Metarhinus, Osborn 1908

Mader, Bryn J., 2008, A species level revision of Bridgerian and Uintan brontotheres (Mammalia, Perissodactyla) exclusive of Palaeosyops, Zootaxa 1837 (1), pp. 1-85 : 20-28

Genus METARHINUS Osborn 1908

= Rhadinorhinus Riggs 1912

= Heterotitanops Peterson 1914b

Age. Uintan.

Subage. Early Uintan.

Type species. M. fluviatilis Osborn 1908 .

Included species. M. diploconus ( Osborn 1895) .

Diagnosis. Medium sized (length P2 to M3 approximately 137–157 mm) dolichorhinine brontothere with prominent orbits and a very deep lateral nasal incision. Often there is no upper diastema and the suborbital protuberance may be small.

Discussion. In 1908 Osborn named a new genus, Metarhinus , for which he recognized three species: M. fluviatilis , M. earlei , and M. diploconus . The type species, M. fluviatilis , was based on a badly crushed and fragmentary skull (AMNH 1500, see Fig. 6 ) from the Uinta Basin, M. earlei was based on a skull (AMNH 13166) from the Washakie Basin (Washakie B of Granger 1909) lacking the nasals and most of the frontal region, and M. diploconus was based on a skull (AMNH 1863) from the Uinta Basin that Osborn had originally described as a new species of " Telmatotherium " (= Telmatherium ) and had named ( Osborn 1895) T. diploconum (note the emendation of the trivial name). It should be noted, however, that although Osborn referred to a taxon " M. diploconus " in the 1908 paper, it is not perfectly clear from the context whether the genus is meant to be Metarhinus or Mesatirhinus . In his monograph of 1929, however, Osborn stated that it was his intention to refer T. diploconum to Metarhinus .

In 1912 Riggs described a large collection of Uintan brontotheres collected two years earlier from the Uinta Basin of Utah by an expedition from the Field Museum. Among these specimens were skulls with deep lateral nasal incisions and prominent orbits, most of which Riggs referred to Metarhinus . Many of the skulls that Riggs referred to this genus were characterized by peculiar, broad, spoon-shaped nasals (see Fig. 7 ), although some of the skulls were incomplete and lacked the nasals entirely. One skull, however, while sharing with the other specimens a deep nasal incision and prominent orbits, had short, distally tapered nasals ( Fig. 8A ) and Riggs assigned this skull to a new genus Rhadinorhinus (type species R. abbotti ). Riggs also assigned Osborn's taxon, Metarhinus diploconus to this new genus and this new generic assignment was accepted by Osborn in 1929.

Mader (1989) provisionally accepted Riggs' (1912) and Osborn's (1929) conclusion that the type species of Metarhinus , M. fluviatilis , represents the brontothere with spoon-shaped nasals and that this form is generically distinct from the brontothere with distally tapered nasals, Rhadinorhinus . Mader questioned whether these two forms should be recognized as distinct genera, however, and suggested that it might be more appropriate to recognize them as separate species of the same genus or even as males and females of a single species. After further consideration I now believe that specimens with both nasal morphologies are similar enough to justify placing them in a single genus, Metarhinus , but for reasons that will be discussed below, I do not believe that they are sexual variants of a single species.

In addition to naming the new genus Rhadinorhinus, Riggs (1912) also named two new species of Metarhinus from supposedly higher levels of the Uinta Formation and representing more "advanced" stages of evolution. The first of these species, M. riparius , was based on a laterally crushed skull (FMNH 12186). Riggs listed the following as diagnostic characters of the species: skull long and narrow, anterior cranial region expanded, sagittal crest short, interorbital region relatively narrow and rounded, rudimentary horn cores above orbits, canines large, molar series short, hypocone usually present on M3, mandible straight in the ramus, and lower canine long and recurved. It should be noted that the type specimen of M. riparius lacks the lower jaws and that the last two characters were based on referred materials.

