by Joshua Ludtke*1
Oreodonts make up an extinct group of small, medium and large hoofed mammals. They are among the most commonly represented mammals in more than 40 million years of the North American fossil record; among completely extinct groups, the oreodonts may be the most abundantly preserved group of fossil mammals. This abundance has allowed them, after their extinction, to spread across the globe. Since at least the 1840s, fossil collectors from around the world have visited the North American west to excavate oreodonts, and their fossils have ended up in both public museums and private collections worldwide (see Fig. 1).
What is an oreodont?
Oreodont is the informal term for any member of the taxon Merycoidodontoidea. Oreodonts are placental mammals; some fossils appear to show fetal oreodonts inside their mothers. More specifically, they are even-toed hoofed mammals, or artiodactyls. Oreodonts have been estimated to vary in mass by two orders of magnitude, with small species weighing around 2 kilograms and large species weighing up to 310 kilograms. Identifying oreodonts any more specifically than as artiodactyls is problematic. Different research teams have argued that oreodonts are related to three different modern groups within Artiodactyla: the pig and peccary group (Suina); the camel and llama group (Tylopoda); and the deer, goat, giraffe and mouse-deer group (Ruminantia). Since 2005, five different research teams independently studying how oreodonts are related to modern artiodactyls have come to four different conclusions (see Fig. 2), one of which is strongly contradictory to molecular evidence. This inability to find the same result seems to indicate that the methods being used are in need of some refinement.
Oreodonts were referred to as “ruminating hogs” by Joseph Leidy, a US vertebrate palaeontologist active in the mid to late nineteenth century who named several oreodont taxa and identified them as belonging to a unique group. This informal name combined four aspects of their morphology known at the time. The cheek teeth of oreodonts are similar to those of ruminant artiodactyls, but their front teeth are more like those of pigs and peccaries. Also like pigs, and unlike some ruminants, they have limbs with four digits to walk on, and no head appendages such as horns or antlers. Ruminating hogs is not a technically accurate term, but it does capture that their head and body morphology can be seen as a mosaic of some modern groups (see Fig. 3).
Oreodonts can also be distinguished by several other anatomical characteristics that are relatively primitive for artiodactyls. Early groups of oreodonts had relatively long tails, which were shortened in later genera. Many artiodactyls with teeth similar to those of oreodonts have a large gap, or diastema, in their tooth row to separate teeth responsible for biting off food (incisors and incisiform canines) and teeth responsible for chewing food (premolars and molars), and an ‘incomplete’ tooth row: some incisors and premolars do not grow. But oreodonts generally had a complete tooth row with a relatively small diastema.
Other aspects of oreodont morphology are similar to those of other ungulates, having developed through convergent evolution. Some groups of oreodonts showed a nasal retraction, with nose bones placed behind where they are usually found, similar to that of modern tapirs. In tapirs, this retraction is linked to the presence of a small proboscis or trunk, and so it can be assumed that some oreodonts also had small trunks, although this should be tested more thoroughly. Oreodonts often had small depressions in the skull called lacrimal and facial vacuities, which may indicate that many had a scent-producing organ in front of the eyes. This could have helped them to mark territory by rubbing their faces on rocks or trees, as modern deer do. Some oreodonts had digits that ended in a claw-like covering, which has led to speculation that they may have dug or climbed to acquire food or hide from predators. However, these oreodonts had very few anatomical traits in common with known modern digging and climbing mammals, so they do not seem to have been specialized for either activity. There have been a few oreodonts preserved in burrows, suggesting that they, similar to the modern warthog, may have used underground dens while raising their offspring. When present, the relatively small diastemata of oreodont jaws make room for the sometimes-enlarged upper canines and lower canine-like first premolars to fit between other teeth in the tooth row. Although sexual dimorphism has not been well studied in oreodonts, these teeth do appear to be part of it; males seem to have had more robust canines than females.
Where and when were oreodonts?
The relative strength of the North American terrestrial mammal fossil record from the Cenozoic Era (66 million years ago to the present) has led palaeontologists to divide the geological timescale into ‘ages’ based on the first and last appearances of North American fossil mammal taxa (see Fig. 4). These North American land mammal ages, NALMAs, are the primary way in which palaeomammalogists discuss relative time in the North American Cenozoic terrestrial record, particularly when talking about areas where fossils are not found alongside volcanic-derived rocks that can be given an absolute date through radiometric dating. Figures 4, 6 and 7 follow recently published calibrations between NALMAs and absolute time, but NALMAs are not tied to particular absolute ages, so these correlations may change as new ways to calibrate them are found.
