Amynodontidae

Amynodonts Temporal range:
Mounted skeleton of Amynodon, American Museum of Natural History
Scientific classification
Kingdom:	Animalia
Phylum:	Chordata
Class:	Mammalia
Order:	Perissodactyla
Superfamily:	Rhinocerotoidea
Family:	†Amynodontidae Scott & Osborn, 18831
Genera and subgroups
†Amynodon †Huananodon †Hypsamynodon †Rostriamynodon †Sharamynodon †Cadurcodontini Wall, 1982 †Amynodontopsis †Cadurcodon †Cadurcotherium †Procadurcodon †Zaisanamynodon †Metamynodontini Kretzoi, 1942 †Megalamynodon †Metamynodon †Paramynodon †Sellamynodon Possible amynodonts †Armania †Caenolophus †Gigantamynodon †Proeggysodon †Teilhardia †Toxotherium

Amynodontidae is an extinct family of perissodactyls related to modern rhinoceroses, known from the Eocene to Oligocene of North America, Europe, and Asia. They were among the dominant groups of herbivorous mammals in the Northern Hemisphere, and were most diverse in Asia. Amynodonts have been nicknamed "hippo-like rhinos"² and "swamp rhinos"³ due to their hippopotamus-like build, with massive bodies and short, robust limbs, as well as their tusk-like canine teeth.

Derived members amynodonts are generally grouped into the lightly built, terrestrial, and tapir-like cadurcodontines and the semiaquatic metamynodontines. Amynodonts increased in body size over the course of their evolutionary history, and ranged in weight from just over a hundred kilograms to close to three tonnes.

Skeleton and life restoration of Metamynodon, from Middle Eocene to Early Oligocene North America

Amynodonts have been nicknamed "hippo-like rhinos" since they had several hippopotamus-like features, such as massive bodies and short, robust limbs.² They were medium-sized to large animals.⁴ It was traditionally believed that the entire family was adapted to semiaquatic life, but different amynodonts are now known to have been adapted to different lifestyles.⁴ The derived metamynodontines had clear adaptations towards a semiaquatic lifestyle but the other group of derived amynodonts, the cadurcodontines, were more lightly built and probably similar to modern tapirs.⁴ Amynodon, not placed in either tribe, has been interpreted as a terrestrial browser that lived in open forest-savanna habitats,⁴ though isotope analyses of its teeth may also indicate a semiaquatic lifestyle.³

The most significant synapomorphy (shared unique trait) in the amynodont skull were preorbital fossae (grooves in front of the eye orbits). The fossae differed in size and position depending on lineage; in primitive amynodonts (such as Amynodon), there are large fossae in front of the orbits.⁴ In the more derived amynodont groups, the condition of the fossae becomes a distinguishing trait; metamynodontines have reduced fossae whereas cadurcodontines have fossae that extend medial to the orbit (towards the middle of the orbits).⁴ Amynodont skulls had sagittal crests, a feature shared with some other early rhinocerotoids, such as Hyrachyus.⁴

Amynodonts had greatly enlarged and tusk-like canine teeth.⁴⁵ This further differentiates them from other rhinocerotoids, who either have small canines or lack them entirely. When other rhinocerotoid groups developed tusks, they were typically developed from the incisor teeth.⁵ The fossil evidence suggests that canine size in amynodonts was a sexually dimorphic trait.⁴

Amynodonts had four toes on their manus (front feet) and three on their pes (hind feet),⁴ a primitive condition among perissodactyls.⁶

The first amynodont to be described was Cadurcotherium cayluxi in 1873, described by Paul Gervais from the Oligocene of France.⁵ The first known North American amynodont fossils were found in the Uinta Formation of Utah and described by Othniel Charles Marsh. In 1875, Marsh named the new species Diceratherium advenum based on a fossil skull. Marsh soon realized that the fossil was distinct from Diceratherium, a previously described true rhinoceros (Rhinocerotidae), and moved it to the new genus Amynodon, as A. advenus, in 1877.⁴