The second species, Metarhinus cristatus , was based on a skull ( FMNH 12194 ) lacking the nasals and most of the upper jaw (maxilla and premaxilla). Riggs listed the following as diagnostic characters of this species: skull length approximately 380 mm, molar series 94 mm, frontal region broad, sagittal crest long and high, molars short-crowned, no hypocone on M3, arches relatively heavy. Because the type skull lacks most of the upper jaw and Riggs did not refer any other specimens to this taxon, it is difficult to understand why he listed an estimate of skull length in the diagnosis of the species, given the extreme degree of error this estimate must necessarily have .

In addition to the two new species of Metarhinus named in the paper, Riggs also referred additional material to Metarhinus earlei Osborn and gave the following revised diagnosis of that species based on both the type and the new referred materials: skull short and broad in frontal region; length 388–405 mm; breadth 245– 255 mm; molar series broad and low-crowned; no hypocone on M3; linea aspera uniting by regular curves above posterior margins of zygomata to form a short, thickened sagittal crest; canines slender; diastema short; P2 oblique; P3 and P4 subrectangular in outline; molars broad and low-crowned; mandible with ramus slightly curved.

Osborn (1929) recognized both Metarhinus riparius and M. cristatus as valid species, although he suggested that M. cristatus might be a junior synonym of M. fluviatilis . The diagnoses of these taxa given by Osborn are taken directly from Riggs (1912). Osborn provided his own diagnosis, however, for M. earlei (1929, p. 426), based in part on his original diagnosis ( Osborn 1908) and that of Riggs (1912). According to this diagnosis M. earlei is defined by, "Skull (AMNH 13166, type) length 393 millimeters, breadth 240, or 388:245, or 405:255; cephalic index 60–63. Occipital condyles narrow (78 mm), premaxillary symphysis elongate, nasals elongate, spreading distally, prominent infraorbital shelf. Type p1–m3, 167 millimeters. Molar series broad and low crowned, no hypocone on m3; canines slender, diastema short."

Although Riggs (1912) and Osborn (1929) cited numerous differences between the cranial morphologies of the various taxa that they recognized, all of the morphological differences that are of any consequence are directly attributable to taphonomic deformation and, to a lesser degree, sexual dimorphism. For example, Osborn (1929) cited a high and prominent sagittal crest as a diagnostic character of Metarhinus fluviatilis , but this character is due entirely to the effects of crushing on the type skull. The same crushing may also account for some of the prominence of the circumorbital ridges also cited by Osborn as a character of the species. Similarly, both Osborn (1929) and Riggs (1912) stated that a long and narrow skull as well as a narrow and rounded interorbital region are diagnostic characters of M. riparius . The type skull of M. riparius , however, has been crushed laterally, exaggerating the cranial proportions. This crushing accounts for both of the characters cited by Osborn and Riggs. The size of the infraorbital shelf (=suborbital protuberance) and canine are probably sexually dimorphic characters in Metarhinus and should not be used in the diagnosis of a species (see Mader 1989 and below).

The dental differences cited by Riggs and Osborn are rather trivial and, in my opinion, do not justify the recognition of distinct taxa. As in other brontothere genera, the presence or absence of a hypocone on M3 and minor differences in the shape of various cheek teeth are poor diagnostic characters, and may generally be attributed to intraspecific variation.

As indicated above, the skulls with tapered and spoon-shaped nasals probably represent different species of a single genus, Metarhinus . That they are not males and females of a single species is suggested by the size variation of the suborbital protuberance in at least the form with tapered nasals (which is probably sexually dimorphic), and by the relatively high value of V for canine size in the form with spoon-shaped nasals (see below), which also suggests sexual dimorphism. The possible identification of males and females in the two groups represented by the two nasal morphologies, suggests that these groups are taxonomically distinct. It is possible, however, that there could be more than one species with spoon-shaped or distally tapered nasals.

The sample of Metarhinus with tapered nasals is very small (n = 4) and those specimens that are available are, for the most part, too poorly preserved to measure in detail. Despite Riggs' (1912) and Osborn's (1929) recognition of two species of Metarhinus with distally tapered nasals (which they referred to the genus Rhadinorhinus ), I find no important morphological differences between any of the specimens with this nasal morphology. Given that all specimens with distally tapered nasals are morphologically alike and that the sample is insufficient to perform a statistical analysis, there is no basis for the recognition of more than one taxon. I, therefore, regard all specimens of Metarhinus with distally tapered nasals as representing a single species, Metarhinus diploconus (see below). If a larger sample of specimens with this nasal morphology should become available, however, further analysis testing this conclusion would be justified.

