The Autochthonous North American Musk Oxen Bootherium, Symbos, and Gidleya (Mammalia: Artiodactyla: Bovidae) JERRY N. MCDONALD and CLAYTON E. RAY SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY • NUMBER 66 SERIES PUBLICATIONS OF THE SMITHSONIAN INSTITUTION Emphasis upon publication as a means of "diffusing knowledge" was expressed by the first Secretary of the Smithsonian. In his formal plan for the Institution, Joseph Henry outlined a program that included the following statement: "It is proposed to publish a series of reports, giving an account of the new discoveries in science, and of the changes made from year to year in all branches of knowledge." This theme of basic research has been adhered to through ttie years by thousands of titles issued in series publications under the Smithsonian imprint, commencing with Smithsonian Contributions to Knowledge in 1848 and continuing with the following active series: Smithsonian Contributions to Anthropotogy Smithsonian Contributions to Astrophysics Smithsonian Contributions to Botany Smithsonian Contributions to the Earth Sciences Smithsonian Contributions to the l^arlne Sciences Smithsonian Contributions to Paleobiology Smithsonian Contributions to Zoology Smithsonian Folklife Studies Smithsonian Studies in Air and Space Smithsonian Studies in History and Technology In these series, the Institution publishes small papers and full-scale monographs that report the research and collections of its various museums and bureaux or of professional colleagues in the world of science and scholarship. The publications are distributed by mailing lists to libranes, universities, and similar institutions throughout the world. Papers or monographs submitted for series publication are received by the Smithsonian Institution Press, subject to its own review for format and style, only through departments of the various Smithsonian museums or bureaux, where the manuscripts are given substantive review. Press requirements for manuscript and art preparation are outlined on the inside back cover. Robert McC. Adams Secretary Smithsonian Institution SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY • NUMBER 66 The Autochthonous North American Musk Oxen Bootheriumy Symbos, and Gidleya (MammaUa: Artiodactyla: Bovidae) Jerry N. McDonald and Clayton E. Ray SMITHSONIAN INSTITUTION PRESS Washington, D.C. 1989 ABSTRACT McDonald, Jerry N., and Clayton E. Ray. The Autochthonous North American Musk Oxen Bootherium, Symbos, and Gidleya (Mammalia: Artiodactyla: Bovidae). Smithsonian Contribu- tions to Paleobiology, number 66, 77 pages, 64 figures, 4 tables, 1989.—^The history of taxonomy of the autochthonous genera of North American musk oxen—Bootherium, Symbos, and Gidleya—is reviewed. The bases upon which taxonomic judgments within the group have been made are identified. These bases are reevaluated in the light of current information on patterns of ontogenesis, sexual dimorphism, postmortem alteration of skeletal remains, and spatial and temporal distribution of musk ox records. The bases used by taxonomists in the past to justify separation of this musk ox group into multiple genera and species can be explained best as indices of sexual dimorphism or postmortem weathering and abrasion. All nominal species within Bootherium. Symbos. and Gidleya are, therefore, placed in synonymy with the senior name in the group, Bootherium bombifrons (Harlan, 1825). A revised diagnosis is provided for the monotypic species. OFFICIAL PUBLICATION DATE is handstamped in a limited number of initial copies and is recorded in the Institution's annual report, Smithsonian Year. SERIES COVER DESIGN: The trilobite Phacops rana Green. Library of Congress Cataloging in Publication Data McDonald, Jerry N. The autochthonous North American musk oxen. (Smithsonian contributions lo paleobiology; no. 66) Bibliography: p. 1. Musk ox, Fossil—North America—Classification. 2. Paleontology—North America. I. Ray, Clayton Ed- ward. II. Smittisonian Institution. III. Title. IV. Series. QE701.S56 no. 66 560 s [569'.73] 88-607937 [QE882.U3] Contents Page Introduction 1 Acknowledgments 2 History of Taxonomy of the Group 3 The Content of Bootherium, Symbos. and Gidleya 24 The Relationship between Gidleya and Symbos 26 The Relationship between Bootherium and Symbos 27 Differences in the Size of Bootherium and Symbos Crania 28 Differences in the Characteristics of the Horn Cores 31 Differences in the Dorsal Surface of the Cranium 40 Differences in the Basioccipital-Basisphenoid Flexion 44 Differences in Depth of the Lacrimal Fossae 46 Differences in the Number of fioo^/zmum and 5ymZ?05 Specimens 46 Similarities in the Morphology and Distribution oi Bootherium and Symbos Specimens 47 Conclusions 51 Systematic Hierarchy 67 Bootherium bombifrons (Harlan, 1825) 67 Appendix I: Abbreviations for Institutional, Departmental, and Private Collections Used in This Study 70 Appendix II: Chronology of Nominal Species Referable to Musk Oxen in the Genera Bootherium. Symbos. and Gidleya 72 Appendix III: Radiocarbon Dates on Bootherium and Symbos Specimens 73 Literature Cited 74 Ul The Autochthonous North American Musk Oxen Bootherium, Symbos, and Gidleya (Mammaha: Artiodactyla: Bovidae) Jerry N. McDonald and Clayton E. Ray Introduction Five genera of musk oxen (low-horned ovibovines) presently are recognized from the Quaternary of North America, Two of these genera (Ovibos and Praeovibos) were Holarctic in distribution, whereas the other three {Bootherium. Symbos, and Gidleya) are known only from the Nearctic. Praeovibos Staudinger, 1908, at one time considered to be the earliest known representative of the musk ox group, is known from fossil skull material from at least 16 sites in Europe and Asia that date from the middle Pleistocene (i.e., the Giinz, Mindel, and possibly Riss stages, and their equivalents) (Kahlke, 1964; Sher, 1974; Cregut-Bonnoure, 1984; Moigne, 1984). Ovibos Blainville, 1816, however, is also known from deposits possibly of Giinz age at Siissenborn, East Germany (Kahlke, 1964; Kurten, 1968; Cregut-Bonnoure, 1984), and was considered by Kahlke (1975) to be the oldest genus of musk ox. Recently, skeletal remains assigned to Praeovibos sp. have been reported from Venta Micena, Spain, and Casa Frata, Italy (Moya-Sola, Augusti, Gibert, and Pons-Moya, 1981; De Giuli and Masini, 1983), both of which, along with at least one of the Kolyma Basin sites in Siberia, are reported as predating those at Siissenborn. The earliest records of Ovibos and Praeovibos from North America are from deposits presumed to be of Illinoian age near Nome and Fairbanks, Alaska, respectively (Pewe and Hopkins, 1967; Harington, 1970a). Nearctic records of Praeovibos are restricted to eastern Beringia (Fairbanks area, Alaska, Old Crow Basin, Yukon Territory, and Porcupine River, Yukon-Alaska), whereas cranial records of Ovibos extend from Alaska southward to Montana, Wyoming, Nebraska, Iowa, Illinois, Ohio, and the continental shelf of New Jersey Jerry N. McDonald, P.O. Box 10308, Blacksburg, Virginia 24062. Clayton E. Ray. Department of Paleobiology, hlatioruil Museum of Natural History. Smithsonian Institution. Washington. D.C. 20560. (Hay, 1923, 1924; Barbour, 1934; Harington, 1978; Walker, 1982; McDonald and Ray, unpubl. data). The North American autochthons, Bootherium (Harlan, 1825), and Symbos (Leidy, 1852), appear simultaneously in deposits usually assigned to the Illinoian glaciation (= late Irvingtonian-early Rancholabrean land mammal ages). The earliest record o( Bootherium is from Cripple Creek Sump (and possibly Gold Hill and Lower Cleary Creek) in the Fairbanks District, Alaska (Pewe and Hopkins, 1967; Pewe, 1975). Remains of Symbos of Illinoian age also have been recovered firom Cripple Creek Sump, as well as firom the North Prong Quarry (Mullen local faunas), Cherry County, Nebraska, and equivocally from the Conard Fissure, Newton Coimty, Arkan- sas (Brown, 1908; Jakway, 1961a; Pew6 and Hopkins, 1967; Martin, 1972). Both North Prong Quarry and Conard Fissure local faunas, however, present some problems of specimen identification or age assignment (see discussion, pp. 48- 51). A Symbos specimen from sediments tentatively assigned to the late Illinoian (i.e., middle Rancholabrean) has been reported from Booth Canyon, Bonneville County, Idaho (White, 1985). The single specimen of Gidleya (Gidley, 1906), from Black Rocks, McKinley County, New Mexico, was considered at one time to date from the early Pleistocene because of its association "with camels and horses, animals of early Pleistocene age" (Hay, 1924:179). Now, however, the Gidleya record, and all Bootherium and Symbos records from localities other than those just mentioned, are considered to be either Sangamonian or Wisconsinan in age. Praeovibos became extinct in Eiu^asia (Sher, 1974; Cregut- Bonnoure, 1984) near the end of the Middle Pleistocene (during the ?Riss glaciation). Praeovibos might have become extinct in North America before the Wisconsinan glaciation, but specimens assigned to this genus have been found at lower Cleary Creek, and have been identified tentatively from essentially late Wisconsinan faunules at Cripple Creek, Dome 1 SMITHSONL\N CONTRIBUTIONS TO PALEOBIOLOGY Creek, Gold Hill, and lower Cleary Creek, Alaska. Bootherium. Symbos, and Gidleya appear to have become extinct by the end of the Wisconsinan glaciation. Ovibos became extinct in Eurasia around 3000 yr B.P., and survived into the historic period only in the higher latitudes of North America and Greenland (AUen. 1912, 1913; Harington, 1970a, 1970b; Cregut-Bonnoure, 1984). The taxonomy of the autochthonous North American musk oxen has been unstable since the earliest recognized taxa were described in the second and third decades of the 19th century. Questions about the relationship between Bootherium and Symbos arose soon after the group was first revised by Leidy (1852a,b) and have continued, unresolved, to the present (e.g.. Nelson and Neas, 1980; White, 1985; Nelson and Madsen, 1987). Another conspicuous taxonomic problem involving musk oxen concerns the status of Gidleya. a genus established originally as Liops on an extensively abraded partial cranium from New Mexico (Gidley, 1906). Although Gidleya has received attention from paleontologists throughout its eighty- year history (Cossmann, 1907; Allen, 1913; Troxell, 1915; Hay, 1922,1924; Ryziewicz, 1933,1955; Frick, 1937; Kretzoi, 1942; Harington, 1961; Romer, 1966), it has failed nonetheless to receive functional validation as a viable taxon. No less than 13 nominal species were erected within Bootherium. Symbos. and Gidleya between 1825 and 1942. Most of these species were established upon weak foundations—isolated teeth, isolated postcranial elements, or fragments of crania. Six of the 13 species already have been synonymized with more securely founded taxa, but the status of the remaining species also needs to be reevaluated. Much of the reason for taxonomic instability at the generic level of the autochthonous North American musk oxen can be attributed to either failure to recognize sexual dimorphism and individual variation or assess differential postmortem alteration of various specimens. This circumstance is in part a result of very few specimens being available at the time for study and comparison. Furthermore, the questions being asked often were different from those being asked today. The arguments that have characterized this debate have endured long past their time in great part because a comprehensive review of the North American fossil musk ox group has not been undertaken since Allen did so in 1913. The taxonomy of the group has not benefited from the information and ideas acquired or developed during the last 75 years, and consequently the uncertainties related to the conceptualization and use of the three recognized genera are based to a surprising but real extent upon 19th century information and ideas. Information acquired since 1913 relevant to the taxonomic status of Bootherium, Symbos. and Gidleya is herein updated and reevaluated. The 226 available specimens is much larger than the 16 or so known to Allen (who, apparently, personally examined only two of these). Consequently, patterns of morphological and geographical variation can be documented and analyzed with substantially greater resolution and confi- dence than was possible in 1913. More reliable dating methods and inter-site correlations permit more confident assessments of chronology and contemporaneity of the three nominal genera. Data obtained from taxa closely related to musk oxen permit comparisons of patterns of individual morphological variation, sexual dimorphism, differential preservation and collection of male and female skeletal elements, and geographi- cal distribution. Recent studies of taphonomy provide much useful information about the effects of weathering and abrasion processes on bones and teeth, and patterns of postmortem biological alteration of skeletal remains. Our purpose is to reassess the relationships among the genera Bootherium, Symbos. and Gidleya. We first review the taxonomic history of the group, directing particular attention to the specific questions that have characterized the instability. In this same section we identify the morphological characters and other data that have been invoked to support the diverse perspectives that have characterized the debate. Next we present our views on the status of the nominal species within each of the three genera. In a third section we consider the relationship between Gidleya and Symbos. Lasdy, we examine the validity of the various elements of the arguments over the relationship between Bootherium and Symbos. incorporating modern data and insights derived from studies of individual variation, sexual dimorphism, ontogenetic change, and tempo- ral and spatial disu^ibution patterns. In this paper the names Bootherium. Symbos. and Gidleya are used as though they represented distinct taxa until we establish and summarize our case for their synonymy in the final section. We have omitted use of the diaerisis in Bootherium in accordance with Article 27 of the 3rd edition of the International Code of Zoological Nomenclature. Abbreviations for institutional, departmental, and personal collections containing specimens used in this paper are identified in Appendix I. ACKNOWLEDGMENTS.—The theme and scope of this paper was established in 1982 and 1983 during McDonald's tenure as a Smithsonian Postdoctoral Fellow, working with Ray, in the DeparUnent of Paleobiology, National Museum of Natural History. The paper was written in 1984 and 1985, since which time it has remained essentially unchanged save for adding new records and references to newly published factual information. The ideas presented in this paper, however, have developed during our more than 30 years of combined research on musk and shrub oxen. Diying that period, we have benefited from the assistance and interest of many colleagues. For facilitating access to collections, for loaning specimens, for sharing unpublished information, and for conversation and debate, we thank the following people: William A. Akersten, Carol W. Allison, Donald Baird, Charles S. BarUett, Jr., Stig M. Bergstrom, Robert C. Bright, Kenneth Caster, Charles S. Churcher, Vickie L. Clay, William A. Clemens, Jr., John Connaway, John P. Cook, Richard G. Corner, Richard A, Davis, A. Gordon Edmund, Ralph E. Eshelman, Anthony Fiorillo, Weldon D. Frankforter, Jr., Larry E. Freeman, Linda Gordon, Russell W. Graham, the late John E. Guilday, Carl E. NUMBER 66 Gustafson, R. Dale Guthrie, Mark S. Hafner, Charles A. Handley, Jr., C. Richard Harington, Toni Herrin, the late Claude W. Hibbard, Robert M. Hunt, Jr., Louis L. Jacobs, David R. Klein, George E. Lammers, Everett H. Lindsay, K. Don Lindsey, Ernest L. Lundelius, Jr., James E. Martin, Earl Manning, H. Gregory McDonald, Malcobn C. McKenna, Susanne J. Miller, John F. Neas, Michael E. Nelson, Sarah W. Neusius, John Palmquist, Ronald Parsley, the late Bryan Patterson, John D. Pinsof, Charles Potter, Mr. and Mrs. Don Rice, Ronald L. Richards, Judith A. Schiebout, Gary Selinger, Holmes A. Semken, Jr., L. Gay Shapiro, Morris F. Skinner, Theresa Skwara, Bob H. Slaughter, Robert E. Sloan, Charles L. Smart, Jr., Ralph Space, John E, Storer, Margot Surovik- Bohnert, Lloyd G. Tanner, Richard H. Tedford, Richard Thorington, Charles W. Totten, WiUiam D. Turnbull, John A. White, Frank C. Whitmore, Jr., Melissa C. Winans, and Susan L. Woodward. Richard A. Davis, H. Gregory McDonald, John D. Pinsof, and Theresa Skwara provided us with unpublished manuscripts and permitted us to refer to specimens described in those manuscripts. Robert S. Hoffmann provided us with a copy of John Neas's M.A. thesis on 2 November 1987, C. Richard Harington, James H. Madsen, Jr., John F. Neas, Michael E. Nelson, and John A. White presently share our research interest in the systematic relationship among Boo- therium, Symbos, and Gidley. We have benefited from interaction with these colleagues. The photographs used herein were taken by Victor A. Krantz, and the illustrations were prepared by Mary Parrish. Gladwyn B. Sullivan was always available when called upon for preparing, casting, photographing, yansporting, and pack- ing specimens; throughout his years of dependable service 'Tutt's" "happy hands" made our work easier, more efficient, and more pleasant than it otherwise might have been. Holmes A. Semken, Jr., and Frank C. Whitmore, Jr.,read and provided critical comments about an earlier version of this paper. We wish to express our deep appreciation for their assistance, but we accept full responsibility for the information and views contained herein. History of Taxonomy of the Group The first pubUshed record of a fossil musk ox fi-om North America was Wistar's description and figures of a cranium with horn cores (ANSP 994; Figures 1-4) collected at Big Bone Lick, Kentucky, by William Clark in 1807 (Wistar, 1818; Rice, 1951). Wistar recognized the resemblance between some characters in his specimen and those of domestic cattle, sheep, and goats, and especially bison, but he gave no indication of having compared the specimen with the tundra musk ox, descriptions of which were available at that time (e.g., Jeremie, 1720; Pennant, 1781,1784; Blainville, 1816). Although Wistar realized that the cranium from Big Bone Lick was "very different from that of any animal now known here" (Wistar, 1818:379),he did not name this newly documented form of life (Wistar, 1818). Seven years later, Harlan (1825), in Fauna Americana, described Bos bombifrons on the basis of the Big Bone Lick cranium (ANSP 994: Figures 1-4). This species shared the genus Bos with B. americanus (now Bison bison) and the new species B. latifrons (now Bison latifrons); the tundra musk ox (Ovibos moschatus) was retained as a monotypic genus, following Blainville (1816). Both Wistar and Harlan placed the Big Bone Lick specimen near Bison because the skulls of both were similar in size, and the horn cores of each were rounded and projected laterally fi'om the cranium distinctly anterior to the occipital crest. Harlan apparently overlooked the remnant lacrimal depression on the left side of the Big Bone Lick skull, for one diagnostic criterion of his genus Bos (and, erroneously, Ovibos) was the absence of a "lachrymal depression" (Harlan, 1825:264,267). In 1828, De Kay described and figured a second fossil musk ox specimen, a partial cranium with the bases of the horn cores, discovered at New Madrid, Missouri, following the famous earthquake of 1812. This specimen, too, was originally identified as "a petrified occiput, and the horn processes of the American Bison" but, in De Kay's opinion, "none of that genus are now to be found in our country, whose crania in the slightest degree resembles the one under consideration" (De Kay, 1828:285). Placing emphasis upon the flattened horn cores, De Kay correctly removed this specimen from the bison group and placed it with the musk oxen. Upon comparing the New Madrid specimen with Cuvier's description of Ovibos moschatus. however, De Kay recognized important differences in the character of the dorsal surface of the frontals (although De Kay allowed that his specimen, with its conspicuously roughened surface, could have been damaged or have lost the entire dorsal surface), the depth of the occipital surface (being greater in his specimen than in Ovibos), and "the most remarkable differ- ence" (De Kay, 1828:287)—the position of the horns. The bases of the horn cores suggested to De Kay (1828:287,289) that the longest axis of the horns has tseen in the same direction with the txxies of the face ... or to speak more definitely, the axis of the longest diameter of the horns, is parallel with the bones of the palate. These observed and inferred differences between the New Madrid specimen and existing descriptions of Ovibos mos- chatus led De Kay to consider affinities with fossil specimens from Siberia originally described by Pallas (1773) and Ozeretskovsky (1811), as reported by Cuvier in an unidentified edition of his Recherches. De Kay found sufficient support in Cuvier's comments about the Siberian specimens to erect a new species. Bos pallasii. to accommodate "the fossil crania of Pallas and Ozeretskovsky, and provisionally, the specimen from the banks of the Mississippi" (De Kay, 1828:291; our emphasis). The Siberian specimens of Pallas, however, earlier had been named Ovibos pallantis by Smith (1827), making fio5 pallasii a junior synonym of the former. The New Madrid specimen was later referred to Bootherium cavifrons by Leidy 5 cm SMITHS0NL\N CONTRIBUTIONS TO PALEOBIOLOGY FIGURE 1.—Holotype of Bos bombifrons Harlan, 1825 (ANSP 994), in dorsal view. 5 cm FIGURE 2.—Holotype oiBos bombifrons Harlan, 1825 (ANSP 994), in right lateral view. NUMBER 66 5cm FIGURE 3.—Holotype of Bos bombifrons Harlan, 1825 (ANSP 994). in ventral view. 5 cm FIGURE 4.