of all John Day oreodonts may be found in the Systematic paleontology section. Specific treatment details varied by taxon.
Agriochoerus guyotianus Cope, 1878 (Cope, Reference Cope1878a), Hypsiops breviceps Lander, Reference Lander1998, Paroreodon marshi Thorpe 1921 (Thorpe, Reference Thorpe1921b), and Oreodontoides oregonensis Thorpe, 1921 (Thorpe, Reference Thorpe1921b) all have consistent taxonomic definitions within the John Day region. Although the exact scientific name accepted may vary, the general conclusion is that for each of these there is only a single species; as such, these taxa were merely identified in the sample.
Diplobunops kardoula Emery, Davis, and Hopkins, Reference Emery, Davis and Hopkins2016 and Ticholeptus zygomaticus Cope, Reference Cope1878b have been specifically identified in detailed work (Emery et al., Reference Emery2016; Maguire et al., Reference Maguire, Samuels and Schmitz2018); we simply confirmed these identifications and applied them to any new material.
We were unable to view the key material of Merycochoerus matthewi Hunt and Stepleton, Reference Hunt and Stepleton2004, but this identification was tentatively confirmed through the original description (Hunt and Stepleton, Reference Hunt and Stepleton2004) and recent new material (Machado et al., Reference Machado, Hopkins and Emery-Wetherell2019). See Systematic paleontology for more details.
Merychyus elegans (Leidy, Reference Leidy1858), Merychyus calaminthus (Jahns, Reference Jahns1940), Merycoides longiceps (Douglass, Reference Douglass1907), and Merycoides pariogonus (Cope, Reference Cope1884) have all been reported from the region previously (Lander, Reference Lander1998; Stevens et al., Reference Stevens, Prothero, Cleveland, Welsh, Marriott, Htun, Balassa, Olson, Watmore and Wheeler2023), but these publications do not report specific specimen numbers. One recent publication lists two specimen numbers of Merycoides pariogonus but does not list reasons for identification, and we were unable to view the specimens to confirm this identification (Retallack and Samuels, Reference Retallack and Samuels2020). Merycoides pariogonus was also recently declared a nomen dubium (Stevens et al., Reference Stevens, Prothero, Cleveland, Welsh, Marriott, Htun, Balassa, Olson, Watmore and Wheeler2023). We suspect reports of Merycoides are misidentifications of Eporeodon occidentalis Marsh, Reference Marsh1875, and the Merychyus are misidentifications of Paroreodon; they have each been treated as discussion points under that taxon.
Promerycochoerus and Eporeodon were given a thorough character analysis and taxonomic assessment by Emery-Wetherell et al. (Reference Emery-Wetherell, Famoso and Samuelsthis issue). All John Day material is identified according to the findings of that paper.
Identification protocol
We identified material in the collection at the John Day Fossil Beds National Monument (JODA) predominantly during summer 2014, with follow-up work and reassessment later. Because oreodont taxa are identified almost exclusively by cranial material, all non-cranial material was removed with one exception—JODA 8881 is an isolated merycoidodontid astragalus from the Whitecap Knoll locality in the lower Big Basin Member. As only two oreodont specimens have been identified from this member and are both the same species, we have included it in our biostratigraphic revision and treated it as Eporeodon (see Systematic paleontology).
Unless an isolated upper dP4 or upper M1 belonged to an agriochoerid, it was removed from our analyses as these two teeth look very similar in merycoidodontid oreodonts and are easily confused. Neither tooth is particularly diagnostic, and it is possible that as a result their size classes could be confused for other oreodonts. For example, a Promerycochoerus dP4 is approximately the same size class as an Eporeodon M1 or M2. Agriochoerid teeth are easier to tell apart from merycoidodontid teeth, however, so they were retained.
Similarly, isolated lower p1s and upper canines were removed because the degree of sexual dimorphism measured in caniniform teeth in artiodactyls suggested it would be difficult to use size to reliably identify these teeth to genus or species (Emery, Reference Emery2016; Emery-Wetherell and Davis, Reference Emery-Wetherell and Davis2018).
Fine-scale locality information
Many specimens came from historical collections. These came with some combination of air photo coordinates, locality information, field notes, or GPS coordinates but were not precisely tied to specific members or geological layers (Kort and Famoso, Reference Kort and Famoso2020). We evaluated air photos and field notes to determine as precise a stratigraphic horizon as was possible for older collections. Recent work by National Park Service interns, seasonal employees, and partners created fine-scale geologic maps of key localities in the Painted Hills and Sheep Rock Units that assisted in the identification of the horizons from which specimens were collected. Some specimens were identifiable to member units (e.g., unit H of the Turtle Cove Member), while others had more generalized stratigraphic assignments. Precision and certainty for each specimen is indicated in the Locality and Time Bin columns of Supplementary Data File 1. Only specimens with certain stratigraphy were retained for our stratigraphic analyses.
Testing for preservation bias
We removed material that was non-diagnostic at the species or genus level, including specimens that were not sufficiently prepared for analysis. If certain localities or formations had worse preservation than others (or less thorough cleaning), it is possible that our analysis was skewed toward representing the biostratigraphy of formations and localities with better preservation.
To test this hypothesis, we evaluated the proportions between formations of our identified versus unidentified samples using a chi square test for goodness of fit in R (Cochran, Reference Cochran1952; R Core Team, Reference Team2019). A statistically significant result on this test would indicate that our analysis was biased toward or against particular formations.
Size change over time
One suggested trend in oreodonts is a change in size over time (Prothero, Reference Prothero1985; Lander, Reference Lander1988). We used recently derived age models to analyze the relationship between time and size for those specimens with precise locality information (Mohr et al., Reference Mohr, Famoso, Samuels, Laib and Schmitz2025). Eporeodon and Promerycochoerus had sufficient sample sizes across time periods for this calculation, yet most specimens were not complete, making it difficult to accurately test size changes with perfectly cross-comparable measurements. To compensate for broken or partial specimens, we evaluated change using the proportional size of each species’ dental and cranial characters. For each measured variable, we calculated the maximum measurement for that species, then calculated each individual measurement’s proportional size (e.g., ninetieth percentile of M1 length). This allowed us to conduct a linear regression on change in size through time with maximal sample size. Variables with fewer than five samples were removed.
Repositories and institutional abbreviations
Specimens examined in this study are deposited in the following institutions: Frick Collection at the American Museum of Natural History (AMNH F:AM), New York, New York, USA; Fossil Mammal Collection at the American Museum of Natural History (AMNH FM), New York, New York, USA; Field Museum of Natural History (FMNH), Chicago, Illinois, USA; Fossil collections at the University of Oregon Museum of Natural and Cultural History (UOMNH F-), Eugene, Oregon, USA; John Day Fossil Beds National Monument (JODA), Kimberly, Oregon, USA; Carnegie Museum of Natural History (CM), Pittsburgh, Pennsylvania, USA; the Princeton University collections (PU), which are now stored at the Yale Peabody Museum (YPM VPPU), New Haven, Connecticut, USA; other collections at the Yale Peabody Museum (YPM VP), New Haven, Connecticut USA; California Institute of Technology collection (CIT), now stored at the Natural History Museum of Los Angeles County (LACM-CIT), Los Angeles, California, USA; Museum of Comparative Zoology at Harvard (MCZ), Cambridge, Massachusetts, USA; Texas Memorial Museum (TMM), Austin, Texas, USA; Vertebrate Paleontology collection at University of Washington’s Burke Museum (UWBM VP), Seattle, Washington, USA.
Systematic paleontology
Class Mammalia Linnaeus, Reference Linnaeus1758
Order Artiodactyla Owen, 1848
Superfamily Merycoidodontoidea Thorpe, Reference Thorpe1937
Family Agriochoeridae Leidy, Reference Leidy1869
Agriochoerus Leidy, Reference Leidy1850
Type species
Agriochoerus antiquus Leidy, Reference Leidy1850 (ANSP 10665) from the Brule Formation in South Dakota.
Agriochoerus guyotianus Cope, Reference Cope1878a
Reference Cope1878a Agriochoerus guyotianus; Cope, p. 77.
Reference Cope1879 Coloreodon ferox; Cope, p. 375, p. 505.
Reference Cope1879 Coloreodon macrocephalus; Cope, p. 376.
Reference Cope1879 Merycopater guyotianus; Cope, p. 375.
Reference Cope1884 Agriochoerus trifrons; Cope, p. 561.
Reference Cope1884 Agriochoerus guyotianus; Cope, p. 563.