In 1887, the new amynodont genus Metamynodon was described based on fossils from the White River Group.⁴ Based on Metamynodon, William Berryman Scott and Henry Fairfield Osborn were able to establish several important family-level characteristics of the amynodonts, including the enlarged sagittal crest and the loss of P1 (the first premolar tooth).⁴⁵

The first discoveries of amynodonts in Asia were made in Oligocene deposits in Pakistan and in Eocene deposits in Myanmar in the 1910s and 1920s. In the 1920s and 1930s, amynodont fossils were also discovered in Mongolia and China.⁵

Marsh believed Amynodon to be a rhinoceros and did not separate it from other rhinoceroses above the genus level.⁴ The separate family Amynodontidae was created in 1883 by Scott and Osborn in order to contain the two genera Amynodon and Orthocynodon, and distinguish them from both the rhinocerotids and the hyracodontids, another extinct rhinocerotoid group.⁴ Orthocynodon would later be designated as a junior synonym of Amynodon by Osborn in 1890.⁷ In 1889, Osborn published the first detailed comparison of Amynodontidae, Hyracodontidae, and Rhinocerotidae, and established the Amynodontidae as clearly distinct based mainly on their large canines and their shortened face (a trait later found to only apply to derived members of the family).⁴

The relationship between amynodonts and other rhinocerotoids is disputed.⁴ Leonard B. Radinsky argued in 1969 that amynodonts were not rhinocerotoids at all, but that they had evolved independently from tapiroid ancestors.⁴ In 1998, Donald Prothero placed the amynodonts in Rhinocerotoidea, as the sister taxon to a clade formed by Hyracodontidae and Rhinocerotidae.⁴ Recent revisions do not reach a consensus on rhinocerotoid relations. Amynodonts have been recovered both as the sister group of paraceratheres,⁸ or in a more basal position as the sister group of the clade containing eggyosodonts, paraceratheres, and rhinocerotids.⁹

The first attempt to establish the relations between amynodont genera was done by Miklós Kretzoi in 1942. Kretzoi established four amynodont subfamilies: Amynodontinae, Cadurcotheriinae, Metamynodontinae, and Paramynodontinae.⁴ In 1954, Vera Gromova rejected Kretzoi's classification scheme and did recognize any subgroups inside Amynodontidae.⁵

In the 1980s, William P. Wall attempted to refine the internal classification of the amynodonts. Wall was the first to perform a phylogenetic analysis of the group, and proposed two amynodont subfamilies: Rostriamynodontinae (containing only Rostriamynodon) and Amynodontinae (containing all other amynodonts). The genus Caenolophus was placed as Amynodontidae incertae sedis.⁴ Wall further divided the Amynodontinae into three tribes, based on the preorbital fossae. The Amynodontini were characterized by a large fossae that does not extend medial to the orbit, the Cadurcodontini by large fossae that did extend medial to the orbit, and the Metamynodontini by reduced fossae.⁴ Wall placed the genera Amynodontopsis, Cadurcodon, Sharamynodon, and Sianodon in Cadurcodontini, and Cadurcotherium, Megalamynodon, Metamynodon, Paramynodon, and Zaisanamynodon in Metamynodontini.⁵

The cladogram below presents the results of Wall's 1989 phylogenetic analysis of amynodont genera:¹⁰

In 1986, Donald Prothero, Earl Manning, and C. Bruce Hanson included Wall's data in a wider analysis of the Rhinocerotoidea. The recovered relationships were identical to Wall's analysis, but Prothero, Manning, and Hanson used different names for the subdivisions, proposing the subfamilies Cadurcodontinae (=Cadurcodontini) and Paramynodontinae (=Metamynodontini).¹¹

In 2017, Alexander Averianov, Igor Danilov, Jin Jianhua , and Wang Yingyong published a new phylogenetic analysis of the amynodonts as part of their description of the species Cadurcodon maomingensis.⁵ Their analysis recovered a sequential branching of basal amynodonts, and two supported clades of derived amynodonts, roughly corresponding to Wall's Cadurcodontini and Metamynodontini, and these names were thus retained.⁵ The names "Rostriamynodontinae", "Amynodontinae", and "Amynodontini" were not retained, with more basal amynodonts merely referred to as early diverging amynodontids.⁵ The results differed from Wall's in placing Cadurcotherium and Zaisanamynodon with the cadurcodontines rather than metamynodontines.⁵ Amynodontopsis was recovered as relatively basal, outside the cadurcodontines.⁵