—Holotype of Bos bombifrons Harlan, 1825 (ANSP 994), in caudal view. SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY (1852b), and, although this specimen has been lost or destroyed, Leidy's decision seems to have been correct. De Kay's contribution to the nomenclatural history of the Pleistocene musk oxen of North America was that he was the first to recognize the occurrence of a fossil musk ox in North America, and he correctly identified the major morphological differences between Ovibos and what would become Booth- erium (in part, then Symbos). Leidy reviewed the extinct North American oxen in 1852. At that time, the only known specimen of Bos bombifrons was the holotype fi-om Big Bone Lick, while at least 12 crania of the kind described by De Kay were known (Leidy, 1852b). In May 1852, Leidy (1852a:71) noted that both forms of musk oxen possessed large larmiers or lachrymal depressions, as in the deer, and if these are possessed by the Ovibos moschatus, the two fossils would belong to the same genus as Ovibos bombifrons and O. cavifrons; but if they are not possessed by Ovibos, as is stated lo be the case by Desmarest, the two latter sp>ecies would form a new genus, for which the name Bootherium is proposed. The genus Bootherium. containing two species (B. bom- bifrons and B. cavifrons), was erected in the formal review (Leidy, 1852b: 12) on the basis of three diagnostic characteris- tics: 1. The OS frontis rises into a hump, or forms a prominent process, from the sides of which arise the hom-cores. 2. The latter arise above and posterior to the orbits, but considerably in advance of the inion, and curve downwards in their course, but do not turn up at the tips, as in Ovibos. 3. The species possess lachrymal depressions, or larmiers, as well developed as in cervine animals. The holotype (orB. cavifrons (ANSP 12995; Figures 5-8) is a relatively well-preserved cranium lacking only the distal tip of the left horn core. This cranium was secured by Thomas Kite of Cincinnati, Ohio, "in the hut of an Indian, in which it was used as a seat and he was informed it was found in a neighboring gravelly bluff, near Fort Gibson, on the Arkansas River" (Leidy, 1852b: 13) in what was then Indian Territory (now Oklahoma). Leidy's descriptions of this specimen and the B. bombifrons holotype are extensive and detailed, but he did not indicate expliciUy those criteria considered to be specifi- cally diagnostic. In the description of B. bombifrons, Leidy stated that the base of the cranium "is sufficiently well preserved to exhibit the peculiarities which associate it in the same genus with Bootherium cavifrons" (Leidy, 1852b: 18), but he did not identify those unifying characteristics. More- over, he did not mention one of the most conspicuous shared characters—the outward, downward, and forward curvature of the horn cores, clearly evident in both type specimens. (Leidy also did not comment on a small circular scar on the holotype of 5. cavifrons located on the dorsal surface of the right frontal, above the orbit, resulting from a pathologic condition. This injury healed before the death of the individual, and resculpting restored the bone surface to near-normal configuration. This condition is apparent in fig. 1, pl. Ill of Leidy, 1852b, and in our Figure 5.) Leidy's review was important in that it transferred Harlan's Bos bombifrons from the bison group to the musk ox group and placed both forms of fossil musk oxen in the same genus. However, Leidy did not establish the distinctiveness and integrity of Bootherium with the three diagnostic criteria he proposed. His first criterion actually consisted of two charac- ters—either a hump or a prominent process of the frontal region—and he did not establish any reason why these two expressions should be shared within the same genus, or that each was species specific (although this is implied by his descriptions of the two specimens). The second and third criteria are characteristics of all low-horned musk oxen and, therefore, are not generically diagnostic. Certainly, the inade- quacy of Leidy's generic diagnoses can be attributed to the fact that he had very little information about the structure of the skull of Ovibos moschatus. and that some of what had been published elsewhere was erroneous. Subsequent information about the morphology of Ovibos proved that the genus did have lacrimal depressions and its horn cores did not turn upward at die tips. In 1854, Leidy repeated his belief that the presence of lacrimal depressions in Bootherium separated that genus from Ovibos, and that B. cavifrons was further separated firom 0. moschatus by differing frontal characteristics—the presence of a deep fissure separating the bases of the horn cores in the latter whereas the horn core bases joined together and covered the entire length and breadth of the frontals in the former (Leidy, 1854). A new species, Ovibos maximus, was named by Richardson in 1852 on a damaged fossil axis (HM 90/2; Figure 9) found at Eschscholtz Bay, Alaska. Richardson compared the fossihzed axis with that of a young male Ovibos moschatus and concluded that the two bones, being of different sizes, represented different species. Without stating reasons, Richard- son considered that the fossilized axis might belong to the same taxon that Leidy had called Bootherium cavifrons and thus proposed the synonymy of 0. maximus and B. cavifrons (Richardson, 1852). Leidy quickly and effectively refuted Richardson's decision, and explicitly stated his opinion that there were no reasonable grounds for considering Ovibos and Bootherium to be synonymous (Leidy, 1854). Rutimeyer (1865) and Dawkins (1867) referred Leidy's genus Bootherium to Ovibos on the grounds that Ovibos, like Bootherium, possessed lacrimal depressions. Rutimeyer (1865) further considered B. cavifrons and B. bombifrons to be actually male and female of the same taxon, and created a new species, 0. priscus. to accommodate the pair. Leidy acknowl- edged that Bootherium might possibly belong within Ovibos. but he did not concur that the two forms were conspecific (Leidy, 1869). In 1872, Dawkins concluded that Leidy's B. cavifrons and B. bombifrons were conspecific and gave them the new combination Ovibos cavifrons (a name proposed earlier by Leidy), despite the fact that bombifrons was the senior specific epithet (Dawkins, 1872). The idea that the two forms were conspecific received mixed reception, some authors—including Lydekker (1885, 1898)—^agreeing with NUMBER 66 FIGURE 5.—Holotype of Bootherium cavifrons Leidy, 1852 (ANSP 12995), in dorsal view. FIGURE 6.—Holotype of Bootherium cavifrons Ijeidy, 1852 (ANSP 12995), in left lateral view. SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY 5cm FIGURE 7.—Holotype of Bootherium cavifrons Leidy. 1852 (ANSP 12995), in ventral view. 5cm FIGURE 8.—Holotype of Bootherium cavifrons Leidy, 1852 (ANSP 12995), in caudal view. NUMBER 66 FIGURE 9.—Holotype of Ovibos maximus Richardson, 1852 (HM 90/2), in A, ventral, B, caudal, and C, right lateral views. Rutimeyer and Dawkins, and others not (e.g., Rhoads, 1895; Osgood, 1905a; Allen, 1913). Rutimeyer (1865), however, had raised a major question about the relationship between B. cavifrons and B. bombifrons and, whether other authors agreed with him or not, none offered systematic arguments against either his or Dawkins' (1872) positions. Rhoads identified another species of musk ox from a small cranial fragment found in a cave in Durham County, Pennsylvania (ANSP 29; Figures 10, 11). The specimen was considered initially to be part of the left horn core and adjacent frontal of a new species of bison to which Rhoads gave the name Bison appalachicolus (Rhoads, 1895). Reexamination of 10 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY B 5 cm FIGURE 10.—Holotype of Bison appalachicolus Rhoads. 1895 (ANSP 29). in A, ventral, and B, rostral views. B 5 cm FIGURE 11.—Holotype of Bison appalachicolus Rhoads, 1895 (ANSP 29), in A. dorsal, and B. caudal views. the specimen led Rhoads to conclude that it belonged to a musk ox, not bison, so he changed the name to Ovibos (Bootherium?) appalachicolus (Rhoads, 1897). Osgood (1905a) reversed the trend toward synonymyzing Bootherium/Ovibos bombifrons and Bootherium/Ovibos cavi- frons when he created Scaphoceros to accommodate a nearly complete skull (USNM 2555; Figures 12-15) from the Yukon Territory that he described as Scaphoceros tyrrelli. Osgood U-ansferred Bootherium cavifrons to Scaphoceros as S. cavi- frons. Osgood rejected Riitimeyer's Ovibos priscus as spurious and reinstated B. bombifrons, leaving it in a genus separate from Scaphoceros. (Following publication, Scaphoceros was found to be preoccupied and was replaced by Symbos; Osgood, 1905b). Osgood (1905a: 181-183) justified and elaborated upon his decisions on the following grounds: Since bombifrons and cavifrons have been considered by several authors as being not only congeneric but conspecific. the establishment of a separate genus for each may appear surprising. While it may be possible, from examination of figures only, to construct a hypothesis to the effect that cavifrons represents the male and bombifrons the female of one species, it is inconceivable that any modem laxonomist would reach such a conclusion after comparing the original types. These are now before me and with them are specimens of S. tyrrelli and of both sexes and young of Ovibos moschatus. From comparisons of these il is evident that, unless the disparity between the sexes in this case was vastly greater than in the recent genus Ovibos, cavifrons and bombifrons do not respectively represent the male and female of one sf)ecies. Neither is bombifrons the young of any species, for the typ)e gives every evidence of maturity. The hom cores of the female Ovibos are essentially of the same character as those of the male. They are excessively flattened and directed downward close to the skull just as those of the male. Their bases approach each other over the top of the frontals increasing with age as in the male, the space between them being merely relatively greater than in the male. They are attached to the frontals only il is tme.bui ihis is the case with the immature male. Therefore the skull of the female has all the essential characters of the male but they are not as highly developed. The skull of Bootherium bombifrons, on the contrary, differs not in degree but in actuality from that of Scaphoceros cavifrons and 5. tyrrelli. The hom cores are not flattened but are actually round or as nearly round as may be in a rough surfaced structure; they are directed away from the skull at a different angle; their attachment to the skuU is entirely different; their bases do not approach each other in the least but on the contrary stand out from the skull on f>edicels and have a distinct burr as in Bison. The frontal region between the hom pedicels is not flattened as in the female Ovibos, but is elevated and convex. The under side of the skull of the type of bombifrons is much injured but one conspicuous character is shown in which it differs from all the olher species. This is found in the basisp>henoid which is not deflected but has its lower surface in the same horizontal plane as that of the basioccipital and it has a sharp median ridge. The hom cores of bombifrons are essentially like those of BLson except that they tum downward instead of upward. However, olher characters, notably the possession of deep lacrymal fossae, serve to distinguish it from Bison. In consideration of these various characters, the genus Bootherium with Bos bombifrons as the type seems to merit recognition. Thus far, only one specimen of this genus, the original type, has been found. Specimens of S. cavifrons and S. tyrrelli, however, have been secured at various localities among which are the following: Fort Gibson, Indian Territory; Council Bluffs, Iowa; New Madrid. Mo.; St Louis, Mo.; Benton Co., Mo.; Tmmbull Co., Ohio; Brook Co.. W. Va.; Pennsylvania; Anvik. Alaska, and Bonanza Creek, near Dawson, Yukon Territory. In this large number of specimens, if there were any females at all it is probable that there would be more than one. In order to give any semblance of certainty to the supfxjsiiion that the differences between bombifrons and cavifrons are sexual, it is necessary to show that these differences are relatively the same that obtain NUMBER 66 11 5cm FIGURE 12.—Holotype of Scaphoceros tyrrelli Osgood. 1905 (USNM 2555). in dorsal view. -"^5!^- \ 5cm FIGURE 13.—Holotype of Scaphoceros tyrrelli Osgood. 1905 (USNM 2555), in right lateral view. 12 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY 5 cm FIGURE 14.—Holotype of Scaphoceros tyrrelli Osgood, 1905 (USNM 2555), in ventral view. FIGURE 15.—Holotype of Scaphoceros tyrrelli Osgood, 1905 (USNM 2555), in caudal view. NUMBER 66 13 between the sexes in living species. This cannot be done, therefore it seems safer to treat the two animals as distinct. A much more reasonable assumption would be that S. cavifrons represenU the male and S. tyrrelli the female of one species. The present objection to this is the fact that both have not been found in the same region. From the foregoing excerpt, it can be seen that Osgood (1905a,b) considered the following facts important in separat- ing Symbos and Bootherium: 1. The dimorphism between skulls of Bootherium and Symbos (Osgood's Scaphoceros) was greater than that between skulls of female and male Ovibos moschatus. (No number of specimens available was given, nor were quantitative compari- sons of the dimorphism made.) 2. The shape of horn cores and the manner of their attachment to and development over the frontals differed qualitatively between Bootherium and Symbos. 3. The frontal region in B. bombifrons was elevated and convex, whereas that of Symbos "between bases of horn cores (was) surmounted by a prominent exostosis with an anterior bounding rim and a deep median excavation" (Osgood, 1905a: 174). 4. The basisphenoid of B. bombifrons was not defiected from the basioccipital, and it contained a sharp median ridge on the ventral surface, whereas the basisphenoid in Symbos was deflected conspicuously and did not exhibit a sharp median ridge. 5. The sex ratio of skulls recovered (Osgood implied) should closely approximate the sex ratio of living wild populations. (No allowance was made for differential preserva- tion or recovery of male and female skulls.) Although Osgood had few specimens with which to work, including only a single specimen of the form he returned to Bootherium bombifrons, his ideas nonetheless have had great influence upon the study of Pleistocene musk oxen during the 20th century, A new genus and species of musk ox was erected in 1906 by Gidley upon a partial cranium (USNM 5100; Figures 16-18) found during the construction of an irrigation dam near Black Rocks, McKinley County, New Mexico. Gidley (1906:165) acknowledged that "its incompleteness and poor condition make it a rather unsatisfactory type, yet there are sufficient distinctive characters preserved to warrant its description" which he gives as follows (Gidley, 1906:165-167): LIOPS. new genius, [sic] Generic characters.—Hom cores set wide apart and well t>ack, as in Ovibos. but much less drooping; continuous with the frontals laterally, with no burrs or rugosities at l>ase; smooth throughout. Parietals forming a large part of the occiput, which is high and narrow above. No tme lambdoidal crest. Foramen magnum about one and one-half times greater in diameter than in Ovibos. Occipital condyles set widely apart, with their borders continuous with the surrounding bones. Tympanic bone roughly triangular in shape, very smooth and flat, with no buUa, and tightly inclosed by the surrounding elements. Post-glenoid process reduced to a low rounded knob. LIOPS ZUNIENSIS. new species. Type, lop and back portion of skull, Cat. No. 51(X), U.S.N.M. colleaion. A striking feature of the portion of the skull preserved is its extreme smoothness. Its angles are free from rugosities, and there are no sharp or roughened processes even in the tympanic and mastoid region. The hom cores are relatively longer, less robust, and less drooping than in Ovibos or Simbos [sic], the latter sUnding directly intermediate between Liops and Ovibos in this respect Another striking feature is the position of the relatively large foramen magnum, which is confmed entirely to the back or occipital face of the skull. Gidley named the new genus Liops, with Liops zuniensis as the type species. Liops was twice preoccupied and, apparendy, Gidley tried to change the generic name to Lissops in 1908. By that time, however, Cossmann had suggested replacing Liops with the valid name Gidleya which remains in current use (Cossmann, 1907; Gidley, 1908). Gidley named two other extinct musk oxen in 1908 when he founded Ovibos yukonensis upon a partial skull (USNM 5728) from the Palisades along the Yukon River in Alaska and Bootherium sargenti upon a partial cranium (dorsal surface, part of left nasal, most of left horn core, and all of right horn core: GRPM 11-423-3101; Figures 19-21) from Moorland Swamp near Grand Rapids, Michigan. The Alaska specimen was correctly assigned to Ovibos and will not be considered further in this paper. Of Bootherium sargenti, Gidley (1908:683-684) said: Species-characters.—Size about two-thirds that of Ovibos moschatus. somewhat larger than B. bombifrons; hom-cores comparatively large, well rounded, long and slender, hom-cores at base horizontally directed at right angles to the skull as in Plate LDC, fig. a, but curving downward and forward in graceful semi-spirals, ending in slender anteriorly directed tips (see Plate LIX, fig. b); orbits comparatively large, depressed below the arching frontals, with thin gently shelving borders.not tubular as in Ovibos. In the general form and contour of the skull and hom-cores this sf)ecies, together with B. bombifrons. is strikingly different from other known species of the Ovibovenae [sic]. The tyjje of B. sargenti. compared with that of B. bombifrons, shows the following resemblances: (1) The fragment preserved indicates a skull but little larger in size and of the same general proportions; (2) the position, form, and contour of the orbits as well as (3) the general appearance of the facial and posterior portions of the cranium (see Plate LIX, fig. c) are essentially alike. The hom-cores are also similarly placed, but the differences in their relative size, form and proportions are very marked. In B. sargenti the base of the hom-core is relatively heavier, is angular in front, and its superior border approaches much nearer the median frontal suture than in fl. bombifrons. In addition, charaaeristic rugosities and markings on the frontals indicate that the hom-covering extended much beyond the hom-core base, nearly or quite meeting the one from the qjposite side in the median line. In B. bombifrons the inter-hom space was apparently covered by a wide skin-band as in Bos. The hom characters seem sufficiently different to separate these species generically, but the other cranial characters denote generic relationship. Moreover it is possible that the extreme difference in type of hom-core may be due in pan at least to difference in sex. NOTES ON THE RELATIONSHIPS OF THE GENUS BOOTHERIUM LEIDY. The genus Bootherium has for some time been considered as closely allied to Ovibos, and by some authorities as synonymous with that genus. But in 1905 Mr. W. H. Osgood re-defined Bootherium, selecting 6. bombifrons as the type, and transferred the remaining species, B. cavifrons, to a new genus, Scaphoceros, of which S. tyrrelli is the type. In the pHiblication first cited Mr. Osgood has shown with good reason the untenabUity of the opinion held by Riilimeyer and others regarding the types of B. bombifrons and Ovibos (Symbos) cavifrons, which they considered the female and male, respectively, of the same, or closely related, species. He has also pointed out that the type of B. bombifrons does not represent an immature male, but a fully adult 14 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY ^'■■^■^■"^ 5cm FIGURE 16.—Holotype of Liops zuniensis Gidley. 1906 (USNM 5100), in dorsal view. individual. By an analogy similar to that employed by Osgood it is equally clear that the typ)e of B. sargenti can not be referred on these grounds to any species of Ovibos or Symbos. The validity of the genus Bootherium therefore seems to be weU established. The separation of the two species originally referred to this genus permits the study of its relationships in a new light. As now known the genus presents quite as many bovine as ovibovine characters,and if referable to the Ovibovinae it is far removed from the olher known genera of the group. From present evidence it seems probable that the finding of more complete material will show that, whether, generically distinct from each other or not, the species B. bombifrons and B. sargenti represent a distinct group, or subfamily, of the Bovidae. To this group may belong also the genus Lissops Gidley. Since, however, so little is known of the species of the group as a whole, owing to the lack of good material, il would be unwarranlablc to separate them al present from the Ovibovinae. A new species, Symbos australis, was erected upon three teeth (RM2, Lm2, Lm3: AMNH 11828; Figure 22) collected by Barnum Brown in 1904 at Conard Fissure, Newton County, Arkansas (Brown, 1908). Other ovibovinc remains collected at the fissure and referred to S. australis included three premolars, an atlas, one or two proximal phalanges, and an ungual phalanx. (Brown reported one proximal phalanx on p. 203, but referred to two on p. 204. Wc arc aware of only one such element among the ovibovinc material from Conard Fissure described by Brown.) Brown determined that the three molars were too large to belong to Ovibos, and sent the RM2 and the atlas to Osgood for comparison with the type specimens of fi. bombifrons, S. cavifrons, and 5. tyrrelli. Osgood concluded (Brown, 1908:203): 'Ihcy evidently belong to a species different from the one on which 1 am working, Symbos (Scaphoceros) tyrelli [sic]. The tooth is somewhat smaller in my specimen, although the difference in condition makes it difficult to get an exact comparison. The atlas is much loo small to fit on the condyle of my specimen or on that of the type of Ovibos cavifrons Leidy, with which I compared it in Philadelphia, 'llie type of O. bombifrons is also in Philadelphia, but is very different from any of the other specimens and hardly needs to be considered. Allen rcvi.scd the North American musk oxen in 1913. Although his U-catmcnt of the fossil taxa was relatively superficial, he followed O.sgood (1905a) in considering Bootherium and Symbos to be valid separate genera (Allen, 1913:209), which, in the light of present knowledge, prove lo be not only not congeneric, but not very closely allied. This interesting discovery is due to Osgood, whose paper on the status and relationships of Bootherium is entitled to high praise. Allen (1913:210) considered that the following character NUMBER 66 15 5 cm FIGURE 17.