Reference Cope1884 Agriochoerus ryderanus; Cope, 566.
Reference Cope1884 Coloreodon ferox; Cope, p. 570.
Reference Cope1884 Coloreodon macrocephalus; Cope, p. 570.
Reference Thorpe1921a Agriochoerus bullatus; Thorpe, p. 115, figs. 2–4.
Reference Lander1998 Agriochoerus antiquus guyotianus; Lander, p. 408.
Reference Ludtke, Prothero and Foss2007 Agriochoerus guyotianus; Ludtke, p. 62.
Reference Albright, Woodburne, Fremd, Swisher, MacFadden and Scott2008 Agriochoerus guyotianus; Albright et al., p. 224.
Reference Albright, Woodburne, Fremd, Swisher, MacFadden and Scott2008 Agriochoerus macrocephalus; Albright et al., p. 228.
Holotype
AMNH FM 7401, a mandible fragment with an isolated m3.
Material
Collections from JODA, AMNH, and UCMP. Holotype included. Full materials list and associated localities in Supplementary Data Set 1.
Diagnosis
Agriochoerid oreodont similar in size to Diplobunops and Eporeodon and larger than Protoreodon. Toothrow length of approximately 62–100 mm (Lander, Reference Lander1998). Upper molars have rounded, bulbous styles unlike in Protoreodon. Molars lack paraconules, unlike in Protoreodon or many species of Diplobunops. Lacks the fully split labial cone and posterolingual conule of the P4 found in Diplobunops kardoula. Muzzle does not flare at the canines as it does in Diplobunops. Smaller than Agriochoerus maximus (Douglass, Reference Douglass1902) and Agriochoerus major (Leidy, Reference Leidy1856). Possesses flexion stops on the occipital condyles, unlike Eporeodon.
Occurrence
In the John Day region, Agriochoerus is found almost exclusively in the lower Turtle Cove Member. A single tooth with questionable locality information was found higher in the section. Agriochoerus is found in all generalized outcrops of the lower Turtle Cove Member, including the John Day Fossil Beds National Monument, Logan Butte, and Sorefoot Creek subregions of the John Day Formation.
Remarks
Our analysis found Agriochoerus only in the lower Turtle Cove Member. The genus disappears above the Picture Gorge ignimbrite save for a single specimen (JODA 4721). While the collection records suggest JODA 4721 was found at JDNM-26, Sutton Mountain (Slater’s Site), this specimen has no field notes or field number associated with it. We therefore consider this occurrence to be questionable. Further remarks on nomenclature are in Supplementary Data File 2.
Diplobunops Peterson, Reference Peterson1919
Type species
Diplobunops matthewi Peterson, Reference Peterson1919 (CM 2974) from the Uinta Formation in Utah.
Diplobunops kardoula Emery, Davis, and Hopkins, Reference Emery, Davis and Hopkins2016
Reference Hanson, Prothero and Emry1996 Diplobunops n. sp.; Hanson, p. 233.
Reference Lander1998 New genus F; Lander, p. 409.
Reference Retallack, Bestland and Fremd2000 Diplobunops sp.; Retallack et al., p. 81.
Reference Lander and Hanson2006 Agriochoerus matthewi crassus (in part); Lander and Hanson, p. 24, fig. 3.
Reference Emery, Davis and Hopkins2016 Diplobunops kardoula; Emery, Davis, and Hopkins, figs. 1–4.
Holotype
UOMNH F-30334, a compressed cranium from the Hancock Mammal Quarry of the Clarno Formation, Clarno, Oregon.
Material
Collections from UOMNH and UCMP. Holotype included. Full materials list and associated localities in Supplementary Data Set 1.
Diagnosis
Agriochoerid oreodont from the Hancock Mammal Quarry, identification as described by Emery et al. (Reference Emery2016). Toothrow length of approximately 80 mm.
Occurrence
Known only from the Hancock Mammal Quarry in the Clarno Formation of Clarno Unit, John Day Fossil Beds National Monument, Wheeler County, Oregon.
Remarks
Diplobunops kardoula is an endemic species of the genus found in a single locality. It was medium in size, approximately the size of Agriochoerus. This species is described in detail by Emery et al. (Reference Emery2016).
Family Merycoidodontidae Cook, Reference Cook1912
Eporeodon Marsh, Reference Marsh1875
Type species
Eporeodon occidentalis Marsh, Reference Marsh1875 (YPM VP 010142) from the John Day Formation in Oregon.
Eporeodon occidentalis Marsh, Reference Marsh1875
Reference Marsh1873 Oreodon occidentalis; Marsh, p. 409.
Reference Marsh1875 Eporeodon occidentalis; Marsh, p. 249.
Reference Bettany1876 Eporeodon occidentalis; Bettany, p. 272.
Reference Cope1884 Eucrotaphus jacksoni; Cope, p. 517.
Reference Cope1884 Eucrotaphus trigonocephalus; Cope, p. 514.
Reference Thorpe1921b Eporeodon bullatus; Thorpe, p. 104.
Reference Thorpe1921b Eporeodon condoni; Thorpe, p. 104, figs. 6–8.
Reference Thorpe1921b Eporeodon leptacanthus leptacanthus; Thorpe, p. 97.
Reference Thorpe1921b Eporeodon leptacanthus pacificus; Thorpe, p. 99.
Reference Thorpe1921b Eporeodon longifrons; Thorpe, p. 103.
Reference Thorpe1921b Eporeodon major; Thorpe, p. 103.
Reference Thorpe1921b Eporeodon occidentalis; Thorpe, p. 96, figs. 1–3.
Reference Thorpe1921b Eporeodon trigonocephalus; Thorpe, p. 101.
Reference Thorpe1937 Eporeodon bullatus (in part); Thorpe, p. 65, figs. 28–30.
Reference Thorpe1937 Eporeodon condoni; Thorpe, p. 66, figs. 31–33.
Reference Thorpe1937 Eporeodon leptacanthus; Thorpe, p. 70.
Reference Thorpe1937 Eporeodon longifrons perbullatus; Thorpe, p. 72, figs. 35–37.
Reference Thorpe1937 Eporeodon longifrons; Thorpe, p. 71.
Reference Thorpe1937 Eporeodon occidentalis; Thorpe, p. 81, figs. 41–43.
Reference Thorpe1937 Eporeodon pacificus; Thorpe, p. 83.
Reference Thorpe1937 Eporeodon trigonocephalus; Thorpe, p. 96.
Reference Thorpe1937 Eucrotaphus jacksoni; Thorpe p. 61.
Reference Schultz and Falkenbach1954 Pseudodesmatochoerus wascoensis; Schultz and Falkenbach, p. 210.
Reference Schultz and Falkenbach1968 Dayohyus trigonocephalus; Schultz and Falkenbach, p. 216, figs. 28, 29.
Reference Schultz and Falkenbach1968 Dayohyus wortmani; Schultz and Falkenbach, p. 217, figs. 28–30.
Reference Schultz and Falkenbach1968 Eporeodon davisi; Schultz and Falkenbach, p. 203, fig. 24.
Reference Schultz and Falkenbach1968 Eporeodon occidentalis; Schultz and Falkenbach, p. 201, figs. 24, 30.
Reference Schultz and Falkenbach1968 Eporeodon (Paraeporeodon) leptacanthus; Schultz and Falkenbach, p. 213, figs. 25–27, 30.
Reference Schultz and Falkenbach1968 Eporeodon (Paraeporeodon) longifrons perbullatus; Schultz and Falkenbach, p. 210, figs. 25–27, 30.
Reference Schultz and Falkenbach1968 Eporeodon (Paraeporeodon) longifrons; Schultz and Falkenbach, p. 207, figs. 25–27, 30.
Reference Schultz and Falkenbach1968 Eporeodon (Paraeporeodon) pacificus; Schultz and Falkenbach, p. 205, figs. 25–27, 30.
Reference Schultz and Falkenbach1968 Pseudogenetochoerus condoni; Schultz and Falkenbach, p. 159, fig. 16.
Reference Schultz and Falkenbach1968 Pseudogenetochoerus covensis; Schultz and Falkenbach, p. 161, fig. 16.
Reference Lander1998 Eporeodon occidentalis; Lander, p. 412.
Reference Lander1998 Eucrotaphus trigonocephalus; Lander, p. 412.
Reference Stevens, Stevens, Prothero and Foss2007 Eporeodon occidentalis; Stevens and Stevens, p. 163, fig. 12.4A.