Riggs (1912) and Osborn (1929) also recognized more than one species of Metarhinus with spoon-shaped nasals, but once again, I have been unable to identify any important morphological differences between specimens of Metarhinus with this nasal morphology. The sample of Metarhinus with spoon-shaped nasals is also rather small, but is large enough to perform at least a preliminary statistical analysis.

All specimens of the variety of Metarhinus with spoon-shaped nasals are from the Wagonhound Member of the Uinta Formation (Uinta Basin) or from chronologically equivalent strata in the Adobe Town Member of the Washakie Formation (Washakie Basin). Because the sample of specimens from both locations is small, they were combined for the purpose of analysis.

Table 6 presents the summary statistics for the sample of Metarhinus with spoon-shaped nasals. Although the average value of V for this sample is within the range of typical extant mammalian species (= 4.9), only slightly more than half of the individual values of V are within the expected range of 4 to 10. Ten individual values of V are below 4 (rounded to the nearest whole number), but seven of these were calculated from samples consisting of four or fewer cases. I believe that these samples are probably too small to show all of the variation that is actually present. If all variables with fewer than four cases are eliminated from consideration, over three quarters of those remaining have values of V between 4 and 10 and the average value for the sample remains relatively unchanged (= 5.0).

The high values of V for the right canine suggests that canine size may be sexually dimorphic in the variety of Metarhinus with spoon-shaped nasals. The sample for the left canine, however, is too small to reflect this. The high values of V for canine size affect the average value of V for the sample only slightly. If canine size is excluded from the calculation of the average value of V, the resulting value is, in general, typical of extant mammalian species (average V = 4.5). If variables with fewer than four cases are eliminated from consideration as well as canine size, the average value of V for the sample becomes 4.8.

Cluster analysis ( Fig. 9 ) of all of the variables listed in Table 6 indicates that there are two size groups present in the form with spoon-shaped nasals. These groups are not greatly different in size, however, and both join at a distance of only 1.9 mm. It is possible that these groups could represent two different species, two groups within a single species (such as males and females), or could possibly be an aberration due to the small sample size.

Based on the present results, I do not believe that there is justification for the recognition of more than a single species of Metarhinus with spoon-shaped nasals. The individual and average values of V for the entire sample are generally within the parameters recognized for a single species and, although cluster analysis suggests two size groups, these groups are separated by only a very small size interval. Normally I would have performed t -test analyses to compare both size groups suggested by the cluster analysis, and values of V would be calculated for each. The small sample size of both size groups, however, makes this impractical. If a larger sample should become available, statistical analyses should be performed again to determine whether the conclusions reached here remain valid. Specifically, it should be confirmed whether the individual and average values of V for the combined sample are within the normal parameters of a single species and whether two size groups are actually present. If two size groups are, in fact, present then tests should be performed to determine whether the group means are significantly different and whether the individual and average values of V for each group are within the range of a single species.

a Based, whenever possible, on an average of left and right measurements.

b Insufficient data.

c Excluding Diastema Length.

Generally, the forms with spoon-shaped and distally tapered nasals are easily distinguished from one another, but if the nasals are lacking, it is difficult to distinguish between them because, exclusive of the nasals, the size and morphology of both are rather similar. The form with spoon-shaped nasals tends to have a relatively small suborbital protuberance compared to the form with tapered nasals, but some specimens with tapered nasals (interpreted as probable females by Mader 1989 and above) also have suborbital protuberances that are weakly developed. Similarly, some specimens with spoon-shaped nasals have a distinctive, sharply downturned zygomatic arch that is similar to that of Palaeosyops but not as robust (see Fig. 7 ). I have found no specimens of Metarhinus with distally tapered nasals that have this zygomatic arch morphology, but the zygomatic arch is often poorly preserved in specimens of Metarhinus . It is not certain, therefore, that all specimens of Metarhinus with spoon-shaped nasals share this zygomatic arch morphology and that at least some specimens of Metarhinus with tapered nasals do not.