In 2018, Jérémy Tissier, Damien Becker, Vlad Codrea, Loïc Costeur, Cristina Fărcaş, Alexandru Solomon, Marton Venczel, and Olivier Maridet performed a new phylogenetic analysis as part of the description of several new amynodont fossils from Eastern Europe. The results were similar to those of Averianov and colleagues, but differed in placing Amynodontopsis at a more derived position, with the metamynodontines, as well as in the assessment of the species Procadurcodon orientalis and Zaisanamynodon protheroi.¹² Averianov et al. considered Procadurcodon to be valid, with Z. protheroi as a possible synonym,⁵ whereas Tissier et al. considered Procadurcodon dubious and Z. protheroi valid.¹² In 2023, a more comprehensive phylogenetic analysis by Léa Veine-Tonizzo, Tissier, Maia Bukhisianidze, Davit Vasilyan, and Damien Becker produced a result similar to the 2018 analysis.⁸

Genus	Temporal range	Location	Notes	Ref
Amynodon	Middle Eocene	United States	Synonym: Orthocynodon7	4
Amynodontopsis	Middle–Late Eocene	China Hungary Romania United States		41213
Armania	Late Eocene–Early Oligocene	Mongolia	Variously classified as an amynodont,1415 a hyracodont,16 or a paracerathere.17 Has been proposed to be a synonym of Gigantamynodon.15	17
Cadurcodon	Middle Eocene–Middle Oligocene	Bulgaria China Kazakhstan Mongolia	Synonyms: Paracadurcodon,5 Sianodon5	512
Caenolophus	Middle–Late Eocene	China	Primitive amynodont or outside the family, possibly a hyracodont.5	5
Cadurcotherium	Oligocene	France Pakistan  Switzerland		5
Gigantamynodon	Late Eocene	China Mongolia	Type species suggested to be a nomen dubium.5 Alternatively suggested to have been a rhinoceros.18	5
Hypsamynodon	Late Eocene	Mongolia	Has been proposed to be a synonym of Cadurcotherium5 or Gigantamynodon15	5
Huananodon	Middle or Late Eocene	China	Based on poor fossil material19 but the teeth are characteristic of amynodonts20	1319
Megalamynodon	Middle Eocene	United States		4
Metamynodon	Middle Eocene–Early Oligocene	United States		4
Paramynodon	Middle–Late Eocene	Myanmar		5
Procadurcodon	Late Eocene	Russia United States		5
Proeggysodon	Late Eocene	China	Generally considered to be an eggysodont. Recovered as an amynodont in a 2020 phylogenetic analysis, but results doubted in the same paper.9	9
Rostriamynodon	Middle Eocene	China		5
Sellamynodon	Late Eocene or Early Oligocene	Romania		12
Sharamynodon	Middle–Late Eocene	China Kazakhstan Kyrgyzstan	Synonyms: Andarakodon,21 Lushiamynodon1221	521
Teilhardia	Middle–Late Eocene	China	Primitive amynodont or outside the family, possibly a hyracodont. Has been proposed to be a synonym of Caenolophus.19	22
Toxotherium	Late Eocene(?)–Early Oligocene	Canada United States	Suggested to have been an "aberrant, diminutive" amynodont by Radinsky (1969).23 Tentatively classified as a lophiodont by Prothero, Manning & Hanson (1986).24	23
Zaisanamynodon	Late Eocene	China Kazakhstan United States		58

Additional amynodont fossils have been reported but their taxonomic attribution at the genus level is questionable.⁵ These include "Amynodon" hungaricus from the Late Eocene of Hungary,¹² "Amynodon" sinensis from the Late Eocene of Henan, China, "Amynodon" watanabei from the Middle–Late Eocene of Japan, and the gigantic "Metamynodon" bugtiensis from the Oligocene of Pakistan.⁵ Revisions have variously retained "M". bugtiensis as an amynodont or referred it to the paracerathere Paraceratherium.⁵¹²