Reference Stevens, Stevens, Prothero and Foss2007 Eporeodon pacificus; Stevens and Stevens, p. 163.
Reference Albright, Woodburne, Fremd, Swisher, MacFadden and Scott2008 Eporeodon bullatus; Albright et al., p. 224.
Reference Albright, Woodburne, Fremd, Swisher, MacFadden and Scott2008 Eporeodon occidentalis; Albright et al., p. 224.
Reference Stevens, Prothero, Cleveland, Welsh, Marriott, Htun, Balassa, Olson, Watmore and Wheeler2023 Eporeodon occidentalis; Stevens et al., p. 18, figs. 10, 11, 13, 14.
Reference Stevens, Prothero, Cleveland, Welsh, Marriott, Htun, Balassa, Olson, Watmore and Wheeler2023 Eporeodon pacificus (in part); Stevens et al., p. 21, figs. 15, 16.
Holotype
YPM VP 010142, skull with RP4-M3, LM3. Broken anterior to P3 alveolus. Zygomatic arches are missing.
Material
Collections from JODA, YPM, AMNH, UWBM, CM, and UCMP. Holotypes of E. occidentalis, E. socialis (Marsh, Reference Marsh1894), E. pacificus (Cope, Reference Cope1884), E. longifrons (Cope, Reference Cope1884), E. condoni (Thorpe, 1921) (Thorpe, Reference Thorpe1921b), E. meagherensis (Schultz and Falkenbach, Reference Schultz and Falkenbach1968), E. dickinsonensis (Douglass, Reference Douglass1907), E. helenae (Douglass, Reference Douglass1902), Dayohus trigonocephalus (Cope, Reference Cope1884), and Pseudogenetochoerus covensis (Schultz and Falkenbach, Reference Schultz and Falkenbach1968) included. Full materials list and associated localities in Supplementary Data Set 1.
Diagnosis
Medium-sized oreodont, with a skull length of 212 ± 22 mm and a tooth row length from P1 to M3 of 99 ± 10 mm (mean and standard deviation). Styles on the upper molars taper upward from the base, as in Promerycochoerus but unlike the straight, untapered molar styles of Merychyus, Paroreodon, and Oreodontoides. Auditory bullae are well rounded and inflated, unlike the flattened bullae of Merycoidodon but similar in shape and size to those of Agriochoerus and Promerycochoerus. Auditory bullae are variable in exact shape and may be slightly conical, laterally compressed, or nearly perfectly spherical. Possesses a large, round pre-orbital fossa that is deeper than in all other comparative taxa. Zygomatic arches are less prominently hooked upward than in Promerycochoerus. Zygomatic arches have well-defined, sharp, and thin outer edges of the zygomatic arch rather than the rounded and thick edges found in some specimens of Promerycochoerus superbus Leidy, Reference Leidy1870. Auditory meatus is elongated and may start as a thin ridge or transition abruptly into a wide trumpet shape but is not surrounded by the pterygoid process as in Promerycochoerus. Post-glenoid processes are thick and similar to those of Agriochoerus and Promerycochoerus but unlike the thin processes in Merychyus and Paroreodon. Differs from Promerycochoerus, Hypsiops, Merychyus, and Paroreodon in having large rather than small stylids on the lower molars. Differs from Merycoides in having small rather than large and rounded incisive foramina. Nasal retraction is only to P1, no deeper, unlike in Merycoides. Premaxillae are thin, similar to Merycoides but dissimilar to Promerycochoerus.
Occurrence
Found in Big Basin, Turtle Cove, and Kimberly members of the John Day Formation in Oregon (Grant, Wheeler, and Crook counties). Found throughout the geographic range of the John Day Formation, including in Logan Butte, Sorefoot Creek, and Lonerock areas.
Remarks
One additional postcranial specimen (JODA 8881, from the lower Big Basin Member) was referred here to this genus. This is a merycoidodontid astragalus approximately the same size and with similar morphology to the astragalus of JODA 16843, a partial Eporeodon occidentalis skeleton found in the Big Basin Member. While we predominantly excluded postcrania from this analysis, these are the only two specimens found in the Big Basin Member and were included despite that criterion. Given that JODA 8881 is only an astragalus, it may well represent a different species of mid-sized oreodont, but at the moment, there is no reason to separate it from Eporeodon. Further nomenclatural remarks are available in Supplementary Data File 2.
Hypsiops Schultz and Falkenbach, Reference Schultz and Falkenbach1950
Type species
Hypsiops brachymelis Douglass, Reference Douglass1907 (AMNH FM 9731) from the Renova Formation in Montana.
Hypsiops brachymelis Douglass, Reference Douglass1907
Reference Douglass1907 Ticholeptus breviceps; Douglass, p. 107, figs. 6, 7.
Reference Loomis1923 Ticholeptus petersoni; Loomis, p. 225.
Reference Thorpe1937 Ticholeptus brachymelis; Thorpe, p. 188.
Reference Schultz and Falkenbach1950 Hypsiops brachymelis petersoni; Schultz and Falkenbach, p. 118.
Reference Schultz and Falkenbach1950 Hypsiops brachymelis; Schultz and Falkenbach, p. 119.
Reference Schultz and Falkenbach1950 Hypsiops johndayensis; Schultz and Falkenbach p. 123.
Reference Lander1998 Hypsiops breviceps; Lander, p. 414.
Reference Stevens, Stevens, Prothero and Foss2007 Hypsiops brachymelis; Stevens and Stevens, Reference Stevens, Stevens, Prothero and Foss2007.
Reference Albright, Woodburne, Fremd, Swisher, MacFadden and Scott2008 Hypsiops sp.; Albright et al., p. 224.
Reference Stevens, Prothero, Cleveland, Welsh, Marriott, Htun, Balassa, Olson, Watmore and Wheeler2023 Eporeodon pacificus (in part); Stevens et al., p. 23, fig. 17.
Reference Stevens, Prothero, Cleveland, Welsh, Marriott, Htun, Balassa, Olson, Watmore and Wheeler2023 Hypsiops brachymelis; Stevens et al., p. 82, fig. 67.
Holotype
AMNH FM 9731, a skull, mandible, and partial skeleton from Montana.
Material
Collections from JODA and CIT evaluated. Holotype for Hypsiops johndayensis (Schultz and Falkenbach, Reference Schultz and Falkenbach1950) included. Full materials list and associated localities in Supplementary Data Set 1.
Diagnosis
Larger than Hypsiops breviceps. Mid-sized oreodont with a skull length of approximately 260 mm and a toothrow length of approximately 100 mm. Possesses a deep and short muzzle unlike Promerycochoerus or Eporeodon. Nasal retraction to approximately P2 or further forward, unlike in Merycochoerus. Lacrimal fossa larger and deeper than in Promerycochoerus, although typically less so than in Eporeodon. Auditory bullae pointed as in Merychyus and Paroreodon, unlike the rounded bullae of Promerycochoerus and Eporeodon. Auditory bullae are not as comparatively large as in Paroreodon. Possesses a lacrimal vacuity, as in Paroreodon.
Occurrence
Within the John Day region, Hypsiops is known only in the Kimberly Member.
Remarks
Unfortunately, Hypsiops in the region is known largely from two skulls and a jaw, CIT 3503 and AMNH FM 7787, both of which have very limited locality information. The suggested location for CIT 3503 may contain Haystack Valley Member in addition to Kimberly Member, and it is possible the range of this taxon is larger than we expect. However, it is still very rare in the John Day Formation generally. Even at the time it was described, only one other specimen was referred to this genus from the John Day basin. We found one additional specimen in the collection that we could clearly refer to Hypsiops, a partial mandible from the Kimberly Member (JODA 11564). Further nomenclatural remarks are in Supplementary Data File 2.
Merycochoerus Leidy, Reference Leidy1858
Type species
Merycochoerus proprius Leidy, Reference Leidy1858 (AMNH FM 14242) from the Runningwater Formation near the Niobrara River in Wyoming.
Merycochoerus magnus Loomis, Reference Loomis1924
Reference Loomis1924 Merycochoerus magnus Loomis, figs. 16, 17.
Reference Schultz and Falkenbach1940 Merycochoerus proprius magnus; Schultz and Falkenbach, p. 286, fig. 17.
Reference Lander1998 Merycochoerus matthewi; Lander, p. 413.
Reference Hunt and Stepleton2004 Merycochoerus magnus; Hunt and Stepleton, p. 51.
Reference Stevens, Stevens, Prothero and Foss2007 Merycochoerus magnus; Stevens and Stevens, p. 163.