As a group, specimens with tapered nasals appear to be slightly larger in size than specimens with spoonshaped nasals (compare the sizes in Fig. 8 ). This size difference is most evident when the basilar length of the skull, length of the cheek tooth series, or length of the cheek tooth series exclusive of the first premolar is considered, although my data indicate that the size difference between the largest individual with spoon-shaped nasals and the smallest individual with distally tapered nasals is no more than 0.5 to 2 millimeters for these linear measurements. Length of the premolar series, length of the molar series, and individual tooth dimensions overlap in the few specimens that are available for comparison and it would not be surprising with a larger sample to find that the skull length and length of the cheek tooth series (with or without P1) overlap as well.

Because of the extremely small size of the holotype, the type species of Metarhinus , M. fluviatilis , probably belongs to the variety of Metarhinus with spoon-shaped nasals, as Riggs (1912) and Osborn (1929) concluded. Similarly, the type of Osborn's " Telmatotherium " diploconum is rather large, and probably represents the variety of Metarhinus with tapered nasals, again as Riggs and Osborn had concluded. Unfortunately, the type of M. earlei is intermediate in size and it is not certain to which morphological group it belongs. It is also impossible to determine which group the type of M. cristatus belongs to based on size because the specimen is insufficiently preserved. Based solely on the size of the type specimens, I provisionally accept M. fluviatilis as representing the form with the spoon-shaped nasals and M. diploconus as representing the form with distally tapered nasals. Metarhinus earlei and M. cristatus are treated as a nomina dubia, because they could be junior synonyms of either M. fluviatilis or M. diploconus .

Peterson (1914b) described the partial skeleton (CM 2909) of a very young brontothere from the Uinta Basin, which he identified as a new genus, Heterotitanops (type species H. parvum ). W.K. Gregory, however, concluded that this specimen probably represents a very young individual belonging to an undetermined species of Metarhinus or Rhadinorhinus ( Osborn 1929, p. 196) and Osborn (1929, p. 425) suggested that it might represent an extremely young (perhaps fetal) Metarhinus fluviatilis . An examination of the skull of CM 2909 shows the extreme depth of the narial incision, the lateral border of which lies in close proximity to the orbit. This morphology is clearly diagnostic of Metarhinus but, unfortunately, the lack of nasals and permanent teeth makes it difficult to determine which species is represented. I provisionally regard Metarhinus parvum as a nomen dubium.

Stock (1937) described the fragmentary right facial region, LACM (CIT) 2037, of a small brontothere from San Diego County, California. According to Stock, this specimen is from the Poway Conglomerate, but the formation from which the specimen was taken is now known as the Friars Formation ( Golz 1976; Golz & Lillegraven 1977). Stock identified the specimen as a new species, which he provisionally referred to Metarhinus , and named Metarhinus (?) pater . Stock gave the following diagnosis for the new taxon:

"Muzzle elongate; naso-maxillary recess deep and reaching back of antorbital [= infraorbital] foramen; antorbital foramen large and situated above anterior border of M2. P1 and P2 small; P4 relatively long in comparison to its width. Size larger than Metarhinus fluviatilis and M. riparius , similar to that of M. earlei and smaller than Mesatirhinus superior [= Sphenocoelus intermedius ]."

In comparing the type of Metarhinus (?) pater to specimens that Riggs (1912) and Osborn (1929) had referred to Metarhinus and Rhadinorhinus, Stock noted that in all these materials the infraorbital foramen is in close proximity to the orbit. Stock further noted that the type of Metarhinus (?) pater and specimens referred to Metarhinus have a postnarial notch that is positioned slightly anterior to the back of M2, and a P1 that is small in size.

According to Stock, Metarhinus (?) pater differs from specimens referred to Metarhinus in having a deeper naso-maxillary recess, a longer snout, longer postcanine diastema, longer P4, and more posteriorly positioned infraorbital foramen (located over the front of M2). Stock distinguished Metarhinus (?) pater from Rhadinorhinus (= Metarhinus diploconus in the present paper), by the former's possession of a P2 that is not subquadrate, longer postcanine diastema, more posteriorly positioned postnarial notch, better developed infraorbital shelf (= suborbital protuberance), and more posteriorly positioned antorbital (= infraorbital) foramen (position relative to M1). Stock also noted that Metarhinus pater lacked the upwardly curved cheek tooth series that Osborn (1929) had cited as a character of " Rhadinorhinus ".

