Embolotherium

Discovery and initial descriptions

Type skull of E. andrewsi (AMNH 26001)

Type skull of E. grangeri (AMNH 26002)

Embolotherium fossils were first discovered in 1925 in the Ulan Gochu Formation of Inner Mongolia, China by an expedition from the American Museum of Natural History, led by Walter W. Granger. Additional and better-preserved fossils were found in 1928 on the museum's Fifth Central Asiatic Expedition, led by Granger and Roy Chapman Andrews.^[2] Cranial elements from at least fourteen Embolotherium individuals were recovered by the expeditions, including several nearly complete skulls. In 1929, Henry Fairfield Osborn described the new genus Embolotherium based on the skulls.^[2] The generic name Embolotherium references the unique bony protuberance on the front of the animal's skull and was derived from the Greek word έμβολή (emvolí), glossed by Osborn as "battering ram",^[2] and the common prehistoric mammal suffix -therium (from the Greek θηρίον, theríon, meaning "beast").^[4] Osborn easily distinguished Embolotherium from the other brontotheres known at the time by its unique protuberance, which Osborn described as a "completely novel form".^[2]

Osborn named three species based on the skulls. The type species E. andrewsi was based on the skull AMNH 26001 and named after Andrews, E. grangeri on the skull AMNH 26002 and named after Granger, and E. loucksii on the skull AMNH 26010 and named after Harold Loucks, who had discovered this skull.^[2] The main feature distinguishing the three species was the shape and position of the ram. In E. andrewsi, the ram rises up directly above the orbits (eye sockets) and curves slightly backwards, and in E. grangeri the ram rises from the middle portion of the skull, slightly behind the orbits, and curves forwards and downwards.^[2] E. loucksii was distinguished by the ram rising up directly above the orbits, forward and upwards without curving. Both E. grangeri and E. loucksii were noted to have an elongated premaxillary rostrum, which Osborn did not observe in E. andrewsi.^[2]

In 1943, Granger and William King Gregory revised the brontotheres known from Mongolia and Inner Mongolia. Granger and Gregory amended the diagnoses of Embolotherium and the species described by Osborn, changed the species designation of some of the fossils, and referred several new fossils to E. andrewsi, including skulls and lower jaws, all from the Ulan Gochu Formation.^[5] Granger and Gregory also described the new species E. ultimum, based on the partial skull AMNH 21604, found in the slightly later Baron Sog Formation. Granger and Gregory distinguished their new species from the others mainly by its larger size, the relatively great size of the third molar, and the greatly expanded zygomatic arches.^[5] They also described a new genus of large brontothere from Inner Mongolia, Titanodectes, with two species, based on lower jaws. Granger and Gregory noted that the species T. ingens could represent lower jaws of E. grangeri but actual comparison was impossible since T. ingens and E. grangeri had no known overlapping fossil material.^[5]

Osborn had believed that the ram of Embolotherium was a completely unique anatomical structure.^[2][5] Granger and Gregory instead demonstrated that the ram was a highly derived form of the horns seen in other horned brontotheres.^[5]

Further discoveries

Granger and Gregory's revision continued to be the standard work on Asian brontotheres for much of the 20th century. Additional work was carried out on the brontotheres of Mongolia and Inner Mongolia by Russian and Chinese paleontologists, but without new comparison to the large collection of Asian fossils in American museums (which contain the majority of brontothere holotypes from the region).^[5] Instead, new work largely followed on from Granger and Gregory's revision, which hindered understanding of the diversity and evolution of Asian brontotheres.^[5]

Two additional species of Embolotherium were described in the late 20th century. Demberelyin Dashzeveg described the species E. ergilense in 1975, and N. M. Yanovskaya described the species E. insigne in 1980, both based on skulls found in the Ergilin Dzo Formation of Mongolia.^[3] Yanovskaya also for the first time described partial postcranial fossils from E. andrewsi.^[6]

In 2000, Wang Ban-Yue described a new well-preserved skull and associated lower jaw (IVPP V.11959) of E. andrewsi from the Ulan Gochu Formation in Inner Mongolia.^[7]

Taxonomic revisions

In a major 2008 revision of brontotheres, Matthew C. Mihlbachler published a new detailed diagnosis of Embolotherium. Mihlbachler recognized only two species of Embolotherium as valid: E. andrewsi and E. grangeri,^[3] with the other historically named species considered synonyms. E. ultimum and E. ergilense were designated as synonyms of E. andrewsi, and E. loucksii and E. insigne were designated as synonyms of E. grangeri.^[3] Mihlbachler also designated both species of Titanodectes as synonyms of E. grangeri.^[3]

In 2018, Bai et al. agreed with Mihlbachler's assessment that only E. andrewsi and E. grangeri were valid species of Embolotherium, but rejected synonymization with Titanodectes, which they instead treated as a valid genus. Bai et al. noted that no reliable lower dentition was known from E. grangeri, which makes actual comparison to the Titanodectes fossils impossible, and that Titanodectes and Embolotherium are known from different localities, and different horizons.^[8]

Ram

Horned brontotheres are all characterized by the presence of frontonasal bony protuberances (horns).^[3] The protuberances of Embolotherium were fused into a single structure,^[3] the most ornate and highly derived protuberance of all brontotheres.^[11] The protuberance in Embolotherium is commonly referred to as the "ram", in-line with Osborn's original description.^[5][3] A similar ram is found only in the closely related Protembolotherium, though that of Protembolotherium is not as elaborate.^[3] The ram evolved from the dual horns seen in other horned brontotheres. Over the course of the evolution of Embolotherium, the horns migrated upwards on the skull, and a connecting crest developed between the horns and grew forward and upwards together with them.^[5] Protembolotherium preserves a more primitive state of ram development, and an even more primitive precursor structure is seen in Nasamplus.^[3]

Various interpretations of the unique structure have been published over the decades. In 1929, Osborn interpreted the structure as functioning as an actual horn, used for "battering, assaulting, charging, and tossing".^[2] 21st-century reinterpretations of the structure suggest that the ram of Embolotherium was incorporated into a large soft-tissue snout, and served to support a large nasal cavity. Deep channels on the lower surface of the ram, as well as the marker for the cartilaginous nasal septum, a thin ridge running along the ram, indicate that the nasal cavity of Embolotherium extended all the way to the top of the ram.^[3] E. grangeri must have had an enormous nasal cavity, far larger than that of E. andrewsi, indicated by the ram beginning further back on the skull and curving more forwards, creating a large space between itself and the jawbone.^[3]

The position of the nostrils in Embolotherium is uncertain. Life restorations traditionally depict them very low, right above the premaxilla and in a normal, rhinoceros-like position.^[3] Due to the extension of the nasal cavity, it is also plausible that the nostrils were positioned on the peak of the ram.^[3][7]

Sexual dimorphism

An unpublished quarry sample of E. grangeri from Mongolia reportedly shows evidence for sexual dimorphism, since the population includes some individuals with paired horns at the top of their rams, and others without these horns.^[11]

Embolotherium did not exhibit sexual dimorphism in certain traits that are dimorphic in several other brontotheres, such as canine size.^[11]