Oreodonts are first known from the Uintan NALMA, approximately 46.6 million years ago, in the Lutetian age, part of the Eocene epoch in the Palaeogene period and their last appearance is from the Hemphillian NALMA, approximately 7.5 million years ago, in the Tortonian age, Miocene epoch and Neogene period. Their known fossil record is confined to four countries in North America: Canada, Mexico, Panama and the United States (see Fig. 5). They are most commonly found in rocks from the late Eocene period and early Oligocene period (roughly 34 million years ago) of the western United States, and are a common component of the White River Fauna of South Dakota, Nebraska and Wyoming, but their known spatial distribution includes a large extent of North America.
What did oreodonts do?
Among herbivorous mammals, multiple groups have transitioned from having low-crowned teeth to having high-crowned teeth. This allows these herbivores to survive a diet that wears down the teeth — usually one containing a lot of grasses and other near-the-ground plants. Oreodonts are one of these groups: their earliest Eocene forms have brachydont (low-crowned) teeth, indicating that they tended to browse, or eat bushes and other off-the-ground plants. Several lineages of oreodonts during the Oligocene and Miocene appear to have developed hypsodont (high-crowned) teeth independently as a selective pressure to a grazing diet. The degree of hypsodonty is slightly more complex than acting as an indicator of whether a mammal was primarily a browser or grazer; the openness of an environment, the quantity of sand and volcanics in the soil and the relative concentration in the diet of phytoliths, glass-like crystals that some plants grow for structural support, are all thought to make mammals more hypsodont. One research team has looked at the overall sharpness of teeth (mesowear) and the height of teeth (hypsodonty) in oreodonts from different time periods to see if there were long-term trends in either of these dietary proxies. Both measures showed trends towards more abrasive diets over time. Interestingly, however, the mesowear results showed a temporary drop in abrasiveness from the late Eocene to the early Oligocene; this has been interpreted as late Eocene oreodonts having a significant amount of fruit-eating in their diet, which was lost as fruit-bearing trees became less common in their preferred environments.
Unlike with several other groups of North American ungulates, such as horses and camels, there was no evolutionary trend for oreodonts to become cursorial — that is, for their limbs to become longer, relative to their body size, over time. This may have limited oreodonts’ ability to adapt to the increasingly open environment of North America during the Neogene. Oreodonts may have preferred to live in closed forests, which became less common over time.
Several modern groups of artiodactyls have complex stomachs that allow them to slowly and thoroughly digest plant material using foregut fermentation. Tylopods and ruminants add a further step by partially regurgitating their food after it has been processed by the stomach. They chew their cud, allowing saliva and chewing to break it down further, before re-swallowing it. This process, called rumination, is what the ruminant artiodactyls are named for. This is unintentionally confusing because camels and llamas, which are not ruminants, ruminate. Unfortunately, there does not seem to be any ways to tell from the skeleton that an animal has a complex stomach or ruminated, so the dietary anatomy and physiology of oreodonts and other fossil artiodactyls cannot be known for certain.
Two groups of researchers have looked at stable isotope data from two genera of late Eocene/early Oligocene oreodonts. Carbon-isotope results, which serve as a proxy for environmental openness, suggest that these two oreodonts probably inhabited open woodlands. Oxygen-isotope results serve as a proxy for drought-tolerance; these show that one sampled oreodont, which had claws, was more drought-tolerant than the other oreodont, which did not. This may indicate that the clawed oreodont used a form of foregut fermentation, or it may indicate that it tended to acquire water from its diet, rather than from drinking fresh water. A study of stable isotope data from Miocene oreodonts from Panama suggested, but could not strongly support, that these animals lived in a closed forest.
“Who’s who” among oreodonts?
The taxonomy (study of what constitutes valid species) and systematics (how are they related) of oreodonts are complex, because different groups of researchers have come to very different ideas. As a result, there are many more names for oreodont taxa than are probably valid. For the sake of convenience, this article will use the most recently published work on merycoidodontid oreodonts as its basis for which subfamilies and genera are valid, and will not discuss individual species. The subfamilies are listed in approximate order of their appearance in the fossil record; in Figures 6 and 7 their known biochronological record is indicated. The geographical record of these subfamilies was checked against recently published work and the author’s own observations. Verified locations are those for which other authors have positively identified specimens to that subfamily, reported locations are those where oreodont fossils are suspected to belong to a subfamily but not verified. The first two subfamilies have been put in the family Agriochoeridae, whereas the other 12 have been placed into the family Merycoidodontidae. It is generally accepted that the Agriochoeridae are ancestors of the Merycoidodontidae, but more research is needed. The relative sizes of the taxa are based on reconstructed mass: animals that would have been 1–10 kg are termed small; 10–100 kg are termed medium; and 100 kg or more are termed large.