Amynodonts probably originated in Asia²⁴⁵ during the Early Eocene.⁵ The Early–Middle Eocene deposits of the Irdin Manha Formation in China have yielded fossils of the oldest well-defined member of the group, Rostriamynodon.⁴ Caenolophus, contemporary with Rostriamynodon, has been suggested to be the most primitive amynodont but its placement in Amynodontidae is disputed; Caenolophus has also been classified as a hyracodont, and appears to be anatomically intermediate between the two families.⁵ Teilhardia, known from the same area, has also been suggested to be either a hyracodont or a primitive amynodont but it is fragmentarily known and possibly synonymous with Caenolophus.¹⁹

Amynodonts spread to North America shortly after their initial appearance² and achieved a Holarctic distribution.⁴ They became one of the dominant herbivore groups of their time.⁵ In North America, the temporal range of the amynodonts extends from the Middle Eocene to the Early Oligocene.⁴ The earliest known North American example of the group is Amynodon, from the late Bridgerian and Uintan land mammal ages.⁴⁵ The amynodonts were at their greatest taxonomic diversity in Asia, especially in Central Asia.⁵ Amynodonts increased in body size over the course of their evolutionary history, the smallest known amynodont is the relatively basal "Amynodon" sinensis, estimated at 127 ± 15 kg (280 ± 33 lbs), whereas the largest are the derived Zaisanamynodon borisovi, at 2442 ± 257 kg (5384 ± 567 lbs), and Procadurcodon orientalis, at 2720 kg (5997 lbs).⁵

Migration across Beringia was highly important in amynodont evolution.⁴ The ancestors of the North American Amynodon and Amynodontopsis bodei were probably amynodonts that migrated from Asia separately; Amynodontopsis bodei is more derived than Amynodon⁵ but likely descended from the Asian species Amynodontopsis jiyuanensis.¹³ Phylogenetic analyses suggest that there may have been as many as four or five dispersal events where new amynodonts spread from Asia into North America.⁵ There were also several dispersal events into Europe. Cadurcotherium is the only genus recorded from Western Europe⁵ but Amynodontopsis, Cadurcodon, and Sellamynodon are all known from Eastern Europe.¹² Amynodonts do not appear to have reached Northern Europe.¹² The migration of amynodonts into Europe in the Oligocene may have been facilitated by the preceding Grande Coupure extinction event.¹²

The Middle to Late Eocene was a time of great environmental change, which may have facilitated the evolutionary radiation of the amynodonts. In North America, the environment changed during this time from lush semitropical forests to being dominated by open forests and grasslands.⁴ The timeframe corresponding to the late Uintan and the Duchesnean land mammal ages in North America saw the emergence of the cadurcodontines and metamynodontines, illustrating diversification and pursuits of different lifestyles.⁴ Similar to North American examples, Asian amynodonts also appear to diversify around the boundary between the Eocene and Oligocene.⁴ The metamynodontines experienced evolutionary radiation in North America, represented by the genera Megalamynodon and Metamynodon.⁵ The record of carcudontines in North America is sparse and depends on the phylogenetic position of Amynodontopsis and the assessments of the species Zaisanamynodon protheroi and Amynodon reedi.⁵¹²

Amynodonts were more successful in Asia than in North America, perhaps due to the increasing aridity in North America going into the Oligocene.⁴ Metamynodontines survived in North America into the Oligocene; Metamynodon lived in river system across most of the continent, but was a rare element of the Oligocene fauna and was extinct by the end of the Whitneyan land mammal age (late Early Oligocene).⁴ Amynodonts in Europe disappeared in the latest Oligocene.¹² Amynodonts survived the longest in Asia,⁴⁵ where Cadurcotherium persisted in modern-day Pakistan until the latest Oligocene.⁵ The last Asian amynodonts were previously believed to have reached into the early Miocene, but this dating has proven incorrect.¹²