—Holotype of Liops zuniensis Gidley, 1906 (USNM 51(K)), in nght lateral view. xV^-.:- Tjjj^ 5 cm FIGURE 18.—Holotype of Liops zuniensis Gidley. 1906 (USNM 51(X)). in caudal view. 16 SMITHSONIAN CON IRIBUIIONS TO PALFORIOI OGY 5cm FIGURE 19.—Holotype of Bootherium sargenti Gidley. 1908 (GRPM 11 -423-3101). in dorsal view. FIGURE 20.—Holotype of Bootherium sargenti Gidley. 1908 (GRPM 11-423-3101), in right lateral view. differences satisfactorily separated Bootherium and Symbos: 1. Bootherium had Bison-like (round or sub-rounded) horns, whereas the horns of Symbos, unlike those of Bison, were flattened. 2. Bootherium had a smooth and sharply convex dorsal frontal surface, whereas the same surface in Symbos was an elongated trough covered with exostosis. 3. The venU'al surface of the basisphenoid was continuous with that of the basioccipital in Bootherium, and it supported a high ridge, whereas in Symbos the venu^al surface of the basisphenoid was deflected from that of the basioccipital. 4. Bootherium had small but deep and sharply defined lacrimal fossae, whereas those of Symbos were shallow and less well defined. 5. Bootherium was much smaller than Symbos, based on the assumption that the holotype of Bootherium bombifrons was the skull "of a very old male, with all the sutures of the preserved part of the skull wholly obliterated by anchylosis." 4. NUMBER 66 17 FIGURE 21.—Holotype of Bootherium sargenti Gidley, 1908 (GRP.M 11-423-3101), in caudal view. Allen went on to say, however, that Bootherium did resemble Symbos in some characters, including those of the occipital condyles, the occipital plane, and the length and depth of the skull. Nonetheless, he concluded that Bootherium and the recendy described shrub ox Preptoceras (Furlong, 1905) were "not closely related, but more nearly so than is eiUier to any other known genus" (Allen, 1913:212), a decision that was based more upon the superficial impression of horn core similarity than a thorough systematic comparison of cranial sUTicture in the two forms. Bootherium sargenti, however, was Uansfcrred from Boo- therium lo Symbos by Allen (1913:215). The species was founded on an imperfect skull . . . found in a swamp near Grand Rapids, Michigan. In the description comparison is strangely made with Bootherium bombifrons, with which it shares no essential feature. The homcores are attached to the skull as in the female of Ovibos, with about the same relative area of exostosis extending from the base over the lateral third or more of the frontals and not, as in Bootherium, supported on a pedicel and terminating in a burr as in Bison. The relationship of Bootherium sargenti is entirely with Symbos, and well fulfills the conditions that would be expected in the female of S. cavifrons. Allen did not specify what conditions would be expected in the female of Symbos, nor did he consider the possibility, when discussing 5. tyrrelli, that this form could be the female of S. cavifrons. Allen did, however, refer S. australis to S. cavifrons "as probably representing the female of that species" (Allen, 1913:215). Allen upheld the distinctiveness of Gidleya (which he discussed under Liops). His description of the specimen (Allen, 1913:216) reads, in part, as follows: The dense smooth natural surface of the bone is preserved over the greater part of the dorsal aspect of the skuU, except laterally in the postorbital region; the surface elsewhere consists of the cancellous stmclure of abraded bone, the abrasions being in places quite superficial, as in the case of the homcores and upper surface of the skull, and elsewhere so deep as to greatly obscure or wholly obliterate imp)ortant features, as the condyles, the characters of the mastoid and tympanic regions, and the occipital angles. The unabraded dorsal surface of the skull shows that it must have been that of an old animal, and probably that of a male, the sutures being entirely obliterated by anchylosis. Continuing, Allen (1913:216) acknowledged that Liops was nearer Symbos than any other taxon, as Gidley (1906) had suggested. A striking difference between Liops and both Symbos and Ovibos is the smoothness of the surface of the basal portion of the homcores and the entire absence of exostosis between the hom bases over the top of the skull, which is here smooth, with the same dense surface as that of the interorbital and preorbital [>ortions. The ventral, caudal, and lateral surfaces of the skull, Allen staled, were greatly abraded. Hay (1915) erected Bootherium nivicolens on the basis of a partial cranium (USNM 23241; Figure 23) from Eschscholtz Bay, Alaska, that possessed character states he considered different from or midway between B. bombifrons and B. sargenti. Most important among these were the outward orientation of the horn cores and the way in which they had burrs situated on distinct pedicels, as in B. bombifrons. The specimen also had exostosis extending over the dorsal surface of the pedicel onto the lateral edges of the dorsal surface of the frontals, as in B. sargenti. Hay also rejected Allen's syn- onymizing of B. sargenti and S. cavifrons, arguing that it was unlikely that, among the 25 or so skulls known and assigned to S. cavifrons, only one would have belonged to a female. More likely. Hay maintained, the smaller specimens with exostosis- covered frontals and more feebly developed horn cores that were assigned to Sytnbos probably represented females. Hay (1915:527) also pointed out that one of the characteristic features of Bootherium, as identified by Allen, was the abmpt downward slope of the dorsal outline of the skull posterior to the hom cores. The type skull of fl. sargenti has a slope of the same region which lacks but a few degrees of being equal to that found in Bootherium .... Elsewhere, Hay (1915:527) slated: Xotwiihslanding the immense development of the hom-cores of the males of Symbos cavifrons, there is no such elevation of the region behind the orbits as we see in the case of 6. sargenti.... Four new species of Pleistocene musk oxen were named 18 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY D .V- . ^ i ^ I J . ^ f^r ■^ B H 'ifea »v*'. FIGURE 22.—Type of Symbos australis Brown, 1908 (A.MNH 11828): R.M2 in A, lingual, B, occlusal, and C. labial views; ljn2 in D, labial, E, occlusal, and F, lingual views; and ljn3 in G.labial, H, occlusal, and I, lingual 1cm NUMBER 66 19 5 cm FIGURE 23.—Holotype of Bootherium nivicolens Hay, 1915 (USNM 2324), in dorsal view. B 1 cm FIGURE 24.—Holotype of Symbos promptus Hay, 1920 (USNM 9120), in A, labial. B, occlusal, and C, lingual views. from 1920 to 1942. Symbos promptus was founded upon an upper left third molar (USNM 9120; Figure 24) from near Afton, Oklahoma, the diagnosis given as simply "Upper molars with the external styles less strongly developed than in S. cavifrons; the fossetts less angular" (Hay, 1920:125). Barbour (1934) named Symbos convexifrons upon a nearly complete skullcap and right horn core (UNSM 39001; Figures 25-28) found in the North Prong Quarry in southern Cherry County, Nebraska. "The brow of this specimen is notably convex, and this, coupled with the slope, length, and sweep of the horn cores, constitutes the main feature of this new sjjecies" (Barbour, 1934:295). ?Ovibos giganteus was created by Frick (1937) with a large right humerus (F:AM 30498; Figure 29) from near Fairbanks, Alaska, as the holotype. Frick probably was stimulated to name this species after seeing what was (probably erroneously) then identified as the humerus of a 20 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY 5 cm FIGURE 25.—Holotype of Symbos convexifrons Barbour. 1934 (UNSM 39001), in dorsal view. FIGURE 26.—Holotype of Symbos convexifrons Barbour, 1934 (UNSM 39001), in right lateral view. NUMBER 66 21 5 cm FIGURE 27.—Holotype of Symbos convexifrons Barbour. 1934 (UNSM 39001), in ventral view. FIGURE 28.—Holotype of Symbos convexifrons Barbour, 1934 (UNSM 39001), in caudal view. giant ovibovinc from near the American Falls Reservoir, Idaho (Frick, 1937). The last North American musk ox to be named was Bootherium brazosis, erected upon a damaged partial cranium with partial horn cores (TAMC 2553; Figures 30-32) that was found in Brazos County, Texas (Hesse, 1942). Hesse considered his specimen to resemble B. bombifrons more than any of the other nominal taxa, but it was differentiated from B. bombifrons by minor differences, some of which undoubtedly resulted from Hesse's having misoriented the specimen during study. Hibbard and Hinds initiated a resurgence of attention to the relationship among Bootherium and Symbos species that has 22 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY B 5 cm FIGURE 29.—Holotype of ?Ovibos giganteus Frick, 1937 (A.MNH F:AM 30498), in A, cranial and B, caudal NUMBER 66 23 5 cm FIGURE 30.—Holotype of Bootherium brazosis Hesse, 1942 (TAMC 2553). in dorsal view. 5 cm FIGURE 31.—Holotype of Bootherium brazosis Hesse. 1942 (TAMC 2553), in ventral view. FIGURE 32.—Holotype of Bootherium brazosis Hesse. 1942 (TAMC 2553), in caudal view. 24 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY continued to the present. These authors stated "it is very likely that Bootherium is the female woodland musk ox since all specimens of Symbos based on skulls are considered as those of bulls" (Hibbard and Hinds, 1960:107). Harington (1961) disagreed, instead favoring Hay's (1915) separation of the two genera. Semken, Miller, and Stevens (1964) apparendy adopted Allen's views, concluding Ihat Bootherium bombifrons was not a female Symbos but that B. sargenti, which was erroneously placed within Bootherium. might be. Ray evaluated the status of fi. appalachicolus and B. brazosis and concluded that both were better considered junior synonyms of B. sargenti than separate species. Ray also acknowledged die merit of the suggestions that B. sargenti could represent the female of S. cavifrons, and stated that failure to allow for weathering damage had vitiated much comparison and unwarranted differentiation in the past (Ray, 1966a,b). By 1977, Harington had come to accept the synonymy of S. cavifrons and B. sargenti. but "presuming the specimen of Bootherium bom- bifrons does not represent an abnormal individual, that species is probably not closely related to Bootherium sargenti or Symbos cavifrons" (Harington, 1977:880-881). Regarding the synonymy of B. sargenti and S. cavifrons. Harington (1977:882-883) wrote: .. .the evidence supporting this view is very strong. The similar basic conformation of the homcores (as far as orientation and curvature are concerned); the smaller, thinner-roofed cranium; and the broad space between the homcore bases in the former species parallel the differences between male and female Ovibos moschatus: thus Sargent's muskox resembles what a female Symbos cavifrons would be expected to look like. In addition, Bootherium sargenti and Symbos cavifrons had simUar geographic and habitat preferences (e.g. Alaska, Yukon Territory, Indiana, Michigan, Virginia (using Bootherium sp.), Utah. Missouri and Nebraska), and sometimes even from the same site and deposit (e.g. near Great Salt Lake, Utah, in the Bonneville sands and gravels (Stokes and Hansen 1937, p. 63), and evidently in the Goldslream Formaticm of Wisconsin age near Fairbanks, Alaska). In the conterminous United States, both species are cwicentrated within the same latitudinal range—usually south of the late Wisconsin fossil localities of the tundra muskox. Ovibos moschatus. Unlike Soergelia, Praeovibos, and Ovibos, which had Holarctic distributions during the Pleistocene. Bootherium sargenti and Symbos cavifrons are only known from North America. Geochronologically. both species appear during the Dlinoian and become extinct near the close of the Wisconsin glaciation. In the preceding excerpt, Harington advocated the compari- son of patterns of bodi sexual dimorphism (without demanding exact duplication of character states between the sexes, as had earlier writers) and geographic and chronologic distribution in assessing the relationships between these two taxa. Nelson and Madsen (1978) and Nelson and Neas (1980) recognized Bootherium and Symbos as distinct genera, but acknowledged that the status of most nominal species widiin Bootherium was still uncertain. Following Harington, Kurten and Anderson (1980) placed B. sargenti in S. cavifrons while allowing B. bombifrons to stand, although widi the admonition Uiat "The status of the genus Bootherium is in doubt; it has often been considered to be congeneric widi Symbos, but this is undemonstrated" (Kurten and Anderson, 1980:334). Nelson and Madsen (1987) concurred with Harington (1977) and Kurten and Anderson (1980) in synonymizing 5. cavifrons and B. sargenti (as S. cavifrons) while maintaining the validity of 5. bombifrons as a distinct taxon. Neas has reversed his opinion held in 1980 diat Bootherium and Symbos were distinct taxa and, with McDonald and Ray, views all nominal taxa within Bootherium. Symbos. and Gidleya as synonyms o( Bootherium bombifrons (McDonald, 1986; Neas, 1986; McDonald and Ray, 1987, in press; Neas and Hoffmann, 1987; Neas and Parker, 1987). The Content of Bootherium, Symbos, and Gidleya Twelve nominal species of autochthonous North American musk oxen were erected between 1825 and 1942 widiin, or are referable to, the genera Bootherium, Symbos. and Gidleya. Bootherium. in the modern sense, at one time or another contained die five species B. f= Bos) bombifrons (1825), B. (= Bison) appalachicolus (1895), B. sargenti (1908), B. nivicolens (1915), and B. brazosis (1942). Ray placed B. appalachicolus and B. brazosis in synonymy with B. sargenti. and Harington did the same with B. nivicolens (Ray, 1966a,b; Harington, 1977). Allen (1913) attempted to place fi. sargenti in synonymy widi Symbos cavifrons, but this decision has met with mixed acceptance and the status of B. sargenti must, therefore, be reevaluated. Bootherium bombifrons was consid- ered conspecific with Symbos cavifrons by Rutimeyer (1865), Dawkins (1872), and dieir followers during the late part of the 19th century, but the two taxa have been considered generically distinct for most of their history. At present, most workers tend to recognize only a single sjjecies of Bootherium. B. bom- bifrons, while acknowledging that uncertainty in the taxonomic status ofthe group, and especially of fi. sargenti. clearly exists. Symbos has contained six species during its 135-year history, including Symbos (= Bootherium) cavifrons (1852), 5. tyrrelli (1905),5. australis (1908),5. promptus (1921),5. convexifrons (1934), and S. (= ?Ovibos) giganteus (1937). Allen (1913) placed S. australis in synonymy with 5. cavifrons. and Jakway (1961b) did the same for S. convexifrons. Among the remaining four species, only S. cavifrons is in use. Gidleya has contained only a single species, G. (= Liops, Lissops) zuniensis (1906). Although diis taxon has been recognized by several paleontologists since it was established, it has never received validation as a viable taxon. The relationship of Ovibos maximus is uncertain; this taxon probably is properly placed within Ovibos. aldiough conclusive determination of diat fact must await further comparative study of die axes of die various ovibovinc genera. At presenL then, there are seven nominal species standing within the three genera, five of the 12 named species having been placed previously, without subsequent rejection, in synonymy (Appendix II). The standing species are B. bombifrons (Harlan, 1825); 5. cavifrons (Leidy, 1852); S. tyrrelli Osgood, 1905; Gidleya zuniensis (Gidley, 1906); B. sargenti Gidley, 1908; S. promptus Hay, 1920; and S. (= ?Ovibos) giganteus (Frick, 1937). We are of the opinion that B. (= Bos) bombifrons, fl. NUMBER 66 25 (= Bison) appalachicolus, B. sargenti, B. nivicolens. and B. brazosis are conspecific. Bootherium appalachicolus. B. bra- zosis, and B. nivicolens have been synonymized with B. sargenti by Ray (1966a,b) or Harington (1977). Bootherium bombifrons and B. sargenti may be considered conspecific on the basis of morphological similarity of their cranial and horn core structure, especially (1) horn core size, shape, orientation, and cranial attachment, and (2) frontal configuration (Figures 1-3,19-21). Aside from modest differences in the exact shape and size of character states—all of which can be attiibuted more satisfactorily to individual variation than to taxonomic distinctiveness—diere are no significant morphological fea- tures present on the holotypes of these taxa sufficient to justify their taxonomic separation. Despite numerous statements and implications to the contrary, the differences in die size, shape, surface details, and orientation of character states between B. sargenti and S. cavifrons are much greater, qualitatively and quantitatively, than are those between B. bombifrons and B. sargenti. Detailed comparisons of character state differences, and discussions of die reasons for diose differences among individuals and between sexes and taxa, are presented below (pp. 25-51). We also are of die opinion dial all nominal species widiin Symbos (including ?Ovibos giganteus) should be considered conspecific. Symbos convexifrons was established upon the dorsal half of a cranium and nearly complete right horn core (Figures 25-28). The dorsal surface of the right half of the braincase is preserved, whereas the frontoparietal sinuses are exposed on the left side (in ventral view). Several small fragments of bone have been reattached to die cranium with plaster. The diagnostic characters for S. convexifrons were the "notably convex" dorsal surface of the cranium, coupled with die slope, length, and sweep of the horn cores. Barbour (1934:295) also noted that "die horns rise from die side of die head." Jakway (1961b), following Frick (1937:562), referred S. convexifrons to S. cavifrons on die grounds diat (1) all odier remains of Symbos from the North Prong Quarry and other sites in Nebraska were assignable to S. cavifrons, and (2) die "convexity between the homcores, the supranuchal, and the internal surface of the cranium" seemed to be pathological instead of taxonomically significant variations (Jakway, 1961b:115). The holotype of S. convexifrons appears to represent a mature individual, based on the size of the specimen, the density of die horn core, and the absence of unfused sutures. This specimen is unusual, however, in several respects. (1) The horn core possesses a distinct burr line which delimits the horn core proper from the exostosis covering die intercornual surface of die cranium. (2) The base of the remaining horn core is situated abnormally low on die side of the cranium and die horn core emanates from die cranium at an unusually low angle but, based on differences in the planes of secondary bone medial to die bases of the horn cores, the left core was neither situated as low nor did it emanate at such an unusually low angle. (3) The longitudinal profile of die dorsal surface of die cranium is flexed to an unusual degree. (4) The transverse profile of die dorsal surface of the cranium is asymmetrical. (5) The exostosis is distiibuted radier thinly and uniformly across die intercornual region. The exostosis extends rostrally only a short distance from die level of the horn cores and caudally only a short distance onto the parietal surface. No rim-like or shelf-like build-up of secondary bone is present at either die caudal or rostral edges, respectively, of the exostosis. (6) The horn core attaches only to die frontal bone. We are of die opinion diat die asymmetry of die ti-ansverse profile of the dorsal surface of die cranium in the holotype of S. convexifrons is probably attributable to a pathology of die right horn core. We were unable to locate any other natural asymmetries and consider die cranium—excluding the horn padiology and subsequent atypical development of the intercor- nual boss and associated exostosis—to be normally developed. Two fragments of the cranium—one including the insertion surface for the left M. semispinalis capitis and the other a portion of die left frontoparietal bones—were reattached at die wrong location during restoration. These artificial errors probably contributed to Jakway's recognition of asymmetries in die supranuchal ridge and internal surface of the cranium. Aside from the asymmetry of the dorsal surface of the cranium in transverse profile and die low level of emanation of the right horn core, die holotype of S. convexifrons sti-ongly resembles a cranium with horn cores (USNM 23548) found in the Ohio River below Augusta, Bracken County, Kentucky. The Kentucky specimen has a similarly flexed (longitudinally) dorsal surface of the cranium and—over the caudal part of the dorsal surface—a similar pattern of exostosis developmenL Odier crania identified as S. cavifrons (e.g., AMNH F:AM 33129 from Lower Coldstream, Fairbanks DistricL Alaska) also exhibit marked flexion of the longitudinal profde of the dorsal surface, as do specimens identified as Bootherium (e.g., ANSP 994 from