Reference Stevens, Prothero, Cleveland, Welsh, Marriott, Htun, Balassa, Olson, Watmore and Wheeler2023 Merycochoerus magnus; Stevens et al., p. 79, fig. 58, 61.
Holotype
AMNH FM 14242, a full skull and mandible from northeast of Agate, Nebraska.
Material
Collections from JODA evaluated. Full materials list and associated localities in Supplementary Data Set 1.
Diagnosis
As described previously, a large oreodont with elongated fused premaxilla and nasal retraction (Hunt and Stepleton, Reference Hunt and Stepleton2004; Stevens et al., Reference Stevens, Prothero, Cleveland, Welsh, Marriott, Htun, Balassa, Olson, Watmore and Wheeler2023). Merycochoerus and Promerycochoerus are of similar size, larger than other John Day oreodonts. The auditory meatus is partially surrounded by the pterygoid process in both Merycochoerus and Promerycochoerus but not in other oreodonts.
Occurrence
Known in the John Day region only from the Rose Creek Member of the John Day Formation (Grant and Wheeler counties).
Remarks
Merycochoerus magnus is established in the Rose Creek Member by UCMP 76848, a partial skull with sufficient premaxilla to establish it had the long posterior fusion typical of this genus and enough dentition to place it in the size class of M. magnus. We were not able to examine this specimen. A recently recovered skull in the UOMNH collections adds more certainty to this identification (Machado et al., Reference Machado, Hopkins and Emery-Wetherell2019), but a full description of that material is forthcoming as part of a dissertation project. Additional large merycoidodontid dental material has been found in the Rose Creek Member, but the dental material is very similar to that of Promerycochoerus superbus; it is unclear at this time whether the two species were coeval.
Oreodontoides Thorpe, Reference Thorpe1921b
Type species
Oreodontoides oregonensis Thorpe, Reference Thorpe1921b (YPM VP 012329) from the John Day Formation in Oregon.
Oreodontoides oregonensis Thorpe, Reference Thorpe1921b
Reference Thorpe1921b Eporeodon (Oreodontoides) oregonensis; Thorpe, p. 107, figs. 11–13.
Reference Loomis1924 Merychyus (Oreodontoides) oregonensis; Loomis, p. 31.
Reference Thorpe1937 Oreodontoides oregonensis; Thorpe, p. 213.
Reference Schultz and Falkenbach1947 Oreodontoides oregonensis; Schultz and Falkenbach, p. 251.
Reference Lander1998 Oreodontoides oregonensis; Lander, p. 414.
Reference Albright, Woodburne, Fremd, Swisher, MacFadden and Scott2008 Oreodontoides oregonensis; Albright et al., p. 224.
Reference Stevens, Prothero, Cleveland, Welsh, Marriott, Htun, Balassa, Olson, Watmore and Wheeler2023 Oreodontoides oregonensis; Stevens et al., p. 31, fig. 26.
Holotype
YPM VP 012329, a cranium with C1–M3. Locality description of Turtle Cove, John Day Valley, Oregon.
Material
We evaluated material identified as Oreodontoides in various museum collections (JODA, YPM and UWBM), but most was referrable to Paroreodon or Eporeodon. Full materials list and associated localities in Supplementary Data Set 1.
Diagnosis
Smaller than all other John Day oreodonts, with toothrow length of 71 mm from P1 to M3 and a skull length from canines to the back of the occipital bone of approximately 182 mm. Possesses straight, non-tapering styles on the upper molars, as seen in Merychyus and Paroreodon. Muzzle is longer and thinner than in Paroreodon or Merychyus. Does not possess a lacrimal vacuity, unlike Paroreodon. Differs from Paroreodon and Merychyus in the relatively longer, thinner muzzle. The holotype possesses no frontal or sagittal crests, unlike in all other John Day oreodonts. The basicranial axis is flatter than in Paroreodon (155° versus ~130°), although this is estimated as much of the basicranium is missing on the holotype.
Occurrence
Endemic to the John Day Fossil Beds and found within the Kimberly Member and Haystack Valley Member.
Remarks
This taxon is comparatively rare, with only the holotype as a complete skull. It is unclear what exact size range is therefore expected in its dentition. While the specimens we have identified as Oreodontoides are much smaller than those of Paroreodon, the dentition is quite similar and a small Paroreodon and a large Oreodontoides could be confused. We also often found that Oreodontoides’ small size was underestimated; small juvenile Eporeodon were regularly misidentified as Oreodontoides within the collection.
Paroreodon Thorpe, Reference Thorpe1921b
Type species
Paroreodon marshi Thorpe, Reference Thorpe1921b (YPM VU 012425) from the John Day Formation in Oregon.
Paroreodon marshi Thorpe, Reference Thorpe1921b
Reference Thorpe1921b Paroreodon marshi; Thorpe, p. 109, figs. 14–16.
Reference Thorpe1921b Eporeodon trigonocephalus parvus; Thorpe, p. 101, figs. 4, 5.
Reference Thorpe1937 Eporeodon parvus; Thorpe, p. 177, figs. 12, 14.
Reference Schultz and Falkenbach1947 Oreodontoides (Paroreodon) marshi; Schultz and Falkenbach, p. 256.
Reference Schultz and Falkenbach1947 Oreodontoides (Paroreodon) stocki; Schultz and Falkenbach, p. 257.
Reference Schultz and Falkenbach1968 Epigenetochoerus parvus; Schultz and Falkenbach, p. 164.
Reference Lander1998 Merychyus elegans arenarum (in part); Lander, p. 14.
Reference Lander1998 Paroreodon parvus; Lander, p. 414.
Reference Stevens, Stevens, Prothero and Foss2007 Paroreodon parvus; Stevens and Stevens, p. 164.
Reference Albright, Woodburne, Fremd, Swisher, MacFadden and Scott2008 Paroreodon “advanced species”; Albright et al., p. 232.
Reference Albright, Woodburne, Fremd, Swisher, MacFadden and Scott2008 Paroreodon marshi; Albright et al., p. 224.
Reference Albright, Woodburne, Fremd, Swisher, MacFadden and Scott2008 Paroreodon parvus; Albright et al., p. 230.
Reference Stevens, Prothero, Cleveland, Welsh, Marriott, Htun, Balassa, Olson, Watmore and Wheeler2023 Paroreodon parvus; Stevens et al., p. 94, figs. 78, 79.
Holotype
YPM VP 012425, cranium with intact P4–M3. Historical collection, locality labeled as “the upper Oligocene Middle John Day Haystack Valley–Turtle Cove area.”
Material
Collections from JODA, UWBM, and YPM evaluated. Holotypes for Paroreodon marshi and Eporeodon trigonocephalus parvus included. Full materials list and associated localities in Supplementary Data Set 1.
Diagnosis
Small oreodont, larger than Oreodontoides, with skull length from the canines to the back of the occipital condyles of approximately 146 ± 7 mm and toothrow length from P1 to M3 of approximately 74 ± 4 mm (mean and standard deviation). Small lacrimal vacuity, restricted to the lacrimal bone. Auditory bullae are large and inflated, with widths similar to the maximum width of the occipital condyles, much larger than in species of Merychyus. These auditory bullae are pointed at their apex as in Merychyus and Hypsiops and unlike in Eporeodon. The hypocone and protocone of the upper molars are similar in shape and placement, giving the tooth a squarer appearance than in Eporeodon. Steep basicranial axis (~130°), steeper than the axis seen in Eporeodon (~155°–160°).
Occurrence
This genus and species is endemic to the John Day Formation. Material ranges from the lower Turtle Cove Member up through the Johnson Canyon Member (Grant and Wheeler counties). The best cranial remains were collected without precise stratigraphic information but are most likely from the Haystack Valley Member. One exception is UCMP 76607, which is from the Johnson Canyon Member.
Remarks
The holotype of Paroreodon parvus (YPM VP 012425) was initially named Eporeodon trigonocephalus parvus by Thorpe (Reference Thorpe1921b), who then named Paroreodon marshi (YPM VP 012425) as a new species and new genus several pages later in the same document. Later, both species were placed into Paroreodon (Lander, Reference Lander1998; Stevens and Stevens, Reference Stevens, Stevens, Prothero and Foss2007). The ICZN states that typically, there is no influence of position on which name should come first when evaluating precedence on two species described in the same work but that position priority should be given for designating the type species of a genus (International Commission on Zoological Nomenclature, 2020). Since Paroreodon marshi was established as the only member of the genus by Thorpe (Reference Thorpe1921b), it should be considered the type species and therefore YPM VP 012415 as the holotype. This is counter to previous taxonomic interpretations that place P. parvus as the type species and therefore YPM VP 012425 as the type specimen (Lander, Reference Lander1998; Stevens and Stevens, Reference Stevens, Stevens, Prothero and Foss2007). In addition, restoring P. marshi as the species name will no longer cause confusion with Protoreodon parvus, an agriochoerid oreodont. Further remarks are in Supplementary Data File 2.