Protoreodontines. Protoreodon and Diplobunops are the only genera in this group of small and medium-sized oreodonts, the earliest evolutionary radiation of oreodonts. The rest of the oreodonts probably descended from this group. Protoreodontines are known from Arkansas, California, Colorado, Montana, New Mexico, Oregon, South Dakota, Texas, Utah and Wyoming in the United States, Chihuahua in Mexico and Saskatchewan in Canada; they are a very common component of Uintan and Duchesnean aged terrestrial localities in North America.
Agriochoerines. The presence of claws in the only genus in this group, Agriochoerus, suggests a burrowing, digging or climbing function, but the skeleton shows no specialization for any of these behaviours. These medium and large oreodonts are known from California, Colorado, Montana, Nebraska, North Dakota, Oregon, South Dakota, Texas, Utah and Wyoming in the United States, Chihuahua in Mexico and Saskatchewan in Canada; they are not as common as protoreodontines, but still show a wide geographical range.
Aclistomycterines. Aclistomycter is a medium-sized, poorly known oreodont representing one of the earliest branches of the Merycoidodontidae but retaining several ‘primitive’ aspects in its head and teeth morphology. Aclistomycterines are known from Texas in the United States and Chihuahua in Mexico.
Oreonetines. Oreonetes and Bathygenys are both small oreodont genera that seem to be descended from smaller species of Protoreodon. They are known from Colorado, Montana, Nebraska, Texas and Wyoming in the United States, Chihuahua in Mexico and Saskatchewan in Canada.
Leptaucheniines. Limnenetes, Leptauchenia and Sespia are small and medium oreodonts with extremely hypsodont teeth — especially in the latter two genera — and have been suggested to be adapted to arid environments. They are known from California, Colorado, Montana, Nebraska, North Dakota, South Dakota, Texas and Wyoming, and have been reported from Arizona in the United States and Saskatchewan in Canada.
Miniochoerines. Miniochoerus is a medium-sized oreodont similar in skull shape to the larger merycoidodontines. It is known from Colorado, Montana, Nebraska, North Dakota, South Dakota and Wyoming. Its absence from known Chadronian aged localities of Texas suggests that some kind of geographical or faunal competitive barrier limited it to more northern localities.
Merycoidodontines. Merycoidodon and Mesoreodon are the most commonly represented medium and large oreodonts in the fossil record, and are thus the stereotypical oreodonts. The merycoidodontines have been hypothesized to be ancestral to the seven later-appearing oreodont groups. They are known from California, Colorado, Florida, Idaho, Montana, Nebraska, North Dakota, Oregon, South Dakota, Texas, Washington and Wyoming in the United States, Oaxaca in Mexico and Saskatchewan in Canada; they are extremely common in the late Eocene and Oligocene North American terrestrial fossil record.
Eporeodontines. Eporeodon and Merycoides are medium and large genera that generally resemble merycoidodontines and are hypothesized to be ancestral to merychyines, ticholeptines, ustatochoerines and brachycrucines. They are known from California, Idaho, Montana, Nebraska, Oregon, South Dakota, Washington and Wyoming, and reported from Nevada and Texas.
Merychyines. Oreodontoides, Merychyus and Paramerychyus are the valid genera in this group. Although medium and large compared to earlier oreodonts, they were some of the smaller-bodied oreodonts of the late Oligocene and the Miocene (see Figure 8). Merychyines are known from California, Colorado, Florida, Montana, Nebraska, New Mexico, Oregon, South Dakota, Texas and Wyoming in the United States and Aguascalientes in Mexico, and reported from Arizona, Idaho, Kansas, Louisiana, Nevada and North Dakota in the United States, Oaxaca in Mexico and Saskatchewan in Canada. Merychyus is abundant in late Oligocene and early Miocene fossil assemblages, but many published references to it may have confused it with other oreodont genera.
Promerycochoerines. One of two groups of large oreodonts of the late Oligocene and early Miocene, promerycochoerines are represented by Desmatochoerus, Megoreodon and Promerycochoerus. Merycochoerines and promerycochoerines are hypothesized to be descended from large merycoidodontines, evolving similar cranial morphologies through parallel evolution. Promerycochoerines are known from California, Idaho, Montana, Nebraska, New Mexico, Oregon, South Dakota, Texas, Washington and Wyoming in the United States, and reported from Saskatchewan in Canada.
Merycochoerines. Hypsiops, Submerycochoerus and Merycochoerus are included in this group of large oreodonts. Merycochoerines, unlike promerycochoerines, seem to have some nasal-bone retraction, possibly indicating a proboscis. Known from Colorado, Montana, Nebraska, Oregon, South Dakota and Wyoming in the United States and Saskatchewan in Canada, and reported from Idaho and North Dakota in the United States.