Big Bone Lick, Kentucky; USNM 23264 from Saltville, Virginia). We believe that the holotype of S. convexifrons represents a pathologic adult male with abnormal placement and orientation of the right horn core and the development of the keratinous boss over the dorsal surface of the cranium. When die evidence of this pathologic condition is put aside, we see nothing to justify recognition of S. convexifrons as a valid taxon distinct from S. cavifrons. Allen's (1913) referral of S. australis (founded upon three associated teeth, AMNH 11828; Figure 22) is reasonable, in part, but not conclusive. The RM2 and Lm2 are undifferen- tiable from the equivalent teeth in other specimens of S. cavifrons. The Lm3, however, differs somewhat from odier m3s assigned to S. cavifrons. Specifically, the shape of the enamel border of the entoconulid-hypoconuhd is conspicu- ously rounded (i.e., elliptical) in the holotype of S. australis (Figure 22). The same character in other m3s of Symbos is more complexly shaped, with an undulating lingual surface and a distinct terminal styUd at the caudal end. The unusual shape of die entoconulid-hypoconulid in the S. australis m3 could represent (1) an atypical tooth of Symbos, (2) a character state 26 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY typical of early Symbos diat subsequently changed to the form found in late Wisconsinan Symbos, or (3) a toodi belonging to anodier large ovibovinc, such as Euceratherium. (The teedi that constitute die type of 5. australis are within the size range of die teeth of Euceratherium. The breaddi of die occipital condyles in the holotype of Euceratherium collinum is 103 mm, and die breadth of die cranial articular cavities in die atlas from Conard Fissure is 107 mm.) Dental variation within die ovibovines, and especially shrub oxen, is not known sufficiently well to conclusively determine die identity of die teedi of die S. australis type specimen. The possibility exists diat die diree teeth forming the type of S. australis might represent more than one animal and, perhaps, more dian one taxon. We feel that S. australis should be left in synonymy with S. cavifrons for now, but we also recognize that some or all of these teeth might, at some later date, be assignable to one or more odier taxon/taxa. Symbos tyrrelli was erected upon a nearly complete but unusually small skull (USNM 2555, die smallest skull for which quantitative information was available in 1905; Figures 12-15) from Bonanza Creek, Yukon Territory, at a time that few Symbos crania were known (Osgood, 1905a). Osgood used diree character states to differentiate die Bonanza Creek specimen from S. cavifrons: die relatively small size, resfricted development of exostosis, and shallow depth of the frontoparietal region. The holotype does not, however, possess any character of taxonomic significance dial would set it apart, quahtatively or quantitatively, from S. cavifrons as that species is now conceived, based upon more than 150 known specimens. Specifically, the size of the skull and its various characters, the degree of exostosis development, and the depth of die frontoparietal sinus region of die Bonanza Creek specimen are all widiin the range of variation known for S. cavifrons. We feel that S. tyrrelli is conspecific widi S. cavifrons. Symbos promptus was founded upon an upper left diird molar (USNM 9120; Figure 24) from near Afton, Oklahoma (Hay, 1920). This tooth is well worn; about one-third of die crown remains. The two criteria upon which the species was founded were the less extensive lateral development of die buccal styles and die more crescentic, less angular shape of die enamel border of the internal fossettes, relative to other known teeth of Symbos. Apparendy, Hay compared die tooth from Afton only with two other specimens—die upf)er right second molar in die S. australis type toodi series and die superior molars in a Symbos cavifrons skull from near Manchester, Michigan (Hay, 1920). The tooth from Afton was worn more dian either of the teeth widi which it was compared. As wear progresses in the superior molars of Symbos cavifrons, the lateral extension of the buccal styles is reduced and the shape of die enamel border of the fossettes changes from a more angular or chevron shape to a more rounded or crescentic shape. The toodi from Afton is widiin the size range of die tccdi of Symbos cavifrons (the size of the M3 of Bootherium is still unknown), and appears to be nodiing more dian a relatively well worn loodi assignable to that taxon. We feel diat S. promptus is properly a junior synonym of S. cavifrons. Frick erected ?Ovibos giganteus upon a right humerus (AMNH F:AM 30498; Figure 29) from an unidentified site in die Fairbanks District, Alaska (Frick, 1937). Frick gave no reason for considering the holotype to be other than Symbos or Bootherium. nor were any diagnostic criteria given in the type description. Some of die incentive for establishing diis species probably came from the discovery by J.W. Gidley, in 1929 and 1930, of what were considered to be remains of a giant ovibovinc from soudiern Idaho. Two crania of Symbos cavifrons possessing unusually large horn cores were found in a gravel quarry near the American Falls Reservoir, and these might have inspired some notion of gigantic ovibovines fix)m dial area. In addition, a robust radioulna and an unusually long humerus from soudiern Idaho were also identified as ovibo- vines. The humerus was shown to Frick by Gidley, and might have been the specific stimulus that led Frick to recognize an extinct taxon of giant musk ox from Alaska. Actually, die radioulna (figure 26 in Gidley, 1930) was probably Bison latifrons and die humerus was probably Camelops sp. In the collection of fossils from soudiern Idaho in the National Museum of Natural History (USNM), there are no ovibovinc limb bones, but diere is a large Bison radioulna (USNM 13710) that can be identified as the one in Gidley's figure 26 and a Camelops humerus (USNM 392114) of simUar size to that mentioned by Frick from Idaho (lengdi estimated by Frick: 470 mm; length of USNM 392114, whose proximal end is abraded slighdy: 458 mm). The humerus from near Fairbanks is ovibovinc, it is similar in size and morphology to the humerus of an associated partial skeleton (F:AM A-204-4254) of Symbos cavifrons from Little Eldorado Creek, Alaska, and we refer it to that species. Gidleya zuniensis was founded upon an extensively abraded cranium and carries only a single specimen, its holotype (USNM 5100; Figures 16-18). The seven standing species of musk oxen within die genera Bootherium. Symbos, and Gidleya have been reduced to diree species in three monotypic genera: Bootherium bombifrons, Symbos cavifrons. and Gidleya zuniensis. In the next two sections of diis paper, we evaluate die relationships between die genera (1) Gidleya and Symbos and (2) Bootherium and Symbos. respectively. The Relationship between Gidleya and Symbos The genus Gidleya (= Liops. Lissops. sensu Gidley, 1906, 1908) was founded upon a partial skull (USNM 5100; Figures 16-18) found in 1905 during die construction of an irrigation dam on die Zuni Indian Reservation at Black Rocks, McKinley County, in western New Mexico. This specimen, along with a small collection of other vertebrate fossils (including Mam- muthus columbi. Equus sp. indet., and camel) was salvaged by the site engineer John B. Harper and transferred to the Smidisonian Institution through die efforts of F.E. Leupp, Commissioner of Indian Affairs. NUMBER 66 27 The musk ox specimen from Black Rocks had been damaged extensively by abrasion. Most of die occipital surface is missing, bodi of the horn cores have been shortened and reduced in diameter by the loss of surface bone, the dorsal surface of the cranium has been abraded to die extent that several of die parietal and frontal sinus cells are exposed. The remainder of the specimen has had most surface detail completely removed or extensively rounded by abrasion. Most of the dense outer bone appears to have been removed prior to the time this specimen was unearthed, leaving cancellous bone exposed over much of the surface. The cells of die cancellous bone apparendy filled with a fine grained chemical precipitate diuing or prior to the last phase of abrasion, giving die specimen an unnaturally smoodi surface. Other parts of die remaining siu-face are not smooth, probably because of damage incurred during die excavation, drying, and handling of die specimen. (The smoodi siu^face bone easily separates from die deeper bone at a nearly uniform depth.) The poor condition of die specimen was acknowledged by Gidley in his type description of the genus (as Liops) and its sole species G. (= Liops) zuniensis (Gidley, 1906). The characters that Gidley considered to be taxonomically signifi- cant were all products of post-mortem alteration—die "extreme smoothness" of die skull, its angles "free from rugosities," and absence of "sharp or roughened processes even in die tympanic and mastoid region," horns "relatively longer, less robust and less drooping" than in Ovibos or Symbos. and "die relatively large foramen magnum, which is centered entirely to the back or occipital face of die skull" (Gidley, 1906:166-167). Widi uncharacteristic oversight, Allen apparentiy regarded the abraded dorsal sm^face as being natural, writing "The dense smooth natural surface of the bones is preserved over the greater part of the dorsal aspect of the skull, except laterally in the postorbital region" (Allen, 1913:216). Allen correctiy pointed out that Gidley had failed to recognize the extent to which the caudal, lateral, and ventral surfaces of the skull had been damaged, and that as a result Gidley had inu-oduced errors into his generic diagnosis. Allen also noted that, while the Black Rocks specimen was about one-half the size of Symbos (Allen personally had examined only one Symbos specimen, AMNH 14365 from Hebron, Porter County, Indiana), its proportions were similar to those of Symbos (Allen, 1913). Patterson commented fiu^ther upon the insecure foundation of Gidleya in some unpublished notes that he prepared after he received a cranium of Symbos cavifrons from near Grand Mesa, Delta County, Colorado (McDonald, 1985a), diat was abraded somewhat like the type specimen of Gidleya zuniensis: On comparing this (i.e., the Colorado) fragment with specimens of Ovibos and with published figures and descriptions of the various Pleistocene forms I was strack by its resemblance to Gidley's "Liops" zuniensis (Gidleya zuniensis Cossman [sic], 1907). The type of this species was found near Zuni. New Mexico and was also uncovered, curiously enough, during excavations for a dam. Dr. C. Lewis Gazin kindly arranged for a loan of the specimen, U. S. N. M. no. 5100. Careful comparison of the two specimens permits no doubt that they are congeneric and little doubt, despite the somewhat larger size of the Colorado specimen, that they are conspecific. The type is about as waterworn as is possible for a fossil to be and yet retain some character, so scoured is it that some portions could almost be described as polished. It is quite evident that Gidley completely underestimated the extent of the abrasion and was therefore badly led astray. I quote his diagnosis here, placing brackets around those "characters" that are certainly artificial. Hom cores set wide apart and well tiack, as m Ovibos. but much less drooping, [continuous with the frontals laterally, with no Ixirrs or rugosities at base; smooth throughout.] Parietals forming a large part of the occiput, which is high and narrow above. [No tme lambdoidal crest.] Foramen magnum [about one and one half times] greater in diameter than in Ovibos. [Occipital condyles set widely apart, with their borders continuous with the surrounding bones.] Tympanic bone roughly triangular in shape [very smooth and flat with no bulla and tightly inclosed by the surrounding elements. Post-glenoid process reduced to a low rounded knob]. None of the characters that remains after this elimination distinguishes Gidleya from Symbos (Osgood 1905A and B) (Notes attached to letter: B. Patterson to C. E. Ray, 4 November 1968). Gidley originally described diis taxon under the name Liops (Gidley, 1906), but diat name was preoccupied so Cossmann (1907) suggested replacing it with Gidleya. Gidley, perhaps realizing Liops was unavailable and not knowing of Coss- mann's replacement, inserted—widiout explanation or refer- ence to a specific specimen—the name Lissops in a subsequent manuscript when he alluded to die genus based on the Black Rocks specimen (Gidley, 1908). Allen (1913) and Troxell (1915) retained use of Liops. but most subsequent authors have referred to die genus as Gidleya (e.g.. Hay, 1922, 1924; Ryziewicz, 1933, 1955; Kretzoi, 1942; Harington, 1961; Romer, 1966). Prick's placement of Gidleya zuniensis in Ovibos, as 0. zuniensis. was done widiout explanation and has had no following (Frick, 1937). Although, based upon usage of die name, Gidleya has been considered a viable genus, the Uterature gives no indication that any specimen odier than die holotype of G. zuniensis has ever been placed in the genus. If allowance is made for die extensive loss of bone by battering and fine abrasion, the cranium upon which Gidleya was founded is undifferentiable from crania that would routinely be assigned to S. cavifrons. As Allen (1913) suggested and Patterson (n.d.) stated emphatically, the propor- tions and configuration of the vestigial characters are like those in Symbos cavifrons. The Black Rocks specimen is smaller dian average (a condition exaggerated by its abraded condition), but it is still clearly within the range of variation for cranial characters of S. cavifrons. At least two other crania of S. cavifrons are now known from the Colorado Plateau (Mc- Donald, 1985a; McDonald, Neusius, and Clay, 1987), so— aldiough the Black Rocks specimen is a boundary record—it was found near die odierwise documented range of the species. We conclude that the Black Rocks specimen is simply an extensively abraded cranium representing Symbos cavifrons. Gidleya zuniensis should, therefore, be considered a junior synonym of Symbos cavifrons. The Relationship between Bootherium and Symbos The uncertainty about the biological relationship between Bootherium and Symbos is a direct result of different 28 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY interpretations of (1) morphological and preservational differ- ences exhibited in die known skull characters and (2) sample sizes of representatives of die two groups. Most of die debate about die relationship of diese two forms has involved a few specific morphological features of die cranium, the horn cores, and the lacrimal bones—all features that were known as early as Leidy's technical description of Bootherium cavifrons and Bootherium bombifrons (Leidy, 1852b). Specifically, die morphological characteristics diat have been invoked to support opinions about die relationship between Bootherium and Symbos include (1) die general size of adult crania (Osgood, 1905a; Allen, 1913); (2) die size, shape, attachment, direction, and angle of emanation of die horn cores (Osgood, 1905a; Allen, 1913; Hay, 1914); (3) die configuration of die dorsal surface of the cranium (Leidy, 1852b; Osgood, 1905a; Allen, 1913; Hay, 1914, 1915); (4) the degree of flexion between die venu-al surfaces of the basisphenoid and basioc- cipital bones at dieir junction, and die configuration of die venu-al surface of die basisphenoid (Osgood, 1905a; Allen, 1913); and (5) the relative depdi of die lacrimal fossae (Allen, 1913). In addition, the pronounced difference in die number of recovered Bootherium and Symbos skulls has been used as a basis for declaring die two forms taxonomically distinct (Osgood, 1905a; Hay, 1914). Presentiy, diree arrangements of the relationship among nominal taxa widiin Bootherium and Symbos are advocated by different writers, including (1) Bootherium (including all nominal species) and Symbos (including all nominal species) are taxonomically distinct; (2) Bootherium bombifrons is taxonomically distinct from Symbos (including all nominal taxa in Symbos. as well as B. sargenti and its synonyms); and (3) Bootherium and Symbos are sexually dimorphic forms of the same taxon (at least genus, possibly species). In essence, the morphological bases for the debate have not changed since Leidy's initial revision of die North American fossU oxen (Leidy, 1852b). Additional information on the cranial morphology of Bootherium and Symbos diat has been produced since Leidy's time has served, instead, to document better the ranges of variation found within any given character rather than to identify new differences. Likewise, the larger number of specimens now known has allowed a better documentation of the disfribution of the two forms widiout altering significandy the disproportionate numerical represen- tation of the two forms. The position taken by specific authors who have been active in the debate has been reviewed in the second section of this paper. Here, we consider the various specific issues separately, first by reviewing die argument(s) as used by the principal advocate(s) and dien discussing the validity of the argument in the light of currendy available information. DIFFERENCES IN THE SiZE OF Bootherium AND Symbos CRANIA Difference in the size of crania of Bootherium and Symbos has been interpreted as representing both sexual dimorphism and a taxonomic characteristic. Although Leidy placed die two forms in different species, he did not mention that size difference alone was a specific consideration in leading him to do so. Rutimeyer (1865) and Dawkins (1872) attributed die size difference to sexual dimorphism. Dawkins (1872:29) stated that the skull of fi. bombifrons t)ears exactly the same relation to that of fl. cavifrons. as the male to the female [sic; he has these reversed] Musk Sheep. It is therefore highly probable that B. cavifrons and fl. bombifrons are the male and female of the same species. Osgood, however, concluded that die disparity in size between the two forms "was vasdy greater than in the recent genus Ovibos" (Osgood, 1905a: 182) and proceeded to separate the two forms into Bootherium and Symbos. Allen (1913) accepted Osgood's conclusion about the taxonomic distinctive- ness of Bootherium (as represented by B. bombifrons) and Symbos. but he moved die recendy named B. sargenti (Gidley, 1908) to Symbos. stating "the relationship of B. sargenti is entirely widi Symbos, and well fulfills die condition that would be expected in die female of S. cavifrons" (Allen, 1913:215). Allen did not enumerate the expected conditions, but he clearly was aware that die female skull should be smaller dian dial of die male. Hay (1914,1915) felt diat the range of sizes among specimens assigned to Symbos was adequate to accommodate both male and female individuals, and rejected Allen's placement of B. sargenti with Symbos. Hibbard and Hinds, however, returned to the position of Rutimeyer (1865) and Dawkins (1872) when they wrote "it is very likely that Bootherium is the female woodland musk ox since all specimens of Symbos based on skulls are considered diose of bulls" (Hibbard and Hinds, 1960:107). All of die positions reviewed above were based upon die qualitative differences observed in available crania, which in most cases consisted of relatively few specimens or, in die cases of Rutimeyer and Dawkins, illustrations only. Hay probably used the largest sample, which included some 25 specimens assigned to Symbos and four specimens assigned to Bootherium. None of these authors supported their position with focused discussions of patterns expected in sexually dimorphic forms of die same species. Neidier did diey quantify comparisons of Bootherium and Symbos, nor compare such patterns with those of dimorphism found in Ovibos or other closely related taxa. Sexual dimorphism of die skeleton is an expected character- istic among species of Artiodactyla, widi die skeleton of males typically being larger than that of females (Glucksmann, 1978; Nowak and Paradiso, 1983). Patterns of size distribution within known or inferred single-sex populations have been described for samples of crania representing species such as Bison latifrons. Bison antiquus. Bison bison, Ovis catclawensis. and Rupicapra rupicapra caucasica (Corner, 1977; McDonald, 1981; Koubek and Hrab^, 1983). Patterns of size disu-ibution recorded for historic and fossil populations of Ovibos moschatus. and Bootherium and Sym- bos. are presented in Figures 33 to 37. These histograms and scatterplots show diat die Bootherium and Symbos samples possess about die same quantitative relationship to each odier as do female and male Ovibos moschatus. NUMBER 66 29 W 5 200 < o O _ o ^ o I— LU ^" LU O CL — s ymb OS N H 7 14 1.5 JCl - a 't . 6 b ?(l ■\ ■ V f) . H 4 1 u - 0 H too 1 ;"i 150 Fossil fJ Oy'bos -, rj 33 X • 15 3.4 w - o • 12.20 .'0 ■ V / 'ib t 0 ■ (1 - JM" ^^ 150 175 So o thorium N 1 9 X - 126.8 30 - ■ '^ ' 6.26 20 ■ 0 - Jh 0 5 1 17 5 200 10 0 12 5 15 0 Fossil 9 Ovibos rj 2 X - 119.3 6- 4 - 2 - • 0 - • 17 5 200 ffecenf f5 Ovibos moscltalus Recent 90vlbos moschatus 3 JZ Z 5 N - 2 2 X • 134.2 (.1 - I ■ ^ ' 10,69 0 • . U' 7 . ') 6 0 - jJL N ■ 7 6- X - 117.1 4 - 2 - • • 0 - 12 5 15 0 17 5 200 FIGURE 33.- Symbos, and POSTORBITAL WIDTH OF TRONTALS (in mm) -Histogram showing distribution of measurements of postorbital width of frontals in Bootherium, fossil and Recent samples of male and female Ovibos moschatus. sy m bos O N • 1 6 z 30 - x" CT 266 1 4 4.36 C/) LU ■-20 - (V |io. . 0 - '5 J 225 J 7 5 V 3 16.67 —n '5 iTi OF CAS f all cas Fo ssli (5 0 vib OS N X ■ 1 2 261.5 ENT 030 - se2o - 1 CT ■ V 2 7 15 10.38 O 1 0 - ■ - 0 - II L I ■ Bootherium > • • • Fossil 9 Ov/bos N • ) X 1 H 0 6 - 4 ■ 2 J 0 - • 3 2 5 ,j 7 5 Q-^ 175 22 5 2 75 325 375 ca 7 necent i Ovibos moschatus LU CQ ^1 Li N X • 8 2 8 0-266 V • 3.21 N 8 X ■ 7 4 6 ■ 4 ■ • • 2 - • • • • • • 100 12 5 150 JO 75 100 125 150 OCCIPITAL HEIGHT - OPISTHION-NUGHAL CREST (in mm) FIGURE 35.