Promerycochoerus Douglass, Reference Douglass1901
Type species
Promerycochoerus superbus Leidy, Reference Leidy1870 (AMNH FM 7431) from the John Day Formation in Oregon.
Promerycochoerus superbus Leidy, Reference Leidy1870
Reference Leidy1870 Oreodon superbus; Leidy, p. 111.
Reference Bettany1876 Merycochoerus leidyi; Bettany, p. 273, fig. 18.
Reference Bettany1876 Merycochoerus temporalis; Bettany, p. 272, fig. 17.
Reference Bettany1876 Oreodon major (in part); Bettany, p. 272.
Reference Bettany1876 Oreodon superbus; Bettany, p. 272.
Reference Cope1884 Meryochoerus superbus; Cope, p. 522.
Reference Cope1884 Merycochoerus macrostegus; Cope, p. 526.
Reference Merriam1901 Paracotylops superbus; Merriam, p. 277.
Reference Douglass1901 Promerycochoerus superbus; Douglass, p. 82.
Reference Matthew1901 Promerycochoerus superbus; Matthew, p. 101.
Reference Thorpe1921c Promerycochoerus (Desmatochoerus) curvidens; Thorpe, p. 241.
Reference Thorpe1921c Promerycochoerus inflatus; Thorpe, p. 235.
Reference Thorpe1921c Promerycochoerus latidens; Thorpe, p. 232.
Reference Thorpe1921c Promerycochoerus marshi; Thorpe, p. 135, figs. 92, 93.
Reference Thorpe1921c Promerycochoerus microcephalus; Thorpe, p. 239.
Reference Thorpe1937 Promerycochoerus superbus; Thorpe, p. 141.
Reference Schultz and Falkenbach1949 Promerycochoerus (Pseudopromerycochoerus) inflatus; Schultz and Falkenbach, p. 123.
Reference Schultz and Falkenbach1949 Promerycochoerus latidens; Schultz and Falkenbach, p. 102.
Reference Schultz and Falkenbach1949 Promerycochoerus macrostegus; Schultz and Falkenbach, p. 118.
Reference Schultz and Falkenbach1949 Promerycochoerus superbus; Schultz and Falkenbach, p. 107.
Reference Schultz and Falkenbach1954 Desmatochoerus curvidens; Schultz and Falkenbach, p. 182.
Reference Schultz and Falkenbach1954 Superdesmatochoerus lulli; Schultz and Falkenbach, p. 215, figs. 16–18.
Reference Schultz and Falkenbach1954 Superdesmatochoerus microcephalus; Schultz and Falkenbach, p. 215, figs. 16–18.
Reference Stevens, Stevens, Prothero and Emry1996 Merycochoerus minor (in part); Stevens and Stevens, p. 536.
Reference Lander1998 Merycochoerus superbus pinensis (in part); Lander, p. 413.
Reference Lander1998 Merycochoerus superbus superbus; Lander, p. 413.
Reference Albright, Woodburne, Fremd, Swisher, MacFadden and Scott2008 Merycochoerus minor; Albright et al., p. 224.
Reference Albright, Woodburne, Fremd, Swisher, MacFadden and Scott2008 Merycochoerus superbus; Albright et al., p. 224.
Reference Albright, Woodburne, Fremd, Swisher, MacFadden and Scott2008 Promerycochoerus macrostegus; Albright et al., p. 224.
Reference Albright, Woodburne, Fremd, Swisher, MacFadden and Scott2008 Promerycochoerus sp.; Albright et al., p. 224.
Reference Stevens, Prothero, Cleveland, Welsh, Marriott, Htun, Balassa, Olson, Watmore and Wheeler2023 Desmatochoerus curvidens; Stevens et al., p. 55.
Reference Stevens, Prothero, Cleveland, Welsh, Marriott, Htun, Balassa, Olson, Watmore and Wheeler2023 Megoreodon grandis; Stevens et al., p. 72.
Reference Stevens, Prothero, Cleveland, Welsh, Marriott, Htun, Balassa, Olson, Watmore and Wheeler2023 Mesoreodon minor; Stevens et al., p. 146.
Reference Stevens, Prothero, Cleveland, Welsh, Marriott, Htun, Balassa, Olson, Watmore and Wheeler2023 Promerycochoerus chelydra; Stevens et al., p. 68.
Reference Stevens, Prothero, Cleveland, Welsh, Marriott, Htun, Balassa, Olson, Watmore and Wheeler2023 Promerycochoerus leidyi; Stevens et al., p. 62.
Reference Stevens, Prothero, Cleveland, Welsh, Marriott, Htun, Balassa, Olson, Watmore and Wheeler2023 Promerycochoerus superbus; Stevens et al., p. 64.
Holotype
The original holotype has been lost. AMNH FM 7431 is considered the neotype (Schultz and Falkenbach, Reference Schultz and Falkenbach1949; Stevens et al., Reference Stevens, Prothero, Cleveland, Welsh, Marriott, Htun, Balassa, Olson, Watmore and Wheeler2023).
Material
Collections from JODA, UCMP, YPM, CM, and AMNH evaluated. Type materials of Promerycochoerus montanus pinensis (Schultz and Falkenbach, Reference Schultz and Falkenbach1949), P. thomsoni (Loomis, Reference Loomis1924), P. barbouri (Schultz and Falkenbach, Reference Schultz and Falkenbach1949), P. superbus Chelydra (Cope, Reference Cope1884), P. montanus Cope, Reference Cope1884), P. hollandi (Douglass, Reference Douglass1907), P. vantasselensis (Peterson, Reference Peterson1907), P. hatcheri (Douglass, Reference Douglass1907), P. grandis (Douglass, Reference Douglass1907), P. inflatus (Thorpe, 1921) (Thorpe, Reference Thorpe1921b), P. leidyi (Bettany, Reference Bettany1876), P. latidens (Douglass, Reference Douglass1907), P. marshi (Thorpe, 1921) (Thorpe, Reference Thorpe1921c), P. grinnelli (Koerner, Reference Koerner1940), and Desmatochoerus curvidens (Thorpe, 1921) (Thorpe, Reference Thorpe1921b) are included. Full materials list and associated localities in Supplementary Data Set 1.
Diagnosis
A large oreodont with a skull length of 311 ± 40 mm and a toothrow length from P1 to M3 of 150 ± 26 mm (mean and standard deviation). Differs from all other John Day Formation oreodonts in having large, hooked dimorphic zygomatic arches that are typically rounded on the edges. The zygomatic arches curve upward more than 45°, unlike in Eporeodon. The lacrimal fossa is a shallow depression that opens anteriorly into a hollowed connection between the nasals and the maxilla. There is only a small gap between the paraoccipital process and the post-glenoid process as the lateral edge wraps around and nearly hides the auditory meatus, unlike the visible auditory meatus of Eporeodon and Hypsiops. Labial cingula are typically present on the upper molars as in Hypsiops but unlike most Eporeodon. Presence of flexion stops as in Agriochoerus and the holotype of Hypsiops johndayensis and unlike in Eporeodon. Large, rounded incisive foramen, similar to that seen in Merycochoerus but significantly larger than the thin slits found in Eporeodon.
Occurrence
Within the John Day region, occurs in the upper Turtle Cove, Kimberly, and Haystack Valley members of the John Day Formation in Oregon (Grant and Wheeler counties).
Remarks
It seems as though Promerycochoerus shows considerable sexual dimorphism, a problem that many previous researchers have discussed (e.g., Peterson, Reference Peterson1914; Loomis, Reference Loomis1924; Schultz and Falkenbach, Reference Schultz and Falkenbach1949; Stevens et al., Reference Stevens, Prothero, Cleveland, Welsh, Marriott, Htun, Balassa, Olson, Watmore and Wheeler2023). In particular, the zygomatic arches show considerable modification and variation consistent with some form of display function. Further discussion of this point can be found in Emery-Wetherell et al. (Reference Emery-Wetherell, Famoso and Samuelsthis issue) and Supplementary Data File 2.