Ticholeptines. Ticholeptus, Paroreodon and Phenacocoelus are the valid genera in this group of medium and large oreodonts which did not show as much development of high-crowned teeth as other lineages. Known from California, Colorado, Florida, Idaho, Montana, Nebraska, Nevada, Oregon, South Dakota, Texas and Wyoming in the United States, Aquascalientes and Oaxaca in Mexico, and Saskatchewan in Canada, and reported from Utah in the United States.
Ustatochoerines. Ustatochoerus and Mediochoerus are the genera in this group of large oreodonts that showed a slight trend towards nasal retraction. Ustatochoerines are known from California, Colorado, Kansas, Montana, Nebraska, Nevada, New Mexico, Oklahoma, Oregon, South Dakota, Texas and Wyoming in the United States.
Brachycrurines. This group of large oreodonts is represented only by the genus Brachycrus, which most strongly resembled modern tapirs in head shape, with the nasal bones retracted to just above and in front of the eye sockets. Brachycrus is known from California, Colorado, Idaho, Montana, Nebraska, Nevada, New Mexico, North Dakota and Wyoming in the United States and from Panama.
In addition to the above localities, oreodonts have also been found in British Columbia in Canada and Delaware in the United States.
Why are oreodonts important?
The extreme abundance of oreodont fossils has made them difficult to study comprehensively; any researcher looking to do so has to visit multiple museums across North America and observe thousands of specimens. In the twentieth century, four research teams embarked on such an endeavour. They all had different ideas on the boundaries between species and how to name groups, so they came up with four different answers to basic systematic questions: how many species of oreodonts are there, and how are they related?
In the twenty-first century, with the expansion of digital imaging and with much easier and faster communication between potential collaborators, it should be easier to understand the evolutionary relationships of oreodonts for palaeontological research. Once that is done, many other projects will become possible. The extreme abundance of oreodonts makes them good study organisms for anyone looking to research species boundaries or variation within species in the fossil record. The fact that they are found over a long time period and a wide area makes them useful for studying changes in tooth and head shape in response to the changing plants and climate of North America during much of the past 66 million years. Their extensive north–south distribution makes them an excellent test of similar changes at different latitudes in different parts of North America. Their apparent convergent and parallel evolutionary trends make them an interesting system to test ideas of character evolution and phenotypic evolvability, the ability of traits to evolve in different lineages which face similar pressures from their environments. Fossils of fetal and juvenile oreodonts known and suspected to have been found with presumed parents provide opportunities for studying development from the womb to adulthood in fossil mammal species.
The immense fossil record of oreodonts could be a powerful tool for studying evolution. It just needs some preliminary refinement and understanding to be as useful as it can be.
Coombs, M. C. 1983. Large mammalian clawed herbivores: a comparative study. Transactions of the American Philosophical Society 73, 1–96. (doi:10.2307/3137420)
Joeckel, R. M. 1992. Comparative anatomy and function of the leptaucheniine oreodont middle ear. Journal of Vertebrate Paleontology 12, 505–523. (doi:10.1080/02724634.1992.10011477)
Lander, E. B. 1998. Oreodontoidea. In Evolution of Tertiary Mammals of North America. Volume 1: Terrestrial Carnivores, Ungulates, and Ungulatelike Mammals (eds. C. M. Janis, K. M. Scott & L. L. Jacobs), pp. 402–425. Cambridge University Press.
Ludtke, J. A. 2007. Family Agriochoeridae. In The Evolution of Artiodactyls (eds. D. R. Prothero & S. E. Foss), pp. 151-156. John Hopkins University Press.
Mihlbachler, M. C. & Solounias, N. 2006. Coevolution of tooth crown height and diet in oreodonts (Merycoidodontidae, Artiodactyla) examined with phylogenetically independent contrasts. Journal of Mammalian Evolution 13, 11–36. (doi:10.1007/s10914-005-9001-3)
Schultz, C. B. & Falkenbach, C. H. 1968. The phylogeny of the oreodonts, Parts 1 and 2. Bulletin of the American Museum of Natural History 139, 1–498.
Stevens, M. S. & Stevens, J. B. 2007. Family Merycoidodontidae. In The Evolution of Artiodactyls (eds. D. R. Prothero & S. E. Foss), pp. 157–168. John Hopkins University Press.
Thorpe, M. R. 1937. The Merycoidodontidae: An extinct group of ruminant mammals. Memoirs of the Peabody Museum of Natural History 3, 1–428. (doi:10.5962/bhl.title.10230)
1Department of Biological Sciences, University of Calgary, Calgary, Alberta, Canada.