—Histogram showing distribution of measurements of occipital height in Bootherium, Symbos, and fossil and Recent samples of male and female Ovibos moschatus. Symbos Bootherium I 0 80 'HI Fossil rJ Ovibos Fossil 9 Ovibos • • • • • • • 0 •• • • . O 6 0 7 0 8(1 Recent 60vlbos moschatus 'O t.O /[I HO 'Ul Recent 9 Ovibos moschatus LEAST FRONTAL BREADTH: ORBITAL BREADTH X100 FIGURE 36.—Histogram showing dislribulion of ratio of least frontal breadlh:orhilal breadth x 100 in Bootherium, Symbos, and fossil and Recent samples of male and female Ovibos moschatus. 'ITic greater the value, the less the orbits extend beyond the postorbital level of the frontals. NUMBER 66 31 £110 E « 90- • - Recent 9 Ovibos moschatus •'- Recent ci Ovibos moschatus O - Fossil <^ Ovibos o ^ 80- 70 o 60 16U 180 200 220 MASTOID BREADTH (in mm] ^ 1 30 -| o - Bootherium • - Symbos 120- X 1 00 90 - 80 oo o • ^ O I < O O o 140 160 180 200 220 240 MASTOID BREADTH (in mm) FIGURE 37.—Scatterplot showing sexual and taxonomic differences in the size and proportions of the occiput, as expressed by the height and breadth of the region: Ovibos moschatus (lop); Bootherium and Symbos (bottom). DIFFERENCES IN THE CHARACTERISTICS OF THE HORN CORES Osgood (1905a) stressed differences in the horn cores of Bootherium and Symbos as partial justification for placing die two forms in separate genera. Osgood noted the similarity between die horn cores of male and female Ovibos moschatus. especially in their being "excessively flattened and directed downward close to the skull" with bases approaching "each other over the top of die frontals Therefore the skull of die female has all the essenUal characters of the male but they are not as highly developed" (Osgood, 1905a: 182). Osgood used the male/female parallel in Ovibos moschatus to compare and evaluate die horn cores of Bootherium and Symbos. The horn cores of Bootherium were seen to differ from diose of Symbos in four respects: (1) they were rounded, not flattened, at die base; (2) they emanated from die cranium at a different angle; (3) their bases were ringed by a burr and not fused with exostosis; and (4) they were attached to the skull via pedicels attached to the frontals instead of merging indistinctly widi die frontal and parietals. "A much more reasonable assumption would be that Symbos cavifrons represents die male and Symbos tyrrelli the female of one species. The present objection to this is the fact that both have not been found in the same region" (Osgood, 1905a: 183). In describing die type specimen of B. sargenti. however, Gidley (1908) noted differences between die hom cores of that specimen and the B. bombifrons type, including the larger size, more angular base, and extension of the dorsal edge onto die frontals toward die median plane. The characteristics of the horn cores in the Bootherium sargenti holotype led Allen (1913) to consider it to be die female of Symbos cavifrons. Hay considered die differences between B. bombifrons and B. sargenti to be taxonomically insignificant, and felt that the latter—along widi his new species Bootherium nivicolens (possessing laterally directed horn cores and modest develop- ment of exostosis over the dorsal edge of the pedicel)—was properly placed in Bootherium (Hay, 1914,1915). Some facts are important in better understanding the role that horn core differences have had in the debate. The holotype of Bootherium bombifrons (Figures 1-4) represents an extreme among described specimens in die roundness of its horn cores at die base, the presence of a distinct burr at the base, and die placement of die horn core base upon a distinct, relatively elongated pedicel. The holotype of Bootherium sargenti (Figures 19-21) is extreme in the degree of development of exostosis over die dorsal surface of the frontals toward the median plane, in die rectangularity of the dorsal half of die base of the horn core, and the preservation of detailed features of the original bone surface. These two specimens, which still represent the extremes of many characters within the genus, were the first two crania of Bootherium to be reported and dius represent die foundation upon which conceptualization of the genus, and its first two nominal species, was based. (There has been no similar difficulty in evaluating the generic placement of nominal species within Symbos that were based upon cranial material.) Bootherium specimens collected or recognized since 1915, when Hay described Bootherium nivicolens. have shown diat the nature of horn core attachment to the frontals varies between die condition seen in the holotype of Bootherium bombifrons to that seen in Bootherium sargenti (Figures 1,19, 38^2). The development of horns and horn cores within Ovibos moschatus has been described by Lonnberg (1900), Allen (1913), and Henrichsen and Grue (1980). The general pattern of horn core development is that the horn core appears upon the frontal bone by about 5-7 months of age and differentiates into a core and pedicel in both sexes by -1.5 years. During the subsequent period of development, the direction at which the horn emanates from the frontals changes. The horn cores emanate initially in a lateral direction, but—^after about 1.5 years of age—the cores shift to a more ventral orientation, reaching their maximum ventral deflection by about 4 years of age. The adult horn cores in bodi sexes have about the same orientation. Throughout die juvenile period (to -3-4 years of age) die horn cores in both sexes remain attached to the frontals 32 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY 5 cm FIGURE 1%.—Bootherium sp. (F:AM 33195; Ester Creek, Alaska). 5 cm FIGURE 39.—Bootherium sp. (F:A.M 33199; Fairbanks Creek, Alaska). NUMBER 66 33 FIGURE 40.—Bootherium sp. (BYUG 834; Utah County, Utah). FIGURE 41.—Bootherium sp. (USNM 347315; Dare County, North Carolina). 5 cm 34 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY 5 cm FIGURE 42.—Bootherium sp. (USNM 23264; SaltvUle, Virginia). by a distinct pedicel while enlarging in length and diameter (especially in antero-posterior diameter). The horn cores of males enlarge more dian do those of females. The adult horn core is slightly more rounded in females than in males. When the horn cores have reached dieir maximum ventrad extent (reached at -3 years in females; 4 years in males), continued enlargement of the horn sheath at the base is accompanied by die deposition of secondary bone over all or part of the pedicel and onto part of die dorsal surface of the cranium (Figures 43-51). The deposition of most of die secondary bone appears to take place over a relatively brief period (1-2 years) and is much more pronounced and extensive in males dian in females. The deposition of secondary bone at the base of die horn core in females appears to be limited primarily to rough pittcd-and-ridgcd bone that Leidy (1852b) called "exostosis." This exostosis develops first over die dorsal edge of die pedicel, extending mediad, rosU'ad, and ventrad beyond die edge of die pedicel. Later, die exostosis spreads caudad onto the frontal bone. The base of die horn core, and the associated exostosis, is confined to the frontal bone in adult females. Thin laminae of bone that probably developed between the integument and die ba.se of die keratinous horn shcalh have been observed in well-prepared and cautiously handled specimens (e.g., USNM 283599; USNM 291025) (Figure 52). In males, the secondary bone consists of bodi rough pitted-and-ridged exostosis where die horn shcadi lies near or in contact widi die bones of the cranial surface, and dense bone, especially bcncadi die rosQ-ad and laterad edges of die exostosis. Secondary bone in males occurs over much of the dorsal surface of die cranium, from the base of the horn core to within a few millimeters of the median plane, and from about or near die caudal edge of die parietals rostrad over the frontals to about die level of die caudal edge of die orbits. Typically, a smooth-surfaced longitudinal groove located over die median separates the secondary bone surfaces (Figures 45, 51), but occasionally dicsc secondary surfaces coalesce over part of the length of the median groove. A small shallow depression often occurs in the median groove just rostrad of die frontoparietal suture (in female skulls, a somewhat larger shallow depression often occurs centered on die median of die parietal surface). The development of secondary bone at the base of the horn core in males enlarges the base to die extent that it spreads onto the lateral edge of die parietal bones and thus, in the adult, the base of the horn core atlachcs to bodi the frontal and parietal bones. The horn cores of Bootherium exhibit some characteristics diat parallel the patterns seen in female Ovibos moschatus, but differ in odiers. The range of differing shapes of the base of the horn cores in all known specimens of Bootherium, correlated with the extent of secondary bone deposition, is similar to die ontogenetic pattern seen in Ovibos moschatus. Bootherium FIGURE 43 (top).—Ovibos moschatus (USN.M 291025; Prince Patrick Island, N.W.T,Canada), female, in dorsal view. liGLRE 44 (bottom).—Ovibos moschatus (USNM 291028; Prince Patrick Island, N.W.T, Canada), male, in dorsal view. NUMBER 66 35 ^\ 5 cm 5cm 36 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY FIGURE 45.—Ovibos proximus (= Ovibos moschatus) Bensley, 1923 (ROM Mam 31E.6.4; East Toronto, Ontario), male, in dorsal view. specimens with exostosis developed have more rostrocaudally elongated (angular or rectangular) shapes of die bases of die horn cores than do specimens without exostosis. The horn cores become increasingly rounded distally in all specimens. The pattern of development of exostosis is also similar in Bootherium and female Ovibos moschatus. Secondary bone appears first over the dorsal surface of die pedicel, then spreads mediad, rosfrad, and ventrad until it finally obscures the pedicel and forms direcUy on the frontal surface. Exostosis in Bootherium does not extend onto die parietal surface, even in die holotype of Bootherium sargenti (GRPM 11-423-3101; Figure 19). The horn cores of a5 yrs: F(N=5) 160.2° M (N=22) 150.2° Osgood and Allen mentioned is a variable character that appears on all of the Bootherium crania for which the character can be assessed. A trace of a ridge occurs on die ventral surface of the basisphenoid of some female Ovibos moschatus. This character appears to be typical of specimens assigned to Bootherium. but we consider it of no taxonomic value. DIFFERENCES IN DEPTH OF THE LACRIMAL FOSSAE Leidy observed die presence of lacrimal fossae in both Bootherium cavifrons and Bootherium bombifrons, describing that of the former as "a deep lenticular depression, or larmier, such as exists in the Deer and Sheep" and that of the latter as "a remarkably deep fossa, or larmier, which appears to have been hemispherical, but, in the specimen, the lower portion is broken away" (Leidy, 1852b: 13,18). One characteristic of Bootherium given by Allen (1913:210) was "die presence of small but deep and sharply defined lacrymal fossae." These are die only allusions to this character in a taxonomic context of which we are aware. The specific shape and size of the lacrimal fossae are variable among individuals, but generally they are larger and better defined in specimens assigned to Symbos than those assigned to Bootherium. We consider differences in die size of die lacrimal fossae to be of no taxonomic value. DIFFERENCES IN THl NUMBER OF Bootherium AND Symbos SPECIMENS The number of known specimens referred to Symbos increased slowly but continuously during the 19th century and die early part of the 20di Cenuiry. Although De Kay (1828) had but a single cranial specimen, Leidy (1852b) had access to eight such specimens, Allen (1913:214) acknowledged knowing of "at least 11" localities from which Symbos specimens had been reported (actually only eight of these localities had produced Symbos remains, and they a total of 12 whole OT partial crania), and Hay claimed to have examined 25 specimens by 1915 (Hay, 1914, 1915). The only Bootherium cranium known to exist widi certainty for some 90 years was die holotype for Bootherium bombifrons (Wistar, 1818; Harlan, 1825; Leidy, 1852b). A second specimen was described by Rhoads in die 1890s, originally as Bison appalachicolus, then Ovibos (Bootherium?) appalachicolus (Rhoads, 1895, 1897). Gidley described a third specimen as Bootherium sargenti in 1908. Other records followed, but always fewer than Symbos. Bodi Osgood (1905a) and Hay (1914, 1915) invoked die difference in the number of specimens of Bootherium and Symbos as probable evidence that die two forms represented different taxa. Osgood, noting that Symbos specimens had been reported from 10 localities by the time of his writing, stated as one reason that Bootherium should be separated from Symbos: "In the large number of specimens, if there were any females at all it is probable that diere would be more dian one" (Osgood, 1905a: 183). Hay objected to Allen's (1913) putting Booth- erium sargenti in Symbos and defended diis position by writing "If the Grand Rapids skull is the female of Symbos cavifrons, it is very remarkable that only one female should be discovered among 25 specimens" (Hay, 1914,1915:527). Many additional crania referred to either Bootherium or Symbos have been found since 1915. The most productive locaUty has been die Fairbanks, Alaska, mining district where, from 1937 to 1960, Otto Geist collected literally diousands of vertebrate fossils for Childs Frick and, to a lesser extent, die University of Alaska (Keim, 1969). Isolated specimens or small numbers of specimens have, however, been collected at numerous other localities throughout much of the United States and western Canada. The proportion of Symbos to Bootherium records, however, has not changed substantially. We have examined 175 Symbos and 51 Bootherium crania for use in diis paper. Our Bootherium.Symbos ratio of .29 is not gready different from Allen's (3:12; .25) ratio. Among bovids the skull of females typically is smaller, lighter, and less strongly fused by suturing than is die skull of males. These differences in size and structure render the female skull more susceptible dian that of die male to destruction by weadiering, abrasion, decomposition, carnivory, gnawing, or trampling. Three female and 38 male records of fossilized crania of Ovibos moschatus are known from North America, giving a female: male sex ratio of .08 for a group whose sexual dimorphism is well documented and widely accepted (Mc- Donald, unpubl. data). Howard Hutchison (pers. comm.) observed in the Canadian arctic diat die skidls of female Ovibos moschatus were more easily and readily destroyed by wolves than were die skulls of males. If female skulls are the more easUy destroyed, then it is reasonable to expect them to be recovered less often dian male skulls. There also appears to be a general collecting bias against salvaging die smaller, possibly NUMBER 66 47 Symbos rj 74 1 4 7 H • • • • • • • • • • • ••••••• ••••••• •••••• •••••••••••••• • • • a ■ V 7.56 'j t ?Boo thorium n - a - 11 1 6 .3 . b .5.34 • • • • •• • :..< 3. ? 7CO < O CO LU CO < o Fossil f5 Ovibos N 2 B 14 5.1 . 0' 5.4 2 • • • • • • • V 3.74■'"oss// 9 Ovibos N 2 . X 156 5 • • r I 1 T Recent d Oif/bos mosc/iafos N 2 1 150.9 • • o ■ V 5 9 3.91 • • • • • • • • 1 :'0' ' io- 140" I bo ii'l Recent 9 Ovibos moschatus 1 7U ■ N - 7 ; - 16 1.1 • • ••• > ANGLE OF BASIOCCIPITAL-BASISPHENOID FLEXION FIGURE 62.—Histograms showing the distribution of measurements of the angle of divergence between the basioccipital and basisphenoid bones in Bootherium, Symbos, and male and female samples of fossil and Recent Ovibos moschatus. more fragmented female skull. This has been observed among some collectors of Bison bison crania in Texas, and museum collections of some modern taxa often are biased against females. The National Museum of Natural History collections of Bison bison. Budorcas taxicolor. and Ovibos moschatus, for example, are biased in favor of males (11$ : 27cr males, adult sample only; 59 : 1 Icf; 119 : 23cf, respectively) (McDonald, 1981, unpubl. data). Sex ratios for other samples of Recent and fossil bovids are given in Table 2. Based upon die sex ratio of samples in published reports and museum collections known to us, there are typically fewer females dian male specimens in collections of bovid crania. It is our opinion diat die ratio of Booiherium:Symbos crania is within the limits of what could be expected as a sex ratio for a randomly collected sample of fossil crania in a single taxon. Certainly, the Booih- erium:Symbos ratio presents no valid, persuasive basis for claiming diat die two forms are taxonomically distinct. SIMILARITIES IN THE MORPHOLOGY AND DISTRIBUTION OF Bootherium AND Symbos SPECIMENS Even though the debate over the relationship between Bootherium and Symbos focused upon characters that differed between Bootherium and Symbos, several characters have been 48 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY TABLE 2.—Sex ratios for selected samples of bovid crania. No. of No. of F:M Sample females males ratio Source Bison bison bison 39 153 .254 McDonald. 1981 Bison antiquus occidentalis 30 115 .260 McDonald, 1981 Bison antiquus antiquus 31 107 .288 McDonald. 1981 Bison latifrons 10 80 .125 McDonald, 1981 Ovibos moschatus (Recent) 15 36 .417 McDonald, unpubl. Ovibos .noschatus (Recent) 26 31 .839 AUen. 1913 Ovibos moschatus (Fossd) 3 38 .079 McDonald, unpubl. Rupicapra rupicapra caucasica 13 30 .433 Koubek and Hrabg. 1983 Ovis catclawensis 2 4 .500 Comer. 1977identified in the literature that are similar in the two forms. Comparison of the technical descriptions of Bootherium cavifrons and Bootherium bombifrons provided by Leidy (1852b) indicates many similarities between the types of his two species, including the longitudinal growdi form of the horn cores, the lateral emanation of the horn cores and their placement about midway between the levels of the orbits and the occipital plane, the shape and surface detail of die occiput, die presence of lacrimal fossae and grooves, the shape of the temporal fossae, the foramen magnum and processes of die ventral surface of the cranium, and the configuration and relative size of the occipital condyles. Allen (1913) conceded that several important characters were similar in Bootherium and Symbos. including die occipital condyles, die surface of die occiput, and the general proportions (great depth and length relative to breadth) of the skull. Additional characters can be added to the list of similarities shared by Bootherium and Symbos. The frontoparietotemporal suture in both Symbos and Bootherium typically exhibits the same rostrocaudal orientation (i.e., roughly horizontal). The crania of male and female Ovibos moschatus share a common orientation of this suture line, but the orientation differs between Ovibos moschatus and Bootherium/Symbos. The mandibular dentition of Bootherium is now known from a mummified carcass collected by Otto Geist in 1940 at Fairbanks Creek, Alaska (McDonald, 1984b). Although this individual was a subadult (-2.3 years of age) at death and all permanent teeth had not developed, those diat are present are indistinguishable from die teeth of Symbos. In addition to morphological similarities between die crania of Bootherium and Symbos. the two forms share a similar pattern of spatial and temporal disu-ibution. The spatial distribution of Bootherium and Symbos crania, based almost entirely upon specimens we have examined personally or from photographs, are shown in Figures 63 and 64. (Provenience data are presented in Tables 3 and 4.) These illustrations include all boundary records of which we are aware. Records without figures are not included in our data set; their inclusion would change only the density, not the spatial extent, of records. The major features of similarity between die distribu- tion of records of Bootherium and Symbos include (1) their concentration in Alaska and the mid-latitude belt of ~35°-45° N; (2) dieir soudiernmost extent on die coastal plain of Texas and Louisiana; (3) their absence in the extreme southeastern and southwestern corners of the continental United States; and (4) their absence over most of the glaciated northeastern one-third of the continent. Bootherium and Symbos are known only from die United States and Canada. Although Harington has noted diat the holotype of Ovibos recticornis. from Radotin, Czechoslovakia, closely resembles Symbos (Ryziewicz, 1933; Harington, 1977), we suspect diat diis poorly known species is referable to Praeovibos. The temporal distribution of Bootherium and Symbos is also broadly similar, but reliable stratigraphic and chronologic information exists for relatively few specimens. Available information indicates that Bootherium and Symbos were essentially Rancholabrean in age. Both Bootherium and Symbos have been found in the Cripple Creek Sump loess in central Alaska, a mass of loess (possibly rebedded) of Dlinoian age that was deposited on a down-warped or down-faulted surface of auriferous gravel beneath the present valley of Cripple Creek in the Fairbanks district.... (Pewe and Hopkins. 1967:269). Symbos remains also have been reported from deposits of Illinoian age (and assigned to die late Irvingtonian) in Nebraska and Arkansas, but both of these records are weak. Jakway (1961a,b) reported Symbos in the Mullen local fauna, which he assigned to die early Illinoian. Jakway regarded this fauna as mixed, including some elements that predated the Illinoian and, possibly, some that post-dated the early Illinoian. Martin subsequently reported that, based upon restudy of the Mullen fauna and additional excavations at the locality (UNSM Cr-10), "the reworking was more extensive than Jakway had supposed" (Martin, 1972:174). Martin recognized at least two separate faunas within the Mullen assemblage, based upon the microtine rodent component: Mullen I, which he assigned tentatively to die early Kansan, and Mullen II, which was assigned to die early Illinoian. Martin did not discuss the status of Symbos within his revision of the fauna, but Kurt6n and Anderson (1980) report that Symbos. along with Bison and Alces. were probably inu-usives diat post-dated Mullen II. The Conard NUMBER 66 49 U^ FIGURE 63.