Ticholeptus Cope, Reference Cope1878b
Type species
Ticholeptus zygomaticus Cope, Reference Cope1878b (AMNH FM 8112) from the Deep River Formation in Montana.
Ticholeptus zygomaticus Cope, Reference Cope1878b
Reference Cope1878b Ticholeptus zygomaticus; Cope, p. 129.
Reference Cope1886 Merycochoerus obliquidens; Cope, p. 359.
Reference Douglass1903 Poatrephes paludicola; Douglass, p. 176.
Reference Hay1930 Promerycochoerus obliquidens; Hay, p. 785.
Reference Schultz and Falkenbach1941 Ticholeptus zygomaticus; Schultz and Falkenbach, p. 86, fig. 15.
Reference Downs1956 Ticholeptus obliquidens; Downs, p. 299.
Reference Downs1956 Oreodonta gen. indet. sp. indet.; Downs, p. 301.
Reference Lander1998 Ticholeptus zygomaticus; Lander, p. 415.
Reference McLaughlin, Hopkins and Schmitz2016 Merycoidodontidae indet.; McLaughlin et al., p. 9.
Reference Maguire, Samuels and Schmitz2018 Ticholeptus zygomaticus; Maguire et al., p. 32.
Reference Maguire, Samuels and Schmitz2018 Merycoidodontidae indet.; Maguire et al., p. 34.
Reference Stevens, Prothero, Cleveland, Welsh, Marriott, Htun, Balassa, Olson, Watmore and Wheeler2023 Ticholeptus zygomaticus; Stevens et al., p. 89, figs. 70–72.
Holotype
AMNH FM 8112, a partial skull and partial mandible from the Deep River Formation, Montana.
Material
Collections from JODA evaluated. Full materials list and associated localities in Supplementary Data Set 1.
Diagnosis
Mid-sized oreodont, with a toothrow length of approximately 96 mm and a skull length of approximately 196 mm (Stevens et al., Reference Stevens, Prothero, Cleveland, Welsh, Marriott, Htun, Balassa, Olson, Watmore and Wheeler2023). Incisors much smaller compared with premolars than in other John Day oreodonts. Molars are more hypsodont than in Eporeodon and Promerycochoerus and possess the same straight, even styles found in Merychyus. Molars are mid-sized, approximately the same size as Eporeodon.
Occurrence
Within the John Day region, found only in the Mascall Formation (Grant and Wheeler counties).
Remarks
Oreodont material from the Mascall Formation is less common than in the John Day Formation. Ticholeptus was first described in the John Day under the species name Merycochoerus obliquidens (Cope, Reference Cope1886), which was later named Promerycochoerus obliquidens (Hay, Reference Hay1930) and then again renamed Ticholeptus obliquidens (Thorpe, Reference Thorpe1937). The holotype, AMNH FM 8192, is an isolated jaw. This species has since been subsumed into Ticholeptus zygomaticus by most recent workers (Lander, Reference Lander1998; Maguire et al., Reference Maguire, Samuels and Schmitz2018; Stevens et al., Reference Stevens, Prothero, Cleveland, Welsh, Marriott, Htun, Balassa, Olson, Watmore and Wheeler2023). This species is known from the John Day region primarily from dental material, with some associated maxillary and mandibular components. We recognize previous unidentified merycoidodontid material from Hawk Rim and the Mascall Formation as larger individuals of this species (McLaughlin et al., Reference McLaughlin, Hopkins and Schmitz2016; Maguire et al., Reference Maguire, Samuels and Schmitz2018). Additional remarks and details are in Supplementary Data File 2.
Merycoidodontid ident.
Material
YPM VP 12420, a skull with I3–M3, missing zygomatic arches. Jaw is associated.
Occurrence
Likely lower Turtle Cove Member of the John Day Formation, but the published locality is “Turtle Cove” in general (Schultz and Falkenbach, Reference Schultz and Falkenbach1968).
Remarks
YPM VP 12420 is a cranium that does not quite fit. It has many, although not all, characteristics in common with Paroreodon and indeed has previously been referred to Epigenetochoerus parvus (Schultz and Falkenbach, Reference Schultz and Falkenbach1968), which we would here call Paroreodon marshi. The dentition is within the correct size range for Paroreodon and much smaller than Eporeodon. With few specimens of Paroreodon or Oreodontoides recognized to compare with, it is difficult to determine whether the lack of a lacrimal vacuity and the shape of the auditory bullae are sufficient to classify it as a different species or not. Further remarks are in the Supplementary Materials.
Results
Identification results and preservation bias
The John Day Fossil Beds National Monument (JODA) collection held a total of 719 specimens labeled as containing oreodonts from the Clarno, John Day, and Mascall formations. Within these, we identified 385 specimens of Paroreodon, Hypsiops, Oreodontoides, Eporeodon, Promerycochoerus, and Agriochoerus and confirmed identifications of Diplobunops, Ticholeptus, and Merycochoerus. We also examined an additional 40 specimens that included John Day holotypes and specimens from the Clarno Formation that were stored at other museums (see Supplementary Data Set 1).
Of the remaining 334 unidentified specimens, 179 had too little material to confidently identify, 103 were only postcranial material, 34 were on loan, on display, or had not been prepared, four turned out to be hypertragulids, and one was a piece of pumice. The remaining 13 specimens were either missed or may have been unavailable at the time of study.
Our chi square test for goodness of fit by formation was not statistically significant (p = 0.21). This confirms that preservation did not cause us to bias our analyses toward or against certain formations, instead equally representing the oreodont material from all formations.
Biostratigraphy and occurrences
Thirteen specimens had locality information that tied them to general, member-crossing stratigraphic positions such “Turtle Cove, General” or “upper Turtle Cove/Haystack.” These have been removed from the biostratigraphic analysis. One Agriochoerus specimen was identified from a locality hesitantly ascribed to the upper Turtle Cove or Haystack, but no field number or notes were associated with the specimen, and we suspect the locality was incorrectly assigned. Other than this specimen, the Picture Gorge ignimbrite marks the end of Agriochoerus.
Three oreodont taxa had brief (<2 Myr) temporal ranges (Fig. 1). These taxa were all tied to single geologic members or localities. Diplobunops kardoula is found only in the Hancock Mammal Quarry of the Clarno Formation, Hypsiops brachymelis is found only in the Kimberly Member of the John Day Formation, and Merycochoerus magnus is known only from the Rose Creek Member. Conversely, Eporeodon (14 Myr) and Paroreodon (9 Myr) had very long temporal ranges and were found in multiple members. The number of coeval oreodont taxa ranged from one to five, with the highest oreodont diversity coming from the Kimberly Member of the John Day Formation.
Revised biostratigraphic range of oreodonts in the John Day region. The black bars indicate the range and relative abundance, with darker sections indicating more material. One specimen of Agriochoerus was found in a site ascribed to the upper Turtle Cove or Haystack Valley Member of the John Day Formation, which is marked with an asterisk as this occurrence is unlikely assigned to the correct locality. Colored bars in the background indicate members and formations, and white indicates unconformities. Gray illustrations at the base show habitats as interpreted from paleosol and floral evidence, showing cooling and drying from subtropical rainforest to open grasslands over time (Bestland et al., Reference Bestland, Retallack and Swisher1997; Retallack, Reference Retallack2004; Dillhoff et al., Reference Dillhoff, Dillhoff, Dunn, Myers and Strömberg2009; Black and Samuels, Reference Black and Samuels2016). LTC = lower Turtle Cove; UTC = upper Turtle Cove; PGI = Picture Gorge Ignimbrite; Kim = Kimberly member; HV = Haystack Valley; BC = Bone Creek; JC = Johnson Creek; RC = Rose Creek.
Figure 1. Long description
The chart uses a horizontalx-axis representing time in millions of years, from 44 on the left to 14 on the right. The y-axis lists nine taxa from bottom to top: Diplobunops, Agriochoerus, Eporeodon, Hypsiops, Oreodontoides, Paroreodon, Promerycochoerus, Merycochoerus, and Ticholeptus.
Background colored vertical bars indicate geological units:
- Clarno (orange) from 44 to 39.5 mya.
- Big Basin (pink) from 39.5 to 37.5 and 33 to 32 mya.
- LTC or lower Turtle Cove (light blue) from 32 to 29 mya.
- PGI or Picture Gorge Ignimbrite (thin orange line) at 29 mya.
- UTC or upper Turtle Cove (light green) from 29 to 26.5 mya.