The relatively elongated upper molars in the type of Metarhinus (?) pater suggest that the identification of the specimen as a dolichorhinine brontothere is probably correct. In a scatter plot that I performed (not shown here) comparing length against width for M2, the type of Metarhinus (?) pater plotted among specimens that belong to the subfamily Dolichorhininae and below the diagonal line (see Fig. 1 ) indicating equality of length and width (i.e., among specimens in which the tooth length exceeds the width). Because of the close proximity of the infraorbital foramen to the orbital rim, and because it appears that the lateral nasal incision was probably very deep, I accept Stock's assignment of the specimen to Metarhinus .

Of the two species of Metarhinus recognized as valid in the present paper, the type of Metarhinus pater is most similar in size to the larger species, M. diploconus , which was formerly referred to the genus Rhadinorhinus . The type of M. pater is slightly smaller (based on the length of the cheek tooth series exclusive of P1 and length of the molar series) than the type of M. diploconus , but is larger than any of the specimens I refer to that species, below.

In my opinion, the characters used by Stock to distinguish Metarhinus pater from M. diploconus (i.e., Rhadinorhinus ) are probably not sufficient to justify the recognition of a distinct species. Two of the characters, diastema length and squareness of the P2, are highly variable in brontotheres and are generally not useful for diagnostic purposes. Two other characters, the position of the postnarial notch and position of the infraorbital foramen, are potentially better for diagnostic purposes, but I am not convinced that the difference is significant. Stock's assertion that the suborbital protuberance of M. pater is larger than that of M. diploconus was entirely conjectural because, as he noted (1937, p. 50), the jugal has been completely broken away along the plane of the jugal-maxillary border and the region of the suborbital protuberance is thus not preserved. Finally, as indicated above, Osborn was incorrect when he cited the curvature of the cheek tooth series as a diagnostic character of " Rhadinorhinus " because this observation was based on one or more taphonomically distorted specimens. This is not a valid distinction, therefore, between M. pater and M. diploconus . For the purposes of the present paper I accept M. pater as a possible junior synonym of M. diploconus .

William Turnbull and David Martill (Martill & Turnbull, abstract 1986; Turnbull & Martill 1988) have reported the existence of a monospecific brontothere assemblage from the Adobe Town Member of the Washakie Formation (Washakie Basin). The specimens range from juvenile to very old individuals and are believed to have been killed in a single mass mortality event. The specimens were provisionally referred to the genus Mesatirhinus but, after a preliminary examination I believe that they probably represent the genus Metarhinus , a conclusion also reached by McCarroll et al. (1996) and accepted by Mihlbachler (2005). This identification is further suggested by the stratigraphic level (TWKA2), which represents early Uintan age. An analysis of this quarry sample may help to solve some of the taxonomic uncertainties discussed above by more clearly establishing the size and morphological variation present in a single species of Metarhinus .

Figures

FIGURE 1. Scatter plots showing length of M2 plotted against width of M2 for five brontothere genera. The solid diagonal line that crosses both graphs represents the points at which the length and width of the tooth are equal. Specimens that plot out above the line have teeth that are wider than long, while specimens that plot out below the line have teeth that are longer than wide. Specimens that plot out directly on the line have teeth in which the length is equal to the width and are thus square in shape. Black symbols, subfamily Dolichorhininae; white symbols, other brontothere genera (Palaeosyops and Telmatherium).

FIGURE 6. Holotype skull of Metarhinus fluviatilis (AMNH 1500) in A, dorsal; and B, lateral views. After Osborn, 1929.

FIGURE 7. Skull of Metarhinus fluviatilis in A, dorsal; and B, lateral views.

FIGURE 8. Skulls of A, Metarhinus diploconus; and B, Metarhinus fluviatilis in dorsal view.

FIGURE 9. Cluster dendrogram for specimens of Metarhinus with spoon-shaped nasals (Metarhinus fluviatilis) resulting from a cluster analysis of all variables listed in Table 6. a, type of Metarhinus riparius.