—^The spatial distribution of cranial records of Bootherium used in this study (cf. Table 3). 50 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY FIGURE 64.—The spatial distribution of cranial records of Symbos used in this study (cf. Table 4). NUMBER 66 51 Fissure local fauna is considered to represent a single biostratigraphic unit assigned tentatively to the Irvingtonian, but—according to Kurten and Anderson (1980:26), who did not enumerate the reasons for their position—this age assesment is "in dispute." This local fauna was originally described by Brown (1908). Graham later collected and studied new material—primarily Blarina, with odier small mammals— from this site, and considers its assignment to the late Irvingtonian to be correct (Graham, 1972, and pers. comm.). The ovibovinc remains from Conard Fissure include three molars which were made the type of a new species, 5. australis. along widi some premolars and postcranial elements that were referred to S. australis by Brown (1908). Most of die teedi reported by Brown are of the size and configuration of those of Symbos, and their assignment to diat genus was reasonable, but one (die Lm3) differs in detail from die corresponding toodi in other specimens assigned to Symbos and all are widiin the size range of corresponding teedi of Euceratherium. In addition, a toodi row collected at Conard Fissure by James H. Quinn is more similar to what are presumed to be the teedi of Soergelia than to those of any other known ovibovinc taxon. The unequivocal presence of Symbos remains in bodi the pre- Rancholabrean fraction of the Mullen, and the Conard Fissure, local faunas is, therefore, not established. Most specimens of Symbos and Bootherium are known or considered to date from die late Wisconsinan (Tables 3 and 4, Appendix III). The geologically youngest record of Symbos. based upon a radiocarbon-dated fourdi lumbar vertebra from die associated skeleton (including skull) found near Scotts, Kalamazoo County, Michigan, is ll,l(X) + 400 yr B.P (M- 1402)(Semken, Miller, and Stevens, 1964). A date of 10,370 ± 160 yr B.P. (1-8582) on Bison bone has been obtained on the Lost Chicken Creek fauna, which includes Symbos. Equus bone from the same fauna, however, yielded a date of 26,760 ± 300 yr B.R (Sl-355), and no dates have been obtained direcUy on remains of Symbos from the site (Harington, 1978). The only radiocarbon dates obtained direcdy from Bootherium specimens were determined from hair (Sl-454:17,210 ± 500 yr BR) and scalp tissue (SI-455:24,140± 2200 yr B.P.) from die frozen carcass of a subadult (F:AM A-293-5268) found at Fairbanks Creek, Alaska, in 1940 (Pewe, 1975; McDonald, 1984b) and from a horn sheadi (SI-292:22,540 ± 900 yr B.P) on a cranium from the same locality (Pewe, 1975). Although not derived from the tissue of the animal itself, a radiocarbon date of 17,200 ± 600 yr B.R (W-1617) was obtained on mau-ix from the cranial cavity of the holotype of Bootherium bombifrons (from Big Bone Lick, Kentucky). The geologically youngest radiocarbon date associated widi Bootherium remains is 13,130 ± 330 yr B.R (A-2985) obtained on spruce wood from die base of a lacustrine mud at Saltville, Virginia, in which die remains of both Bootherium and Symbos have been found (McDonald, 1984a). Specimens of Symbos and Bootherium have been found together at several sites (Tables 3 and 4), including at least eight of the Fairbanks district sites (Cripple Creek and Cripple Creek Sump, Dome Creek, Engineer Creek, Ester Creek, Fairbanks Creek, Gold Hill, lower Coldstream, and upper Cleary Creek); Lost Chicken Creek, Alaska; die Provo Formation near Slate Canyon, Utah County, Utah (Stokes and Hansen, 1937), Natural Trap Cave, Wyoming (Gilbert and Martin, 1984), Big Bone Lick, Kentucky, and Saltville Valley, Virginia (Ray, Cooper and Benninghoff, 1967; McDonald and Bardett, 1983; McDonald, 1984a). These localities are widely distributed diroughout die Bootherium/Symbos range, and both forms have been recovered from other deposits that are known or considered to be contemporaneous. Bootherium and Symbos were, dierefore, contemporaneous as well as sympau^ic. Conclusions In the preceding sections we have reviewed die history of taxonomy of the autochdionous genera of musk oxen in North America, and die arguments that have been invoked during die past 135 years to justify decisions about the relationships between the nominal genera Bootherium and Symbos (includ- ing Gidleya). We have also discussed each issue individually in die light of current information and modern concepts, and presented evidence that allows the interpretation of all issues as characteristics or functions of sexual dimorphism. Specifically, we re-evaluated differences in the absolute size of crania; die size, shape, attachment, direction, and angle of emanation of die horn cores; the configuration of the dorsal surface of die cranium; the degree of flexion between the ventral surfaces of die basisphenoid and basioccipital bones at their junction; die relative depdi of the lacrimal fossae; and the absolute number of recovered skulls of Bootherium and Symbos. In addition, we have introduced new perspectives on the relationship between diese two forms—including similarities in the primary mor- phology of die cranium and horn cores in Bootherium and Symbos. and morphological parallels between Bootheriuml Symbos and Ovibos—and attempted to show that these perspectives, too, allow recognition of the two nominal genera as sexually dimorphic forms of a single genus. Symbos. dierefore, must be suppressed as a junior synonym of Bootherium. leaving Bootherium as the sole genus of musk oxen autochthonous to North America. There is presentiy no strong evidence that more than one species exists widiin Bootherium. The range of quantitative variation widiin the male and female samples is continuous, as is most qualitative variation, presenting patterns of variation harmonious widi diat expected in a representative sample population of a large mammal taxon that ranged over half of Nordi America for some 500,000 years. The caution that die collective sample of Symbos (now male Bootherium) crania might represent more dian one species (McDonald, 1985b) is still applicable, since some qualitative differences (e.g., shape of the frontoparietotemporal suture; depth of die frontoparietal sinus region; configuration of die basioccipital bone) do occur among crania of bodi male and female Bootherium. Instances of atypical characters are rare, and appear to be randomly 52 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY TABLE 3.—^Records of Bootherium mapped in Figure 63. Locality Specimen No. skuUcap and hom cores partial skullcap and partial left hom core cranium with partial hom cores Name Eschscholtz Bay, Alaska Fairbanks area, Alaska Fairbanks area. Alaska Fairbanks area, Alaska Fairbanks area, Alaska Fairbanks area, Alaska Fairbanks area, Alaska Fairbanks area. Alaska Fairbanks area, Alaska Fairbanks area, Alaska Fairbanks area, Alaska Fairbanks area, Alaska Fairbanks area. Alaska Fairbanks area. Alaska Alaska Alaska Alaska Alaska Alaska Alaska Alaska Fairbanks area, Fairbanks area, Fairbanks area, Fairbanks area. Fairbanks area. Fairbanks area, Fairbanks area. Fairbanks area, Alaska Fairbanks area, Alaska Lost Chicken Creek. Alaska Old Crow River, Yukcxi Territory Benton County, Washington Minidoka County. Idaho 7 Power County, Idaho 8 Bannock County, Idaho 9 Utah County, Utah 10 Utah County. Utah 11 Big Hom County, Wyoming No. Description USNM 2324 skuUcap with both hom cores A-284-2044 left hom core with adjacent frontal A-284-8280 partial cranium with partial hom cores A-284-8281 skullcap with hom cores A-293-5268 mummified carcass A-325-8392 partial left hom core A-459-2014 partial left hom core with adjacent frontal A-521-4247 right hom core with adjacent frontal F:AM 30499 partial cranium with base of right hom core F:AM 30500 skullcap with hom cores and supra- orbital region F:AM 30508 cranium with hom cores with both sheaths largely intact F:AIVI 33101 skullcap with partial hom cores F:AM 33102 right hom core F: AM 33103 partial skuUcap with partial right hom core F:AM 33194 skullcap with partial hom cores F:AM 33195 cranium with hom cores F:AM 33196 partial skuUcap and left hom core F:AM 33197 skullcap with partial hom cores F:AM 33198 skuUcap with partial hom cores F:AM 33199 cranium with hom cores F:AM 33207 cranium with partial hom cores; part of face F:AM 33208 skuUcap with base of left hom core F:AM 33220 skuUcap widi partial hom cores UAFV-54-158 cranium with hom cores NMC 10536 partial right hom core and adjacent frontal WSU uncataloged cranium with partial hom cores LACM 16888 cranium with partial hom cores skuUcap with hom cores UVP 083 IMNH 17124 lost; no number BYUG 834 skuUcap and hom cores KUMNH 6135 Provenience on the shores of Eschscholtz Bay. probably at Elephant Point Cripple Creek Engineer Creek Engineer Creek Fairbanks Creek; 64°58'N.147°10'W Engineer Creek Cripple Creek Cripple Creek Upper Qeary Creek creek near Fairijanks (Pewe. 1975) Cripple Creek Cripple Creek Sump Cripple Creek Sump Dome Creek Ester Creek Fairt)anks Creek Fairbanks Creek Fairbanks Creek Fairbanks Creek Fairbanks Creek Gold HUl bank opposite Fox—Coldstream near head of Lost Chicken Creek; 64°03X141°53'W bank of Old Crow River, 67°47'N. 139°57'W UmatiUa Mammoth Site, near Wal- lula Gap, on Columbia R. -1 mi upstream from UmatUla. Oregon from 12' depth, gravel pit 1 mi W of Minidoka Dam. N side of Snake R. (LACM loc. 6671); NE 'A. Sec. 1, T9S. R25E. Lake Walcott Quad. USGS 7.5' series intersection of Oregon Trail and Ele- vator Streets. American FaUs; NE V4. Sec. 29. T7S. R31E, American Falls Quad, USGS 7.5' series from 12' depth, Arimo Gravel Pit; SE 1/4, SW 'A. Sec. 19. TIOS. R37E. Arimo Quad. USGS 7.5' series near Pleasant Grove about 50' below surface, Provo City Upper Gravel Pits, near Slate Can- yon; SW V4, Sec. 8. T7S, R3E. Provo Quad, USGS 7.5' series Natural Trap Cave NUMBER 66 53 TABLE 3.—Continued. (Locality/Specimen No. repeated from left half of table for ease of reference.) Locality No. Specimen No. Stratigraphic unit Geologic age Principal published description(s) USNM 2324 2 A-284-2044 2 A-284-8280 2 A-284-8281 2 A-293-5268 2 A-325-8392 2 A-459-2014 2 A-521-4247 2 F:AM 30499 2 F:AM 30500 2 F:AM 30508 2 F:AM 33101 2 F: AM 33102 2 F:AM 33103 F:AM F:AM F:AM F:AM F:AM F:AM F:AM 33194 33195 33196 33197 33198 33199 33207 F:AM 33208 F:AM 33220 UAF V-54-158 NMC 10536 WSU uncataloged LACM 16888 radiocarbon age: 17,210 ± 500 yr (SI- 454: on hair) and 24.140 ± 2,200 yr (SI-455; on muscle) B.P. radiocarbon age: 22.540 ± 900 yr B.P. (SI-292 on hom sheath) (?)Late Wisconsinan (probably dates "shortly after 13,000 yr B.P.," Lyman and Livingston, 1983) Late Pleistocene Hay. 1915 (pL 31: fig. 1) unpubUshed unpublished unpublished Guthrie, 1972:300 (photograph); McDonald, 1984b unpublished unpublished unpubUshed unpublished unpublished unpublished unpublished unpublished unpublished unpublished unpubUshed unpublished unpublished unpublished unpublished unpublished unpublished unpubUshed unpublished Harington, 1977 (fig. 83) unpublished While. 1985 7 UVP 083 8 IMNH 17124 9 lost; no number 10 BYUG 834 11 KUMNH 6135 ?Equivalent to B layer (= Rainbow Beach Section), American Falls Formation Lake BonnevLUe gravels (?)Provo Formation ?~26,500 ± 500 yr B.R (?)Lake Bonneville age Late Wisconsinan Nelson and Madsen, 1987 (fig. 1) White, 1985 (figs. 2A. 3A) unpublished Stokes and Hansen, 1937 (fig. 1, bottom) Nelson and Madsen, 1987 (fig. 7) 54 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY TABLE 3.—^Records of Bootherium mapped in Figure 63 (continued). Locality Specimen No. Name No. Description Provenience 12 Sheridan County, Nebraska UNSM 9-418-39 left hom core Pit 4, Johndreau Quarry 1. E V2, Sec. 25.T31N.R41W 12 Sheridan County, Nebraska UNSM 10-419-39 partial cranium with base of left hom core Pit 4. Johndreau Quarry 1. E V2, Sec. 25.T31N.R41W 13 Hitchcock County. Nebraska UNSM 18-11-36 cranium with partial hom cores Trenton Gravel Pit 14 Cuming County, Nebraska UNSM 603-46 cranium with hom cores gravel pit at W edge West Point (UNSM loc. Cm-3); SW 'A. NW 'A.Sec. 34.T22N.R6E 15 (?)DougIas County. Nebraska UNSM 193-25-5-27 cranium with partial hom cores (?)near Omaha 16 Atchinson County. Missouri KUMNH Cast 345 left hom core from gravel bar ~4 mi SW of Fairfax; SE Vi. SE 'A, Sec. 36. T64N. R41W 17 Cooper County. Missouri Widel coUection skuUcap with p>artial hom cores dredged from Lamine R. (?)14 mi S of Blackwater, Sec. 22, T46N, R19W 18 Brazos County. Texas TAMC 2553' partial cranium with partial hom cores from sandbar in or along Brazos R. at Pitts Bridge 19 Bolivar County. Mississippi McKay coUection partial cranium with partial hom cores from sand bar in Mississippi R. near Rosedale 20 Muskegon County. Michigan GRPM 11-423-3101 dorsal cranium and horn cores about 2'-3' deep, Moorland Swamp. Charles McKay Farm. 3 mi NE of Moorland. NE 'A. Sec. 16,TION. R14W 21 GibscHi County, Indiana USNM 24885 skuUcap with partial hom cores; asso- ciated occiput and basicranium probably of same individual at 5'-20' depth in gravel pit "some- where" east of East Mt. Carmel 22 Boone County, Kentucky ANSP 994 cranium with hom cores and orbital region Big Bone Lick (Qark-Jefferson Col- lection) 22 Boone County, Kentucky UNSM till fragmentary skullcap and hom cores Big Bone Lick (University of Ne- braska coUection) 23 Smyth County, Virginia USNM 23264 cranium with partial hom cores Saltville Valley 23 Smyth County. Virginia USNM 392115 fragment of proximal right hom core Saltville Valley 23 Smyth County, Virginia Space collection fragment of frontal bone with base of left hom core^ SaltvUle Valley 23 Smyth County, Virginia Stephens collection fragment of frontal with base of left hom core SaltvUle Valley 24 Dare County, North Carolina USNM 347315 cranium with partial hom cores and parts of face in surf. Oregon Inlet 25 Bucks County. Pennsylvania ANSP 29 small part of right hom core and adjacent cranium from a closed limestone crevice in Durham Cave, along bank of Dela- ware R. near RiegelsviUe 'This Sfjecimen is now a jsart of the coUections of the Texas Memorial Museum. University of Texas. Austin. Texas. ^This specimen was in the Rufus Pickle coUection when it was examined by Ray in 1965. The Pickle coUection was subsequently sold to Ralph Space of Sussex.New Jersey. This specimen could not be located when the Space collection was examined by McDonald in December 1984. NUMBER 66 55 TABLE 3.—Continued. (Locality/Specimen No. repeated from left half of table for ease of reference.) LocaUty No. Specimen No. Stratigraphic unit Geologic age Principal published description(s) 12 USNM 9-418-39 unpublished 12 USNM 10-419-39 Middle Pleistocene (Yarmoulhian) unpublished 13 14 USNM 18-11-36 USNM 603-46 Middle to Late Wisconsinan (Frankforter, 1950:45) unpublished unpublished 15 16 USNM 193-25-5-27 KUMNH Cast 345 Bart»ur. 1931 (fig. 146) Neas and Parker, 1987 (fig. 1) 17 Widel coUection Holmes. 1960 (fig. 225) 18 19 TAMC 2553 McKay coUection Hesse, 1942 (pl. 18); Ray, 1966b (figs. 1.2) unpublished 20 GRPM 11-423-3101 Gidley. 1908 (pl. 59) 21 USNM 24885 unpublished 22 22 ANSP 994 USNM 1111 matrix from cranium radioca dated at 17,200 ± 600 yrs B.P 1617) (?)Late Wisconsinan rbon (W- Wistar, 1818 (pl. 9: figs. 10,11) unpublished 23 23 23 USNM 23264 USNM 392115 Space collection (?)Unit W4 (Late Wisconsinan fluvial gravels) (McDonald, 1984a)3 radiocarbon age of Unit W4: between 27.000 ±900 (A-2986) and 14,480 ± 300 (Beta-5701) yrs B.P (?)Late Wisconsinan (?)Late Wisconsinan Ray, Cooper, and Benninghoff, 1967 (pl. 65: figs. 2,3; pl. 66: figs. 1-3) unpublished unpublished 23 Stephens collection (?)Late Wisconsinan unpublished 24 USNM 347315 Ray, 1983:3 (photographs) 25 ANSP 29 Rhoads, 1895.1897; Ray, 1966a ^Based on information obtained from excavations at SaltvUle since October 1980, aU known vertebrate fossds from this locality are presumed to date from the late Wisconsinan or Holocene. No remains of extinct vertebrate taxa have been found in deposits of Holocene age, and no vertebrate fossd bearing deposits have been found that date >27.000 yr B.P (McDonald, 1984a). 56 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY TABLE 4.—Records of Symbos mapped in Figure 64. Locality Specimen No. Name No. Description Provenience 1 Anvik. Alaska USNM 2378 partial right hom core Anvik 2 Point Barrow. Alaska BM(NH) 10462 cranium with partial hom cores Port (= Point) Barrow 3 Lillian Creek District, Alaska BLM uncataloged cranium with hom cores Tucker Mine. LUUan Creek 4 Fairbanks area, AlasLa A-204-4254 cranium with hom cores and left maxiUary, left mandible, and sev- eral postcranial elements Little Eldorado Creek 4 Fairbanks area. Alaska F: AM 30501 cranium with left hom core; part of face lower Qeary Creek; found in place 90' below surface 4 Fairbanks area, Alaska F:AM 30505 cranium with hom cores 9 4 Fairbanks area, Alaska F:AM 30506 cranium with partial hom cores ? 4 Fairbanks area. Alaska F: AM 33124 skuU lacking nasals, part of premaxil- laries; stiU has hom sheaths upper Qeary Creek 4 Fairbanks area. Alaska F: AM 33125 cranium with partial hom cores Lower Coldstream Creek 4 Fairbanks area. Alaska F:AM 33126 skuUcap with partial hom cores lower Coldstream Creek 4 Fairbanks area. Alaska F:AM 33127 cranium with partial hom cores lower Coldstream Creek 4 Fairbanks area. Alaska F:AM 33128 cranium with bases of hom cores lower Coldstream Creek 4 Fairbanks area. Alaska F:AM 33129 cranium and base of right hom core lower Goldslream Creek 4 Fairbanks area, Alaska F:AM 33130 cranium with partial hom cores lower Goldslream Creek 4 Fairbanks area, Alaska F:AM 33131 cranium with partial hom cores Fairbanks Creek 4 Fairbanks area, Alaska F:AM 33132 partial cranium with partial hom Fairbanks Creek 4 Fairbanks area. Alaska F:AM 33133 cores cranium with bases of hom cores Fairbanks Creek 4 Fairbanks area. Alaska F:AM 33134 cranium with partial hom cores Fairbanks Creek 4 Fairbanks area. Alaska F:AM 33135 cranium with partial right hom core Fairbanks Creek 4 Fairbanks area. Alaska F:AM 33136 cranium with partial hom cores Fairbanks Creek 4 Fairbanks area, Alaska F:AM 33137 cranium with partial hom cores Fairbanks Creek 4 Fairbanks area. Alaska F:AM 33138 cranium with bases of hom cores Fairbanks Creek 4 Fairbanks area, Alaska F:AM 33139 left hom core and part of cranium Fairbanks Creek 4 Fairbanks area. Alaska F:AM 33140 cranium with partial hom cores Fairbanks Creek 4 Fairbanks area. Alaska RAM 33141 cranium with partial hom cores Fairbanks Creek 4 Fairbanks area. Alaska F:AM 33142 cranium with hom cores Fairbanks Creek 4 Fairbanks area, Alaska F:AM 33143 cranium with partial hom cores Engineer Creek 4 Fairbanks area, Alaska F:AM 33144 cranium with bases of hom cores Engineer Creek 4 Fairbanks area. Alaska RAM 33145 cranium Engineer Creek 4 Fairbanks area. Alaska F:AM 33146 cranium with partial left hom core Engineer Creek 4 Fairbanks area, Alaska F:AM 33147 partial cranium Engineer Creek 4 Fairbanks area, Alaska F:AM 33148 cranium with hom cores Engineer Creek 4 Fairbanks area. Alaska F:AM 33149 skuUcap with bases of hom cores Engineer Creek 4 Fairbanks area. Alaska F:AM 33150 cranium with bases of hom cores Cripple Creek 4 Fairbanks area, Alaska F:AM 33153 cranium with partial hom cores Engineer Creek 4 Fairbanks area. Alaska RAM 33154 cranium with partial hom cores Engineer Creek 4 Fairbanks area, Alaska F:AM 33155 cranium with partial hom cores Cripple Creek 4 Fairbanks area. Alaska F:AM 33156 skuUcap with partial hom cores Cripple Creek 4 Fairbanks area. Alaska F:AM 33157 cranium with left hom core Cripple Creek 4 Fairbanks area. Alaska RAM 33158 skuUcap with hom cores Cripple Creek 4 Fairbanks area, Alaska RAM 33159 cranium with hom cores; partial face Cripple Creek 4 Fairbanks area. Alaska RAM 33160 cranium with partial hom cores Cripple Creek 4 Fairbanks area, Alaska F:AM 33161 cranium Cripple Creek 4 Fairbanks area. Alaska RAM 33162 cranium with partial hom cores Cripple Creek 4 Fairbanks area. Alaska RAM 33164 cranium with right hom core Cripple Creek 4 Fairbanks area, Alaska F:AM 33167 skuUcap Cripple Creek Fairbanks area, Fairbanks area. Alaska Alaska F:AM 33168 F:AM 33169 Cripple Creek Sump Cripple Creek Sump cranium with partial hom cores cranium with bases of hom cores NUMBER 66 57 TABLE 4.—Continued. (Locality/Specimen No. repealed from left half of table for ease of reference.) Locality Specimen Principal published No. No. Stratigraphic unit Geologic age description(s) 1 USNM 2378 unpubUshed 2 BM(NH) 104 unpubUshed 3 BLM uncataloged unpubUshed 4 A-204-4254 radiocarbon age: >40.0(X) yr B.R (SI-291: on fecal peUets)' unpublished 4 RAM 30501 unpublished 4 F:AM 30505 unpublished 4 RAM 30506 unpublished 4 RAM 33124 radiocarbon age: 25.09011070 yr B.R (SI-850: on hom sheath) unpubUshed 4 F:AM 33125 unpublished 4 F:AM 33126 unpublished 4 F:AM 33127 unpublished 4 F:AM 33128 unpublished 4 F:AM 33129 unpubUshed 4 F:AM 33130 unpublished 4 F:AM 33131 unpublished 4 F:AM 33132 unpublished 4 RAM 33133 unpublished 4 F:AM 33134 unpubUshed 4 F:AM 33135 unpublished 4 F:AM 33136 unpublished 4 F:AM 33137 unpublished 4 F:AM 33138 unpublished 4 F:AM 33139 unpublished 4 F:AM 33140 unpublished 4 F:AM 33141 unpublished 4 F:AM 33142 unpublished 4 F:AM 33143 unpublished 4 F: AM 33144 unpublished 4 F:AM 33145 unpublished 4 F:AM 33146 unpublished 4 F:AM 33147 unpublished 4 F:AM 33148 unpublished 4 F:AM 33149 unpubUshed 4 F:AM 33150 unpublished 4 F:AM 33153 unpublished 4 F:AM 33154 unpublished 4 F:AM 33155 unpublished 4 F:AM 33156 unpublished 4 F:AM 33157 unpublished 4 F:AM 33158 unpublished 4 F:AM 33159 unpublished 4 F:AM 33160 unpublished 4 F:AM 33161 unpublished 4 F:AM 33162 unpublished 4 F:AM 33164 unpublished 4 RAM 33167 unpublished 4 RAM 33168 unpublished 4 RAM 33169 unpublished 'Several fecal pellets resembling those of Ovibos moschatus were found associated with this skeleton, parts of which were articulated by mummified connective tissue. The skeleton, however, does contain three Bison bones and one ?Ovibos moschatus rib, so the association between the Symbos skeleton and the fecal matter, although probable, is not established with certainty. 