- Kim or Kimberly (tan) from 26.5 to 25.5 mya.
- HV or Haystack Valley (yellow) from 24 to 23 mya.
- BC or Bone Creek member (blue-gray) from 23 to 22.5 mya.
- JC or Johnson Canyon (pinkish-gray) from 22.5 to 21.5 mya.
- RC or Rose Creek (thin tan line) at 18 mya.
- Mascall (gray) from 16 to 14 mya.
- White spaces indicate unconformities during the Big Basin member, between the Kimberly and Haystack Valley members, and before and after the Rose Creek member.
Black horizontal bars represent species ranges:
- Diplobunops appears briefly at 40 mya.
- Agriochoerus spans 32 to 29 mya, with an asterisk at 28 mya.
- Eporeodon shows a long range from 39 to 25.5 mya.
- Hypsiops and Oreodontoides are concentrated between 26.5 and 23.5 mya.
- Paroreodon and Promerycochoerus span from 29 to 22 mya.
- Merycochoerus appears briefly at 18 mya.
- Ticholeptus spans 16 to 14 mya.
At the base, gray silhouettes illustrate habitat transitions: large tropical trees on the left (44 mya) transition to swampy lakeshores during the Big Basin member (39 to 32 mya) to smaller deciduous trees and eventually to sparse shrubs and grasses on the right as the climate dried (14 mya).
Size change over time
We did find some statistical significance for a decrease in Eporeodon relative size measurements over time (Fig. 2). This relationship was not statistically significant for dental material (p = 0.75) but was weakly significant for cranial material (p = 0.008, R2 = 0.02, slope = 0.02). This relationship seems to have been driven primarily by skull length (p = 0.006, R2 = 0.27, slope = 0.06) and length of the pre-orbital fossa (p = 0.04, R2 = 0.15, slope = 0.06).
Linear relationships between time and proportional size of different characters of Eporeodon occidentalis. Points are slightly jittered on the x axis to better show the density of the data. Point color reflects sample size of that specific variable across all time periods. For example, the length of the lower p3 has only eight observations, so it is light pink, while the length of the upper M1 has 49 observations and is black. Gray ribbon indicates the 95% confidence interval of the regression line. Only variables with five or more samples are shown. Colored bars in the background and abbreviations follow the format in Figure 1.
Figure 2 Long description
A multi-panel scatter plot with four quadrants titled All Measurements, Cranial, Dental, and Fossa Length.
Axes and Scale:
* The Y-axis represents Proportional Size ranging from 0.2 to 1.0.
* The X-axis represents time in Millions of Years, decreasing from 32 on the left to 26 on the right.
* A legend on the right indicates sample density using a color gradient from light pink (5 samples) to black (50 samples).
* Background shading indicates geological intervals: L T C (light blue), P G I (light green), U T C (light green), and Kim (light tan).
Data Trends:
* All Measurements (Top-Left): Shows a dense cluster of points between 31 and 30 million years. A blue regression line shows a slight linear decrease in size over time.
* Cranial (Top-Right): Data points are concentrated in the L T C and P G I zones. The blue regression line shows a moderate downward slope from 0.8 to 0.75.
* Dental (Bottom-Left): Displays the highest density of data points. The blue regression line is nearly horizontal, indicating stable proportional size around 0.83.
* Fossa Length (Bottom-Right): Contains fewer data points, primarily between 31 and 28 million years. A blue regression line shows a steep linear decrease from 0.65 to 0.5, with a wide grey shaded area representing the confidence interval.
We also found statistical significance and a weak relationship for a decrease in Promerycochoerus relative size over time (p = 0.07, R2 = 0.01, slope = 0.02), although we had fewer specimens and a shorter section of time to analyze (Fig. 3).
Linear relationships between time and proportional size of different characters of Promerycochoerus superbus. Points are slightly jittered on the x axis to better show the density of the data. Point color reflects sample size of that specific variable across all time periods. For example, the height of the auditory bullae has only five observations, so it is light pink, while the length of the upper M1 has 17 observations and is black. Gray ribbon indicates 95% confidence interval of the regression line. Only variables with five or more samples are shown. Colored bars in the background and abbreviations follow the format in Figure 1.
Figure 3 Long description
The four panels are titled All Measurements, Cranial, Dental, and Skull Length. The y-axis represents Proportional Size from 0.2 to 1.0. The x-axis represents Millions of Years from 29 to 23.
Background vertical bands indicate time periods: U T C in light green (29 to 26.5 million years), Kim in light brown (26.5 to 25.5 million years), H V in tan (24.2 to 23.2 million years), and B C in grey (23.2 to 23 million years).
Data points are colored by a Samples legend where light pink represents 5 and dark black represents 20.
* Top-left (All Measurements): A blue regression line shows a gradual linear decrease from approximately 0.85 at 29 million years to 0.75 at 23 million years. Dense clusters of points appear around 28 and 26 million years.
* Top-right (Cranial): Shows a similar downward trend with a blue line starting at 0.8 and ending near 0.7.
* Bottom-left (Dental): The regression line mirrors the All Measurements panel, starting at 0.85 and ending at 0.78.
* Bottom-right (Skull Length): This panel has fewer data points. The blue regression line shows a steeper decline from 0.9 at 28 million years to 0.65 at 23.5 million years, with a wider grey confidence interval shading.
Discussion
Size change over time
Several measurements of Eporeodon and all measurements of Promerycochoerus showed a statistically significant relationship with time (Figs. 2, 3). However, these changes were minimal and gradual. The size ranges almost completely overlapped despite their averages changing, and the gradational nature of these changes create no clear boundary to divide either genus into a smaller and a larger species. The coefficients of variation of the pooled Eporeodon sample remained within the expected range of modern taxa (Emery-Wetherell et al., Reference Emery-Wetherell, Famoso and Samuelsthis issue), supporting a single-species hypothesis. The relationship within Promerycochoerus also may be partially driven by low sample size in younger and older material, as the lack of full skull lengths demonstrates (Fig. 3). As a result, we do not recognize more than one species of Eporeodon or Promerycochoerus in the John Day region.
Biostratigraphy and ecological change
Within the Turtle Cove members of the John Day Formation, we were able to pinpoint precise stratigraphic positions for 210 specimens, allowing a comparison of their relative abundance (Fig. 4). At a specimen level, most oreodonts in the collection came from the lower Turtle Cove Member (68%). This is also the only period where both families of oreodont are represented. Species diversity peaks later, with five overlapping species in the Kimberly Member. This is coincident with the continent-wide peak in diversity of the family Merycoidodontidae recognized by other researchers (Stevens et al., Reference Stevens, Prothero, Cleveland, Welsh, Marriott, Htun, Balassa, Olson, Watmore and Wheeler2023). Locally, it coincides with a continued trend toward open habitats. Paleopedology shows the Kimberly Member preserves a heterogeneous mosaic of sagebrush scrubland, desert, and desert scrub environments (Retallack, Reference Retallack2004). It is possible the species diversity of oreodonts reflects this mosaic of habitats or is simply a byproduct of generally high diversity within the group overall.
Relative abundance of oreodonts in the Turtle Cove Member. Violin plots and points show abundance for oreodonts with fine-scale stratigraphic-level information available. Violin plots interpolate between points, so points are also provided to show exact placement of occurrences. These dots are transparent; darker dots have higher abundance in that stratigraphic division. Colored bars in the background show unit-level assignments.
Figure 4. Long description
The X axis represents time in Millions of Years Ago, ranging from 32 on the left to 27 on the right. The Y axis lists four taxa from top to bottom: Agriochoerus, Eporeodon, Paroreodon, and Promerycochoerus. The background is divided into vertical colored bands labeled at the top with letters A through K2.
* Agriochoerus: Shows a concentrated abundance between 31 and 29 million years ago, peaking in units E1 through E3.
* Eporeodon: Displays the longest temporal range, with occurrences from 32 to 28 million years ago. The highest abundance is indicated by wide violin peaks and dark dots between 31 and 29 million years ago, specifically in units D through F.
* Paroreodon: Shows a thin, consistent distribution with sparse occurrences at 31, 30, 29, and 27 million years ago.
* Promerycochoerus: Appears only in the later stages, starting around 29 million years ago after the Picture Gorge ignimbrite, in unit Gand increasing in abundance toward 27 million years ago in unit K2.
Darker dots indicate higher abundance at specific stratigraphic levels, while the white violin shapes represent interpolated abundance between data points.