58 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY TABLE 4.—Records of Symbos mapped in Figure 64 (continued). Locality Specimen No. Name No. Description Provenience 4 Fairbanks area, Alaska F:AM 33170 cranium with partial hom cores Cripple Creek Sump 4 Fairbanks area, Alaska F:AM 33171 partial cranium with bases of hom Cripple Creek Sump 4 Fairbanks area. Alaska F:AM 33172 cores skuUcap and partial left hom core Ester Creek 4 Fairbanks area, Alaska F:AM 33173 partial cranium with partial hom Ester Creek 4 Fairbanks area. Alaska F:AM 33174 cores skuUcap with partial hom cores Ester Creek 4 Fairbanks area, Alaska F:AM 33175 cranium with partial hom cores Ester Creek 4 Fairbanks area. Alaska F.AM 33176 cranium with base of right hom core Litde Eldorado Creek 4 Fairbanks area. Alaska RAM 33177 cranium with partial hom cores LitUe Eldorado Creek 4 Fairbanks area. Alaska RAM 33178 cranium with partial right hom core Little Eldorado Creek 4 Fairbanks area, Alaska F:AM 33179 cranium with partial hom cores Livengood Creek 4 Fairbanks area, Alaska F:AM 33180 cranium with partial hom cores Livengood Creek 4 Fairbanks area, Alaska RAM 33181 skuUcap with partial hom cores Bear Creek 4 Fairbanks area. Alaska RAM 33182 cranium with hom cores; part of face Dawson Cut. Engineer Creek 4 Fairbanks area. Alaska F:AM 33183 cranium with partial hom cores Dawson Cut 4 Fairbanks area. Alaska F:AM 33184 cranium with bases of hom cores Dawson Cut 4 Fairbanks area, Alaska F:AM 33185 cranium with hom cores Dome Creek 4 Fairbanks area, Alaska F:AM 33186 cranium with partial hom cores Dome Creek 4 Fairbanks area, Alaska RAM 33188 cranium with hom cores (?)natt Creek 4 Fairbanks area, Alska F:AM 33189 cranium with right hom core Gold HOI 4 Fairbanks area, Alaska F:AM 33190 cranium with f)artial left hom cores Gold HOI 4 Fairbanks area. Alaska F:AM 33191 skuUcap Gold HUl 4 Fairbanks area, Alaska F:AM 33192 cranium with partial hom 30.000 yr B.P (GSC-987) probably Sangamonian Skwara and Walker, in press Khan. 1970 (pis. 23-25) Khan, 1970 (pis. 28.29) 11 SMNH Pl00.84 unpublished 11 SMNH Pl 675.1 Khan. 1970 (pis. 26,27) 12 UCMP 58219 unpubUshed 13 IMNH 2598 White. 1985 (figs. 2c, 3c) 14 USNM 13694 White. 1985 14 USNM 13695 White. 1985 15 UUVP 8540 Alpine Formation Early Lake BonneviUe age Nelson and Madsen. 1978 (figs. 2-5) 15 UUVP 8536 (?)Provo Formation (?)Laie Lake BonneviUe age Hay, 1927 (pl l:fig. 3) 16 17 YPM uncataloged BYUG 102 (?)Provo Formation or (?)post-Provo sediments (?)Late Lake BonneviUe age (Bissell. 1963)5 unpubUshed Stokes and Hansen, 1937 (fig. 1, top specimen) 18 BYUG 103 (?)BonneviUe Formation (?)Medial Lake BonnevUle age seU, 1963) (Bis- unpubUshed 18 USNM 17914 Alpine Formation Early Lake BonneviUe age (Bissell. 1963) unpubUshed ^Stokes and Hansen (1937) reported this specimen to be from gravels and sands of late BonneviUe age (= Provo Formation), but NelsOTi and Madsen (1978), foUowing BisseU (1963). consider the specimen to have come from fan sediments of post-BonneviUe age. 62 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY TABLE 4.—Records of Symbos mapped in Figure 64 (continued). Locality Specimen No. Name No. Description Provenience 19 Delta County, Colorado PM526 cranium with bases of hom cores along or near Oak Creek. Sec. 14 or 15. T13S. R96W HeUs Kitchen Quad, USGS 7.5' series 20 Montezuma County. Colorado DAP 23 partial cranium with partial hom cores, complete cervical vertebral series and most of thoracic verte- bral series from loess on Grass Mesa. NE 'A. Sec. 7. T38N. R15W. Trimble Point Quad. USGS 7.5' series 21 McKinley County. New Mexico USNM 5100 cranium with partial hom cores Black Rocks irrigation dam site. 4.5 mi E of Zuni. Zuni Quad. USGS 7.5' series 22 Custer County. Montana Payes coUection partial cranium with partial left hom core from gravel pit -12 mi E of MUes City 23 Bennett County. South Dakota Rice coUection cranium with partial hom cores and caudal part of face from fine gravel lens in loess on Don Rice farm. TuthUl; NE 'A. Sec. 8, T37N. R35W 24 Aurora/Douglas County. Soudi Dakota WHOM 18661 cranium with hom cores from gravel pit -4 mi N of Corsica 25 Cherry County. Nebraska UNSM 39000 skuU from North Prong Quarry (UNSM loc. Cr-102). N side of North Branch of Middle Loup River. -12 miNWofMullen;NW'A,Sec. 18, T25N.R34W 25 Cherry County. Nebraska UNSM 39001 skuUcap with right hom core from North Prong Quarry (UNSM loc. Cr-102). N side of North Branch of Middle Loup River. -12 mi NW of Mullen; NW 'A. Sec. 18. T25N,R34W 26 Frontier County. Nebraska UNSM 1108-48 cranium with partial hom cores Medicine Creek Dam 27 Red WLUOW County, Nebraska UNSM 1112 cranium with hom cores from gravel pit near McCook 28 Furnas County, Nebraska UNSM 4-10-95 cranium with partial hom cores found during constmction of Chi- cago, Burlington and Quincy Rail- road, 2 mi E of Cambridge 29 Franklin County, Nebraska UNSM uncataloged right hom core ? 30 Qay County, Nebraska UNSM 4-10-306 cranium with partial hom cores from sand pit on Spring Ranch 31 Cuming County, Nebraska UNSM uncataloged cranium with partial hom cores East Pit. West Point 32 Otoe County, Nebraska UNSM 20-10-04 cranium with hom cores from gravel bed -1' thick, 32' beneath surface, Eugene Munn farm 5 mi S of Nehawka and 15 mi NW of Nebraska City. NW 'A. Sec. 1, T9N.R13E 33 Gage County, Nebraska UNSM 22-11-30 cranium with partial hom cores from coarse gravel, 12' beneath sur- face, G.O. McClung farm, 4 mi E and 2 mi S of Beatrice 34 Jefferson County. Nebraska UNSM 15-11-00 partial cranium with partial hom near Endicott 35 Pottawattamie County, Iowa SUI 107 cranium with hom cores from loess 12' beneath surface, I3(y above Missouri R. at CouncU Bluffs 36 Wyandotte County, Kansas KUVP 54004 cranium with partial hom cores from N bank of Kansas R. NW 'A, SE 'A, Sec. 29 Tl IS. R24E 36 Wyandotte County, Kansas KUVP 58013 cranium with partial hom cores from channel of Kansas R. NW 'A, Sec. 5,T12S.R23E 37 EUs worth County, Kansas AMNH 12699 cranium with bases of hom cores (?)near WUson^ 38 Russell County, Kansas FHSM 11545 skuUcap and partial hom cores from a sand and gravel quarry along a tributary of Big Creek, 3 mi S and 3 mi E of Goriiam, SW 'A. THS, R15W *This specimen is a cast of the original in the Hastings Museum, Hastings, Nebraska. ''This specimen was purchased from Jacob Fowler at Wilson, Kansas, by Charles Sternberg for E.D. Cope. There is no indication that the specimen was found at or near WUson. NUMBER 66 63 TABLE 4.—Continued. (LocaUty/Specimen No. repeated from left half of table for ease of reference.) LocaUty Specimen No. No. Stratigraphic unit Geologic age Principal published description(s) 19 PM526 20 DAP 23 Mesa Verde (= Sage Plain) Loess found (probably redeposited) in sedi- ments that were probably of Holo- cene age radiocarbon age: 15,970 ± 155 yr B.P (SI-6137:onbone) McDonald, 1985a (figs. 2-6) McDonald, Neusius, and Qay, 1987 (fig. 5) 21 USNM 5100 Gidley. 1906 (figs. 1-3) 22 Payes coUection unpublished 23 Rice coUection unpublished 24 WHOM 18661 Pleistocene Pinsof, 1986 (fig. 15) 25 UNSM 39000 (?)Crete Sand and SUt from Terrace 4 fUl Dlinoian (Jakway, 1961a,b); (?) post- Dlinoian (Kurten and Anderson, 1980:32) unpublished 25 UNSM 39001 (?)Crete Sand and SUt from Terrace 4 fUl Dlinoian (Jakway, 1961a, Dlinoian (Kurten and 1980:32) b); (?)post- Anderson, Barbour, 1934 (figs. 173,174) 26 UNSM 1108-48 27 UNSM 1112 28 UNSM 4-10-95 unpublished Comer. 1977 (fig. 3B) Barbour. 1931 (fig. 140) 29 UNSM uncataloged 30 UNSM 4-10-308 31 UNSM uncataloged 32 UNSM 20-10-04 Grand Island Formation Kansan (Schultz, 1934) unpublished Barbour. 1931 (fig. 144) unpubUshed Barbour, 1931 (fig. 142) 33 UNSM 22-11-30 Barbour, 1931 (fig. 143) 34 UNSM 15-11-00 Barbour, 1931 (fig. 141) 35 SUI 107 unpublished 36 KUVP 54004 Nelson and Neas, 1980 (fig. 3A) 36 KUVP 58013 Nelson and Neas. 1980 (fig. 4A) 37 AMNH 12699 38 FHSM 11545 Late Pleistocene unpublished Nelson and Neas, 1980 (fig. 5B) 'This specimen is a cast of the original in the Hastings Museum. Hastings, Nebraska. 64 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY TABLE 4.—^Records of Symbos mapped in Figure 64 (continued). Locality Specimen No. Name No. Description Provenience WMHS 85.14 Connaway coUec- tion (#284) uncataloged Oost) FHSM 8138 ANSP 12995 SMU-SMP 69127 LSUMZ 17814 39 Comanche County, Kansas 40 (?)Muskogee County. Oklahoma 41 Kaufman County, Texas 42 West Feliciana Parish, Louisiana 43 Bolivar County. Mississippi 44 Coahoma County. Mississippi 45 New Madrid County. Missouri 46 Benton County. Missouri ANSP 12996 46 Benton County. Missouri ANSP 12997 46 Benton County, Missouri ANSP 12998 46 Benton County. Missouri ANSP 12999 46 Benton County. Missouri ANSP 13000 47 Lafayette County. Missouri USNM 8361 48 Mississippi River. Missouri or Dlinois unknown 49 Sangamon County, Dlmois Broughton coUec tion 50 USNM 7800 PM 14679 GRPM 12539 UMMP 3450 Mason County. Dlinois 51 52 Kenosha County. Wisconsin Newaygo County. Michigan Yoder coUection KPM A2146- 61:379 uncataloged 53 Washtenaw County, Michigan 54 St. Joseph County. Michigan 55 Kalamazoo County, Michigan 56 La Grange County. Indiana 57 St. Joseph County. Indiana cranium with partial hom cores cranium with hom cores cranium with bases of hom cores partial cranium and Ijase of left hom core cranium with partial hom cores partial cranium with bases of hom cores cranium with partial hom cores partial cranium partial cranium f>artial cranium partial cranium partial hom core cranium cranium with bases of hom cores partial cranium with partial right hom core cranium with hom cores skuUcap with partial left hom core cranium with left hom core and dorsal parts of face skuU skuU skuU. part of right mandible, and aU or part of 10 postcranial elements cranium with hom cores and parts of face along Salt Foric of Arkansas R.. 12 mi S of Coldwater. Either NW 'A.NW V4.T34S,R18W orNE 'A.NE 'A. T34S. R19W reportedly coUected by an Indian from a graveUy bluff on the Arican- sas R.. near Fort Gibson from shaUow fluvial deposit along E side of Litde Broshy Creek. -2.24 km WSW of City Hall. Kaufman from stream sediments along E side of Bayou Sara, near mouth of Gales Creek from gravel bar in Mississippi R. W of Scott from gravel bar in Mississippi R. at Friars Point New Madrid, probably from aUuvium on floodplain of Mississippi R.; sfjecimen was "ejected by the shock of an earth quake in 1812" (?)along Osage R.. near (?)Tackner (?)along Osage R.. near (?)Tackner (?)along Osage R.. near (?)Tackner (?)along Osage R., near (?)Tackner (?)alcMig Osage R.. near (?)Tackner from bed of recently abandoned chan- nel of Missouri R.. WeUington dredged from Mississippi R. a few mi S of Grand Tower, Dlinois from gravel pool surrounded by clay -65' beneath surface. Sangamon VaUey Sand and Gravel Company gravel pit 8.5 mi E of Springfield. SW 'A, Sec. 36. T16N. R4W. Springfield East Quad. USGS 7.5' series Manito. Sec. 22, T23N. R6W. Manito Quad, USGS 15'series near WUmot from marl pit between Croton Dam and Newaygo from bog. 4' beneath surface of deposit, WiUiam J. Schlicht farm, alxjut 0.5 mi NW of Manchester. Near center of NW 'A. NW 'A. Sec. 1,T4S,R3E recovered by dragline from beneath Marl Lake, on Ray Yoder property, 3 mi SE of White Pigeon. Sec. 17. T8S.R11W from marl deposit 0.5 mi S of Scotts. Sec. 25,T3S.R9W from peat bog on Lester Egli farm, 2.5 mi N of WolcottvUle. NE 'A. NE 'A. SE 'A, Sec. 21. T36N, RIOE. WolcottvUle Quad. USGS 7.5' series from 0.6 mi W of North Liberty NUMBER 66 65 TABLE 4.—Continued. (Locality/Specimen No. repeated from left half of table for ease of referraice.) LocaUty No. Specimen No. Stratigraphic unit Geologic age Principal published description(s) 39 FHSM 8138 L^te Pleistocene NelsOT and Neas. 1980 (fig. 5A) 40 ANSP 12995 Leidy. 1852b (pl. 3. pL 4: fig. 1) 41 SMU-SMP 69127 Wisconsinan. probably dating to be- tween 24,000 and 21.000 yrs B.P., perhaps to as much as 75,0(X) yr McDonald. 1985b (figs. 2-5) 42 43 LSUMZ 17814 WMHS 85.14 probably redeposited from sediments of medial to late Wisconsinan age (-38,000-10.000 yrs B.P.) McDonald and Corkum. 1987 (fig. 2) unpublished 44 45 Connaway CoUec- tion (#284) uncataloged (lost) unpubUshed De Kay. 1828 (pl 6: figs. 1-3) 46 ANSP 129% 46 ANSP 12997 46 ANSP 12998 46 ANSP 12999 46 ANSP 13000 47 USNM 8361 48 unknown 49 Broughton coUec tionunpublished unpublished unpublished unpublished unpublished unpublished Galbreadi. 1974 (fig. 1) Ray, WUls.and Palmquist. 1%8 50 USNM 7800 51 PM 14679 52 GRPM 12539 53 UMMP 3450Ray. WUls. and Palmquist, 1968 unpublished Frankforter. 1966 (pl. 2) Case. 1915 (pis. 1.2) 54 Yoder coUection Semken, MUler. and Stevens. 1964 (pl. 129: fig. 2) 55 KPM A2146- 61:379 56 uncataloged radiocarbon age: 11,100 ± 400 yr B.P. (M-1402: on bone) Semken. MUler. and Stevens. 1964 (pL 129: fig. 3: pl 130) Rarick and Wayne. 1969 (photo- graphs, pp. 10,11) 57 ? Lyon. 1926 (fig. 1) 66 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY TABLE 4.—Records of Symbos mapped in Figure 64 (continued). Locality Specimen No. Name No. Description Provenience 58 Kosciusko County. Indiana CMNH 11744 partial skuUcap with partial hom cores found in ditch being dug in swamp near North Webster 59 Porter County. Indiana AMNH 14365 skuU from -7' beneath surface in mixture of sand and clay. -6 mi E of Hebron 60 Miami County, Indiana MCIIS 170.56 skuU from muck on Homer Fenters farm.S of Macy. AUen Township. S '/a. Sec. 24 or NE 'A. Sec. 25. T29N. R3E. Macy Quad, USGS 7.5' se- 60 Miami County. Indiana USNM 8574 skuU nes Union Township 61 Montgomery County. Indiana USNM 14428 skuU from fluvial sediments along MUl Creek, on Ingersol farm. Ripley Township 62 Bartholomew County, Indiana AMNH 13830 cranium washed from aUuvium overlying 10'-20' of glacial gravel along East Fork of White River, near Wales- boro 63 (?)Hamilton County. Ohio UCM 37882 cranium with left hom core from "local gravels" in or near Cin- cinnati 64 Hamilton County. Ohio CMP55 cranium with partial hom cores from -10' beneath surface in outwash gravels. SE 'A. NE 'A. SE 'A, Sec. 10. TIN. RIE, Hooven Quad, USGS 7.5' series 65 Warren County. Ohio CMP54 cranium with partial hom cores from -14' beneath surface in gravels, near Litde Miami River. S of Foster 66 Bracken County, Kentucky USNM 23548 cranium with hom cores from Ohio R. below Augusta 67 Nicholas County. Kentucky YPM 13133 cranium with bases of hom cores Blue Licks region 68 Bocme County. Kentucky Be-Cr uncataloged cranium with right hom core Big Bone Lick 68 Boone County. Kentucky MCZ 9117 cranium with [partial hom cores Big Bone Lick (Shaler coUection) 69 Spencer County, Kentucky USNM 1988 partial cranium with partial hom along Salt River 70 Smyth County, Virginia Space collection (#3144) cores cranium with partial hom cores SaltvUle Valley 70 Smyth County, Virginia Space collection (#3145) cranium with partial hom cores SaltvUle Valley 70 Smyth County, Virginia USNM 23577 cranium with partial hom cores SaltvUle Valley 71 Continental Shelf off New Jersey PU 16340 pjartial left hom core and frontal Continental Shelf. 40 mi SE of Atlan- tic Citydistributed in time and space; diey do not cluster in a biologically or taxonomically meaningful way. At present, dien, we consider the autochdionous North American musk oxen to be monotypic, consisting only of Bootherium bom- bifrons. Bootherium bombifrons is die largest known Pleistocene musk ox of North America—heavier, longer limbed, and proportionately shorter (craniocaudally) than Ovibos mos- chatus and, probably, Praeovibos spp. The degree of sexual dimorphism appears to have been relatively great within this species, but a more nearly definitive assessment of this condition must await die comparative study of postcranial specimens representing both sexes. The skull of fi. bombifrons is relatively longer and deeper than that of Ovibos; die dorsal half of the cranium is narrow relative to the ventral half; and die orbits protrude little relative to those of Ovibos and Praeovibos. The horn cores of males are longer and deflected downward more dian are diose of females. The bases of the horn sheaths extended over the dorsal surface of die cranium and fused at the midline in males, but probably extended only slightiy over the dorsal surface of the cranium, if at all, in females. Technical descriptions of the cranium of males (given as Bootherium cavifrons or Symbos cavifrons) have been presented by Leidy (1852b) and Semken, Miller, and Stevens (1964); die female skull (as Bootherium bombifrons) has been most dioroughly described by Leidy (1852b). Technical descriptions of parts of die adult male postcranial skeleton have been provided by Semken, Miller and Stevens (1964), McDonald and Bartlett (1983), and McDonald, Neusius, and Clay (1987). NUMBER 66 67 TABLE 4.—Records of Symbos mapped in Figure 64 (continued). (Locality/Specimen No. repeated from left half of table for ease of reference.) Locality Specimen Principal published No. No. Stratigraphic unit Geologic age description (s) 58 CMNH 11744 unpublished 59 AMNH 14365 Hay. 1912 (figs. 49.50) 60 MCHS 170.56 Richards and Wepler. 1985 (fig. 1) 60 USNM 8574 unpublished 61 USNM 14428 Lyon and HaU. 1937 (pis. 1,2) 62 AMNH 13830 unpublished 63 UCM 37882 McDonald and Davis, in press (fig. 2) 64 CMP55 Wisconsinan McDonald and Davis, in press (fig. 3) 65 CMP54 Wisconsinan Anonymous, 1965:31 (photo); Mc- Donald and Davis, in press (fig. 4) 66 USNM 23548 unpublished 67 YPM 13133 unpublished 68 Be-Cr uncataloged (?)Late Wisconsinan unpublished 68 MCZ 9117 (?)Late Wisconsinan unpublished 69 USNM 1988 unpublished 70 Space collection (#3144) unpublished 70 Space coUection (#3145) unpublished 70 USNM 23577 unpublished 71 PU 16340 Parris, 1983 (fig. 4)Systematic Hierarchy Class MAMMALIA Linnaeus, 1758 Order ARTIODACTYLA Owen, 1848 Family BoviDAE Gray, 1821 Genus Bootherium (Harlan, 1825) Bootherium bombifrons (Harlan, 1825) "A head very different from any now known here An animal nearly allied to the bison," Wistar, 1818:379-380. Bos bombifrons Harlan, 1825:271. Ovibos bombifrons.—Leidy. 1852a:71. Ovibos cavifrons Leidy, 1852a:71. [New synonymy.] Bootherium cavifrons (Leidy).—Leidy, 1852b: 12. [New synonymy.] Bootherium bombifrons.—Leidy, 1852b:17. Ovibos priscus Rutimeyer, 1865:328. Bison appalachicolus Rhoads. 1895:248. Ovibos (Bootherium?) appalachicolus Rhoads, 1897:492. Scaphoceros tyrrelli Osgood, 1905a: 173. [New synonymy.] Scaphoceros cavifrons.—Osgood, 1905a:182. [New synonymy.] Symbos tyrelli.—Osgood, 1905b:224 [misspeUing]. [New synonymy.^ Symbos cavifrons.—Osgood, 1905b:224. [New synonymy.] Symbos tyrrelli.—Gidley, 1906:682. [New synonymy.] Liops zuniensis Gidley, 1906:166. Gidleya zuniensis.—Cossmann, 1907:64. [New synonymy.] Bootherium sargenti Gidley, 1908:683. Lissops zuniensis (Gidley. 1908:684). [New synonymy.] Symbos australis Brown, 1908:203. Bootherium nivicolens Hay. 1915:523. Symbols promptus? Hay, 1920:109 [misspeUing]. [New synonymy.] Symbos promptus Hay, 1920:115. [New synonymy.] Symbos convexifrons Barbour, 1934:295. ?Ovibos giganteus Frick, 1937:556. [New synonymy.] 68 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY ?Ovibos zuniensis.—Frick. 1937:557. [New synonymy.] Bootherium brazosis Hesse. 1942:78. Bootherium appalachicolus.—^Ray, 1966:4. Symbos giganteus.—McDonald and Bartlett, 1983:454. [New synonymy.] ?Ovibos australis.—McDonald and Bartlett, 1983:458. ORIGINAL DESCRIPTION.—C. Wistar provided an informal description of what later became the holotype for Bootherium bombifrons in 1818 in die paper "An Account of Two Heads Found in the Morass, Called the Big Bone Lick, and Presented to the Society, by Mr. Jefferson" (Wistar, 1818). The scientific name Bos (= Bootherium) bombifrons was created by Richard Harlan in 1825 in Fauna Americana (Harlan, 1825:271). TYPE SPECIMEN AND REPOSITORY.—Bootherium bom- bifrons is monotypic. The holotype is ANSP 994, a cranium with the greater parts of both horn cores. This sjjecimen is located in die Department of Geology, Academy of Natural Sciences, Philadelphia, Pennsylvania. TYPE LOCALITY.—Big Bone Lick, Boone County, Ken- tucky. DISTRIBUTION.