The two small-bodied oreodonts Paroreodon and Oreodontoides are rare compared with other John Day oreodonts. No specimen of Oreodontoides in our study had highly specific stratigraphic information. Instead, these were referable to either the Haystack Valley Member or the Kimberly Member of the Lonerock region where stratigraphy is less well known and not yet cross-correlated. Similarly, although found from the lower Turtle Cove Member through the Johnson Canyon Member, 86% of Paroreodon was from above the Picture Gorge ignimbrite. Thus, small-bodied oreodonts made up very little of the oreodont faunal composition within the lower Turtle Cove Member (<2%) and some of the upper Turtle Cove Member (15%), and only at the Kimberly Member and above, as oreodonts became comparatively rare, did they become major components (28% or more).
The lower Turtle Cove Member is dominated by the mid-sized oreodonts Eporeodon (80%) and Agriochoerus (19%), although Agriochoerus is more common in lower levels. Eporeodon is a hooved terrestrial browser, but Agriochoerus retains claws and has been interpreted to hold a semi-arboreal browsing niche similar to that of red tree kangaroos (Coombs, Reference Coombs1983). It is possible that as the habitat began to open up, only the terrestrial Eporeodon was able to adapt.
Although little is known about Eporeodon’s specific niche, the presence of an enlarged pre-orbital fossa may provide some clues as to their behavior and ecology. Other modern artiodactyls that rely heavily on scent such as duikers and muntjaks are not often found in large groups (Bowland and Perrin, Reference Bowland and Perrin1995; Odden and Wegge, Reference Odden and Wegge2007). Instead these animals are predominantly solitary or in small family groups, establishing territories in forested regions (Newing, Reference Newing1994; Odden and Wegge, Reference Odden and Wegge2007). Scent is an important component of communication in solitary ungulates, to the extent both duikers and muntjacs form latrine areas in their territories like scent-based message boards (Barrette, Reference Barrette1977; Newing, Reference Newing1994; Odden and Wegge, Reference Odden and Wegge2007). Given the similar morphology, Eporeodon may have been a similarly solitary, scent-driven species. As the habitat began to open (Retallack, Reference Retallack2004; Retallack and Samuels, Reference Retallack and Samuels2020), Eporeodon became a smaller component of the oreodont fauna before eventually disappearing approximately 25 million years ago.
The Picture Gorge ignimbrite (29.042 ± 0.026 Ma) marks the disappearance of Agriochoerus and the arrival of the larger Promerycochoerus. A similar switch from Agriochoerus to Promerycochoerus has been noted in other Arikareean localities as well (Matthew, Reference Matthew1910). Promerycochoerus is a large-bodied oreodont that makes up 52% of the upper Turtle Cove Member oreodont fauna and 64% of the fauna from the Kimberly Member before declining in the Haystack Valley Member (14%) and becoming extirpated. Increased sexual dimorphism and the flashy, antler-like zygomatic arches of Promerycochoerus are similar to the adaptations found in many artiodactyls when competition for mates grows more intense in larger herd structures (Davis et al., Reference Davis, Brakora and Lee2011). Both larger herd structures and more elaborate headgear are also consistent with more open environments (Cabrera and Stankowich, Reference Cabrera and Stankowich2020). Herd structure and size in the fossil record are difficult to demonstrate. Although the John Day region represents long-term accumulation rather than mass deaths, associated pairs of Promerycochoerus have still occasionally been found (Schultz and Falkenbach, Reference Schultz and Falkenbach1949). Outside of the John Day region, Promerycochoerus have been preserved in clusters of up to seven associated skeletons with both juveniles and adults preserved together (Peterson, Reference Peterson1914; Loomis, Reference Loomis1933; Schultz and Falkenbach, Reference Schultz and Falkenbach1949). Whether these represent living herd behavior or taphonomic clustering remains to be tested.
In the John Day region, we see a general trend toward cooling and drying, with more open habitats over time (Bestland et al., Reference Bestland, Retallack and Swisher1997; Dillhoff et al., Reference Dillhoff, Dillhoff, Dunn, Myers and Strömberg2009; Retallack and Samuels, Reference Retallack and Samuels2020). This is reflected within the changing oreodont composition. In the lower Turtle Cove Member, the oreodonts are predominantly either arboreal or solitary, both adapted to live in more densely forested environments. After that, the arboreal species disappear, and the range of oreodont sizes and presumed life habits increases.
The high level of volcanism throughout the Turtle Cove Member may have contributed to the change in environment (Famoso et al., Reference Famoso, Davis, Samuels and Emery2014). Specifically, the fauna before and after the Picture Gorge ignimbrite appears to transition from closed to open habitats (Famoso, Reference Famoso2020). Carbon isotopes from teeth also suggest a change from closed-canopy to water-stressed habitat across the PGI boundary (Reuter et al., Reference Reuter, Wainwright, Hoffman, Clementz, Blumenthal and Hopkins2025). As the climate dried, large-bodied and small-bodied oreodonts grew in abundance while mid-sized oreodonts dwindled and then disappeared, a pattern also seen in oreodonts in the Great Plains regions (Cleveland, Reference Cleveland2021). This may relate to body-size partitioning, changes in ecological roles, or other factors. Further research will be needed to understand this change and eventual decline in oreodonts.
Conclusion
With our revised taxonomy, the John Day region preserves nine species of oreodont, seven in the family Merycoidodontidae and two in the family Agriochoeridae. The John Day region preserves no more than five species of oreodont at any one time. The highest biodiversity of oreodonts occurs in the Kimberly Member, which also preserves the greatest range of oreodont size classes.
Even with our taxonomic revision, the John Day region still contains unusually high artiodactyl diversity, especially during the Oligocene John Day Formation. At their peak, oreodonts alone exceeded the diversity of the four artiodactyl species found in the region today, to say nothing of the additional species of Achaenodontidae, Anthracotheriidae, Camelidae, Dromomerycidae, Entelodontidae, Hypertragulidae, Leptochoeridae, Leptomerycidae, Moschidae, Oromerycidae, and Tayassuidae also found in the area concurrently (Fremd, Reference Fremd2010; Retallack and Samuels, Reference Retallack and Samuels2020; Samuels et al., In Review, Reference Samuels, Schap, Famoso and HopkinsThis Issue). The oreodonts may be less diverse than expected, but the John Day still preserves an incredibly diverse assemblage that rivals any North American ecosystem known today.
Acknowledgments
The initial work on the project in 2014 was conducted during an internship through the Geological Society of America’s GeoCorps America program, which was funded by the National Park Service. Travel was funded by the Field Museum Research Grant from the Field Museum of Natural History in Chicago, the Theodore Roosevelt Grant from the American Museum of Natural History, the Ernst Mayr Grant from Harvard University, and the Thomas Condon Scholarship and Smith Scholarship from the University of Oregon. B. Hunt and E. Stepleton provided excellent stratigraphic and taxonomic advice. Access to and assistance in collections was provided by J. Chupasko (MCZ), R. Banasiak (FMNH), M. Brown (TMM), A. Atwater (MOR), E. Westwig (AMNH), C. Sagabiel (VPL), P. Holroyd (UCMP), G. Wilson (UWBM), and C. Conroy (MVZ). B. McHorse provided housing and assistance with photographs. K. Burns, T. Sorenson, M. Sorenson, and D. Houts, B. Houts, L. Houts, and I. Houts provided much-needed housing during museum trips. T. Fremd helped hunt down locality information for obscure specimens and sites. B. Lander and J. Ludtke provided invaluable taxonomic advice and clarification. A.E. Kort, M. Ruga, M. Bouwens, Y. Patel, S. Masser, S. Lundblad, E. Baker, and M. McKnight assisted with geologic mapping and digitization of analog specimen localities. K. Walters (UO), K. Grande (CWU), S. Canada (CWU), and E. Welsh (NPS) helped collect some initial data. E. Davis (UO), S. Hopkins (UO), S. Frost (UO), D. Warrick (OSU), and J. Roering (UO) contributed invaluable advice to early versions of this research project.
Competing interests
The authors declare no competing interests.
Data availability statement
Supplementary Data Set 1, Supplementary Data Set 2, and the Supplementary R Code used for analyses are available at https://figshare.com/s/43c58e2413a3bd15d942.
A selection of 3D models representing John Day oreodont species has been uploaded and made available for download with permission from the Yale Peabody Museum, John Day Fossil Beds National Monument, and Carnegie Natural History Museum. These are available at https://www.morphosource.org/projects/000786750.
Open access