—Bootherium bombifrons was the most widespread of die musk oxen to inhabit Nordi America during the Quaternary; the odier taxa were Praeovibos spp. and Ovibos moschatus. Of these taxa, only Bootherium bombifrons was autochthonous to North America. Bootherium bombifrons is known from about the Irving- tonian-Rancholabrean boundary (or, from -500,000 yr B.R) to about 10,000 yr B.R Remains of diis species have been recovered from throughout Anglo-America except the north- eastern quarter, and the extreme southwestern and soudieastern corners, of die region. DIAGNOSIS,—Horn cores subcircular to semielliptical in cross section at the base—ventral surface semicircular in rostrocaudal cross section, dorsal surface semicircular through flat to concave in rostrocaudal cross section. Horn cores emanate from lateral edge of dorsal surface of cranium, approximately midway between occipital and orbits. Horn cores emanate in lateral direction, then trend downward, forward, and outward becoming more nearly round distally. Tips terminate rostral to level of bases of horn cores, near or at level of orbits. Diameter of base of horn core along rostrocaudal axis ranges from 61 mm to 146 mm (N = 121). Length of horn core, base (burr line) to tip along dorsal surface, ranges from 210 mm to >400 mm (N = 16). Dorsal surface of cranium flexed; point of flexion near parietofrontal juncture. Orbits situated below level of dorsal surface of frontals, and extend but littie beyond postorbital margin of frontals. Mean postorbital width:orbital width ratio is .64 (N = 33). Line of frontoparietotemporal sutures is typically relatively straight and oriented horizontally. Nasofrontal suture line shaped like flat-bottomed V. Occipital surface about as deep as wide; silhouette of occipital surface is bell-shaped. (In ventral view, basioccipital typically is in shape of pentagonal shield widi caudal half of lateral edges lying nearly parallel to each other and die midline, and the rostral half of lateral edges converging rostrally toward die midline. A ridge of bone oriented rostrocaudally typically occurs over part of the midline. The shape and surface configuration of die basioccipi- tal bone is often considered to be diagnostic, but specimens of Bootherium bombifrons are known in which die basioccipital is indistinguishable from die same element in typical specimens of Ovibos moschatus. Conversely, specimens of Ovibos moschatus are known that possess basioccipital bones indistin- guishable from die same element in specimens of Bootherium bombifrons.) Superior molars typically are without (1) cementum, (2) accessory style or rib external to enamel wall between protocone-hypocone, and (3) accessory internal fossette within dentine field at juncture of protocone-hypocone. (Exceptions to 1 and 2 are rare, and to 3 are uncommon.) DIFFERENTIAL DIAGNOSIS OF MALES.—Horn cores semiel- liptical in cross section at base—dorsal surface flat to concave in rostrocaudal cross section. Bases of horn cores fused to both frontal and parietal bones, and burr line typically blurred by deposits of secondary bone. Diameter of base of horn core along rostrocaudal axis ranges from 86 mm to 146 mm (N = 94; X = 109 mm). Length of horn core, base (burr line) to tip along dorsal surface, ranges from 206 mm to >400 mm (N = 16; x = 266 mm—diis mean is based on a range of 206-369 mm; die longest horn cores known are incomplete, with remaining sections being about 400 mm in length). Dorsal surface of cranium typically covered with deposits of secondary bone (both dense and exostotic bone) from about level of orbits to about level of occiput. DIFFERENTIAL DIAGNOSIS OF FEMALES.—Horn cores sub- rounded to semielliptical in cross section at base—dorsal surface semicircular to flat in rostrocaudal cross section at base. Bases of horn cores fused with frontals only, and preserve distinct burr line. Diameter of horn core at base along rostrocaudal axis ranges from 61 mm to 96 mm (N = 27; x = 74 mm). Length of horn core, base (burr line) to tip along dorsal surface, ranges from 210 mm to 356 mm (N = 5; x = 263 mm). Dorsal surface of cranium with little or no deposits of secondary bone. If present, secondary bone is more extensive at level of rostral edge than at medial or caudal edge of base of horn cores. DIFFERENTIAL DIAGNOSIS OF OTHER NORTH AMERICAN MUSK OXEN.—Horn cores of Ovibos moschatus compressed dorsoventrally; elliptical to semielliptical in cross section at base along rostrocaudal axis; emanate in ventro-lateral direc- tion; and are straight. Tip of horn core terminates below, or just rostral to, level of base of horn core. Diameter of base of horn core along rostrocaudal axis ranges from 62 mm to 204 mm (N = 41). Lengdi of horn core, base (burr line) to tip along dorsal surface, ranges from 180 mm to 311 mm (N = 16). Dorsal surface of cranium not markedly flexed between occiput and level of orbits; secondary bone typically does not coalesce over midline. Tubular orbits extend relatively far lateral to postorbi- tal edges of frontal bones. Mean postorbital width:orbital width ratio is .54 (N = 34). Nasofrontal suture line shaped like the letter W. Silhouette of occipital surface is semielliptical; NUMBER 66 69 transverse width typically is greater than dorsoventral depdi. Superior molars typically (1) lack cementum, (2) contain accessory rib between protocone-hypocone, and (3) contain circular accessory enamel eyelet within dentine field near juncture of protocone-hypocone. Horn cores of Praeovibos spp. are triangular to subelliptical (with long axis oriented dorsocaudally-ventrorostrally) in cross section at the base. Base of horn core is fused widi dorsolateral angle of cranium, and rises above dorsal surface of cranium. Horn cores emanate in a ventrolateral and rostral direction; straight or curved slightiy rosti-ad; tips are below, or rostral to, level of base of horn core. Littie or no secondary bone is deposited on dorsal surface of cranium medial to base of horn core. Frontoparietal sinuses are shallow. Dorsal surface of cranium is not markedly flexed between occiput and level of orbits. Orbits are markedly tubular and extend far lateral to postorbital edges of frontals. The postorbital widdi:orbital width ratio in die only specimen for which measurements are available is .45. Appendix I Abbreviations for Institutional, Departmental, and Private Collections Used in This Study Institutional Collections A Frick Collection (field numbers). Depart- ment of Vertebrate Paleontology, Ameri- can Museum of Natural History, New York, New York AlVINH American Museum of Natural History, New York, New York ANSP Academy of Natural Sciences of Philadel- phia, Philadelphia, Pennsylvania BC Behringer-Crawford Memorial Museum, Covington, Kentucky BCPM British Columbia Provincial Museum, Victo- ria, British Columbia BLM Fairbanks District, Bureau of Land Manage- ment, Fairbanks, Alaska BM(NH) British Museum (Natural History), London, England BYUG Department of Geology Museum, Brigham Young University, Provo, Utah CM Cincinnati Museum of Natural History, Cin- cinnati, Ohio CMNH Carnegie Museum of Natural History, Pitts- burgh, Pennsylvania DAP Anasazi Heritage Museum, Cortez, Colorado F:AM Frick Collection (cataloged specimens). Department of Vertebrate Paleontology, American Museum of Natural History, New York, New York FHSM Sternberg Memorial Museum, Fort Hays State University, Hays, Kansas GMUS Geology Museum, University of Saskatche- wan, Saskatoon, Saskatchewan GRPM Grand Rapids Public Museum, Grand Rap- ids, Michigan HM Haslar Museum, England IMNH Idaho Museum of Natural History, Idaho State University, Pocatello, Idaho KPM Kalamazoo Public Museum, Kalamazoo, Michigan KUMNH Museum of Natural History, University of Kansas, Lawrence, Kansas LACM Natural History Museum of Los Angeles County, Los Angeles, California LSUMZ Museum of Zoology, Louisiana State Uni- versity, Baton Rouge, Louisiana MCHS Miami County (Indiana) Historical Society Museum, Courdiouse, Peru, Indiana MCZ Museum of Comparative Zoology, Harvard University, Cambridge, Massachusetts NMC National Museum of Natural Sciences, Ot- tawa, Ontario PM Field Museum of Natural History, Chicago, Illinois PU Princeton University Museum of Natural History, Princeton University, Princeton, New Jersey (vertebrate paleontology col- lections now at Peabody Museum of Natural History, Yale University, New Haven, Connecticut) ROM Royal Ontario Museum, Toronto, Ontario SMNH Museum of Natural History, Regina, Saskatchewan SMU-SMP Shuler Museum of Paleontology, Depart- ment of Geology, Southern Methodist University, Dallas, Texas SUI Museum of Natural History, University of Iowa, Iowa City, Iowa TAMC Texas A&M College (now University) col- lections (vertebrate paleontology collec- tions now at Texas Memorial Museum, University of Texas, Austin, Texas) UAI University of Alberta, Edmonton, Alberta UAF Otto Geist Museum, University of Alaska, Fairbanks, Alaska UCM Department of Geology Museum, University of Cincinnati, Cincinnati, Ohio UCMP Museum of Paleontology, University of California, Berkeley, California UMMP Museum of Paleontology, University of Michigan, Ann Arbor, Michigan UNSM University of Nebraska State Museum, Uni- versity of Nebraska, Lincoln, Nebraska USNM former United States National Museum, collections deposited in die National Mu- seum of Natural History, Smithsonian Institution, Washington, D.C. UUVP Utah Museum of Natural History, University of Utah, Salt Lake City, Utah UVP Antiquities Section, Division of State His- tory, Salt Lake City, Utah 70 NUMBER 66 71 WHOM W.H. Over State Museum, Vermillion, Soutii Dakota WMHS Winterville Mounds Historic Site, Green- ville, Mississippi WSU Laboratory of Anthropology, Washington State University, Pullman, Washington YPM Peabody Museum of Natural History, Yale University, New Haven, Connecticut Private Collections (Most recent date of record in parendieses) Broughton Robert D. Broughton, Springfield, Illinois (1967) Connaway McKay Payes Rice Space Stephens Widel Yoder John Connaway, Archaeological Survey, Mississippi Department of Archives and History, Clarksdale, Mississippi (1988) McKay Collection, Archaeological Survey, Mississippi Department of Archives and History, Clarksdale, Mississippi (1988) Mitchell L. Payes, Tucson, Arizona (1974) Don Rice, Tudiill, Soudi Dakota (1985) Ralph Space, Sussex, New Jersey (1984) Garland S. Stephens, Wydieville, Virginia (1965) Phillip Widel, Blackwater, Missouri (1967) Ray Yoder, White Pigeon, Michigan (1964) Appendix II Chronology of Nominal Species Referable to Musk Oxen in the Genera Bootherium, Symbos, and Gidleya 1. Bos (= Bootherium. Ovibos) bombifrons Big Bone Lick, Kentucky ANSP 994; cranium witii partial horn cores Harlan, 1825 Stands as senior name in the group. 2. Bootherium f = Ovibos. Scaphoceros. Symbos) cavifrons Near Fort Gibson, Oklahoma ANSP 12995; cranium with partial horn cores Leidy, 1852b Referred, this paper, to B. bombifrons. 3. Ovibos maximus Eschscholtz Bay, Alaska HM 90/2; abraded axis Richardson, 1852 Stands; generic identity uncertain, but probably Ovibos. 4. Bison (= Bootherium. Ovibos) appalachicolus Durham Cave, Bucks County, Pennsylvania ANSP 29; cranial and right horn core fragment Rhoads, 1895 Referred to B. sargenti by Ray, 1966a. 5. Scaphoceros (= Symbos) tyrrelli Bonanza Creek, Yukon Territory USNM 2555; nearly complete skull Osgood, 1905a Referred, this paper, to B. bombifrons. 6. Liops (= Gidleya. Lissops. Ovibos) zuniensis Black Rocks, New Mexico USNM 5100; extensively abraded cranium with partial horn cores Gidley, 1906 Referred, this paper, to B. bombifrons. 7. Bootherium sargenti Moorland Swamp, Michigan GRPM 11-423-3101; skull cap with horn cores Gidley, 1908 Referred to S. cavifrons by Allen, 1913; returned to Bootherium by Hay, 1914; referred, this paper, to B. bombifrons. 8. Symbos (= ?Ovibos) australis Conard Fissure, Arkansas AMNH 11828; 3 molar teeth Brown, 1908 Referred to S. cavifrons by Allen, 1913. 9. Bootherium nivicolens Eschscholtz Bay, Alaska USNM 2324; skullcap with horn cores Hay, 1915 Referred to B. sargenti by Harington, 1977. 10. Symbos promptus Afton, Oklahoma USNM 9120; upper left third molar Hay, 1920 Referred, diis paper, lo B. bombifrons. 11. Symbos convexifrons Cherry County, Nebraska UNSM 39001; skullcap with right horn core Barbour, 1934 Referred to S. cavifrons by Jakway, 1961b. 12. Ovibos (= Symbos) giganteus Near Fairbanks, Alaska AMNH RAM 30498; right humerus Frick,1937 Referred, diis paper, to B. bombifrons. 13. Bootherium brazosis Brazos River, border of Brazos County, Tfexas TAMC 2553; partial cranium with partial horn cores Hesse, 1942 Referred to B. sargenti by Ray, 1966b. 72 Appendix III Radiocarbon Dates on Bootherium and Symbos Specimens Bootherium 1. a. A-293-5268 Fairbanks Creek, Alaska Hair SI-454 17,210 ± 500 yr B.R Pewe, 1975; McDonald, 1984b b. A-293-5268 Fairbanks Creek, Alaska Tissue from beneath scalp SI-455 24,140 ± 2,200 yr B.R Pewe, 1975; McDonald, 1984b 2. F:AM 30508 Creek near Fairbanks Horn sheath SI-292 22,540 ± 900 yr B.R Pewe, 1975 Symbos 1. A-204-4254 Litde Eldorado Creek, Alaska Fecal material Sl-291 >40,000 yr B.R Pewe, 1975 RAM 33124 Upper Cleary Creek, Alaska Horn sheath Sl-850 25,090 ± 1070 yr B.R Pewe, 1975 RAM A-651-3006 Dome Creek, Alaska Horn sheath SI-851 17,695 ± 445 yr B.R P6w6,1975 NMC 25892 Lost Chicken Creek, Alaska Bone 1-10649 20,500 ± 390 yr B.R Harington, 1980 DAP 23 Grass Mesa, Colorado Bone Sl-6137 15,970±155yrB.R McDonald, Neusius, and Clay, 1987 KPM A2146-61:379 Scotts, Michigan Bone M-1402 11,100 ± 400 yr B.R Semken, Miller, and Stevens, 1964 UMMP 34124 Climax, Michigan Bone M-639 13,200 ± 600 yr B.R Hibbard and Hinds, 1960 Notes: This number was given to diree vertebrae (adas, axis, and 3rd cervical) of a large musk ox that was determined to be Symbos cavifrons on die basis of die large atlas fitting the Symbos cavifrons skull (UMMP-3450) from the Schlecht farm in Washtenaw County, Michigan. The atlas was considered to be too large for Ovibos and Booth- erium (Hibbard and Hinds, 1960). 73 Literature Cited AUen, J.A. 1912. The Probable Recent Extinction of the Muskox in Alaska. Science. new series. 36(934):720-722. 1913. Ontogenetic and Other Variations in Muskoxen, with a Systematic Review of the Muskox Group, Recent and Extinct. Memoirs of the American Museum of Natural History, new series. l(4):101-226, plates 11-18.45 figures. 21 tables. Anonymous 1965. Muskox SkuU, KLUer Wasps Donated to Museum Here. Cincinnati Post and Times-Star, 5 September:31,1 figure. Barbour. E.H. 1931. The Musk-oxen of Nebraska. The Nebraska State Museum Bulletin, 1(25):211-233,15 figures. 1934. A New Ovibovinc, Symbos convexifrons. sp. nov. The Nebraska State Museum Bulletin, l(37):295-298,2 figures. Bensley. B.A. 1923. A Muskox SkuU from Iroquois Beach Deposits at Toronto: Ovibos proximus, sp. nov. University of Toronto Studies. Biological Series. 23: 11 pages, 2 plates. 1 table. BisseU, H.J. 1963. Lake BonneviUe: Geology of Soudiern Utah VaUey, Utah. United Slates Geological Survey Professional Paper, 257-B:101-130, figures 23-27, plate 5,12 tables. BlainviUe. H. de 1816. Sur plusieurs especes d'animaux mammiferes. de I'ordre des ruminans. Bulletin des Sciences, par La Societe Philomatique de Paris, 1816:73-82. Brown, B. 1908. The Conard Fissure, a Pleistocene Bone Deposit in Northem Arkansas: with Descripticms of Two New Genera and Twenty New Sjjecies of Mammals. Memoirs ofthe American Museum of Natural History, 9(4): 155-208,3 figures, plates 14-25. Case, E.C. 1915. On a Nearly Complete Skull of Symbos cavifrons Leidy from Michigan. Occasional Papers ofthe Museum of Zoology, University of Michigan, 13:3 pages, 2 plates. Chadboume. P.A. 1871. The Discovery of a SkuU of a Musk-ox in Utah. American Naturalist, 5(5):315-316. Comer, R.G. 1977. A Late Pleistocene-Holocene Vertebrate Fauna from Red WLUow County. Nebraska. Transactions of the Nebraska Academy of Sciences, 4:77-93.4 figures, 6 tables. Cossmann. M. 1907. [Proposal to Replace Liops with Gidleya.] Revue Critique de Paldozoologie, 11th year(l):64. Cregut-Bonnoure, E. 1984. The Pleistocene Ovibovinae of Westem Europe: Temporo-Spatial Expansion and Paleoecological Implications. In D.R. Klein. R.G. White, and S. KeUer, editors. Proceedings ofthe First International Muskox Symposium, Biological Papers ofthe University of Alaska, Special Report, 4:136-144,2 figures. Dawkins, W.B. 1867. Ovibos moschatus (BlainviUe). Proceedings of the Royal Society of London, 15:516-517. 1872. The British Pleistocene Mammalia. Part V: British Pleistocene Ovidae, Ovibos moschatus. Blainville. Palaeontographical Society Monograph, 25: 30 pages. 5 plates. De Giuli. C. and F. Masini 1983. A New Element of the Late VUIafranchian (Tasso Unit) Faunas of Italy: Occurrence of Ovibovini (Bovidae. Artiodactyla. Mammalia) in the Fauna of Casa Frata (Upper Valdamo, Tuscany). Bollettino della Societd Paleontologica Italiana, 22(3):271-280. 1 figure, plates 1-3,1 table. De Kay, J.E. 1828. Notes on a FossU SkuU in the Cabinet of the Lyceum, of the Genus Bos, from the Banks of the Mississippi; with Observations on the American Species of that Genus. Annals of the Lyceum of Natural History of New York, 2:280-291, plate 6. Frankforter, W.D. 1950. The Pleistocene Geology of the Middle Portion of the Elkhom River Valley. University of Nebraska Studies, new series. 5: 46 pages. 15 figures. 1 table, 1 map. 1966. Some Recent Discoveries of Late Pleistocene Fossils in Westem Michigan. Papers of the Michigan Academy of Science, Arts, and Letters, 51:209-220. plates 1-2.2 tables. Frick. C. 1937. Homed Ruminants of North America. Bulletin of the American Museum of Natural History, 69: xxviii + 669 pages, many figures and tables. Furlong, E.L. 1905. Preptoceras. a New Ungulate from the Samwel Cave, California. University of California Publications, Bulletin ofthe Department of Geology, 4(8): 163-169. plates 24-25.1 table. Galbreath.E.C. 1974. A Cranium of Symbos cavifrons (Mammalia) from the Mississippi River between Southem Hlinois and Missouri. Illirujis State Academy of Science, Transactions, 67(4):393-396.1 figure. Gidley. J.W. 1906. A New Ruminant from the Pleistocene of New Mexico. Proceedings ofthe United States National Museum, 30(1447):165—167.3 figures. 1908. Descriptions of Two New Species of Pleistocene Ruminants of the Genera Ovibos and Botherium. with Notes on the Latter Genus. Proceedings ofthe United States National Museum, 34(1627):681- 684,1 figure, plates 57-59.1 table. 1930. Hunting Fossils on the Old Oregon Trad. In Explorations and Field-Work of the Smithsonian Institution in 1929, pages 31-36. figures 24-27. Washington, D.C: Smithsonian Institution [Publica- tion 3060]. GUbert. B.M.. and L.D. Martin 1984. Late Pleistocene Fossils of Natural Trap Cave. Wyoming, and the Climatic Model of Extinction. In P.S. Martin and R.G. Klein, editors. Quaternary Extinctions, A Prehistoric Revolution, pages 138-147, 3 figures. Tucson: University of Arizona Press. Glucksmann. A. 1978. Sex Determination and Sexual Dimorphism in Mammals, ix + 179 pages, 10 figures, 8 tables. London: Wykeham Publications. Graham, R.W. 1972. Biostratigraphy and Paleoecological Significance of the Conard Fissure Local Fauna with Emphasis on the Genus Blarina. tx + 90 pages. 4 figures, 25 tables. Master's thesis. Department of Geology, University of Iowa. 74 NUMBER 66 75 Guthrie, R.D. 1972. Re-creating a Vanished World. National Geographic Magazine 141(3):294-301,10 figures. Halstead, E.C. 1968. The Cowichan Ice Tongue, Vancouver Island. Canadian Journal of Earth Sciences. 5:1409-1415,7 figures, 1 uble. Harington, CR. 1961. History, Distrilxition and Ecology of the Muskoxen. 489 pages, 26 figures, 1 table. Master of Science thesis. Department of Geography. McGUI University. 1968. A Pleistocene Muskox (Symbos) from Dease Lake, British Colum- bia. Canadian Journal of Earth Sciences, 5(5): 1161-1165,4 figures, 1 table. 1970a. A Pleistocene Muskox (Ovibos moschatus) from Gravels of Dlinoian Age near Nome, Alaska. Canadian Journal of Earth Sciences, 7:1326-1331.5 figures. 1 table. 1970b. A Postglacial Muskox (Ovibos moschatus) from Grandview, Mani- toba, and Comments on the Zoogeography of Ovibos. National Museum of Natural Sciences, Publications in Palaeontology (National Museums of Canada), 2: vi + 13 pages, 1 figure, 3 plates. 1 table. 1975. Pleistocene Muskoxen (Symbos) from Alberta and British Columbia. Canadian Journal of Earth Sciences, 12(6):903-919. 10 figures. 2 tables. 1977. Pleistocene Mammals of the Yukon Territory. Ixxi + 1060 pages, 92 figures. 106 tables. Doctoral dissertation, Department of Zoology. University of Alberta. 1978. Quaternary Vertebrate Faunas of Canada and Alaska and Their Suggested Chronological Sequence. Syllogeus (National Museums of Canada. National Museum of Natural Sciences). 15: 105 pages. 15 figures. 1 table. 1980. Pleistocene Mammals from Lost Chicken Creek, Alaska. Canadian Journal of Earth Sciences, 17(2):168-198,16 figures,23 tables. Harlan. R. 1825. Fauna Americana: Being a Description of the Mammiferous Animals Inhabiting North America, i-x, 11-318 pages. Philadel- phia: Anthony Finley. Hay, O.R 1912. The Pleistocene Period and Its Vertebrata. Indiarui Department of Geology and Natural Resources, Annual Report for 1911, 36:539- 784,77 figures, 31 plates, tables. 1914. The Pleistocene Mammals of Iowa. Iowa Geological Survey Annual Report for 1912, 23: 662 pages, 142 figures. 75 plates, tables. 1915. Contributions to the Knowledge of the Mammals of the Pleistocene of North America. Proceedings of the United States National Museum, 48(2086):515-575.5 figures, plates 30-37.53 ubles. 1920. Descriptions of Some Pleistocene Vertebrates Found in the United States. Proceedings of the United States National Museum, 58(2328):83-l46.4 figures, plates 3-11.14 tables. 1922. Report on Work Done on the Pleistocene Epoch and Its Vertebrate Fossils. Carnegie Institution of Washington, Year Book, 20(1921):445-446. 1923. The Pleistocene of North America and Its Vertebrated Animals from the States East of the Mississippi River and from die Canadian Provinces East of Longitude 95°. Carnegie Institution of Washington Publication, 322: 499 pages. 25 figures, tables, 41 maps. 1924. The Pleistocene of the Middle Region of North America and Its Vertebrated Animals. Carnegie Institution of Washington Publica- tion, 322a: 385 pages, 5 figures, tables. 29 maps. 1927. The Pleistocene of the Westem Region of North America and Its Vertebrated Animals. Carnegie Institution of Washington Publica- tion, 322b: 346 pages. 19 figures. 12 plates, tables. 21 maps. Henrichsen. P.. and H. Gme 1980. Age Criteria in the Muskox (Ovibos moschatus) from Greenland. Danish Review of Game Biology, 11(4): 18 pages. 4 figures, plates 1-2,4 ubles. Hesse, CJ. 1942. The Genus Bootherium, with a New Record of Its Occurrence. Bulletin of the Texas Archeological and Paleontological Society, 14:77-87,2 plates. Hibbard. C.W. and F.J. Hinds 1960. A Radiocarbon Date for a Woodland Musk Ox in Michigan. Papers ofthe Michigan Academy of Science, Arts, and Letters, 45:103— 111, 1 figure, plates 1-2.1 uble. Holmes. CD. 1960. Introduction to College Geology. Third printing, xxi + 429 pages. New York: MacmiUan Co. [First printed 1949.] Jakway, G.E. 1961a. The Pleistocene Faunal Assemblages of the Middle Loup River Terrace-FUls of Nebraska. Doctoral dissertation. University of Nebraska. Department of Zoology and Physiology. [Not seen.] 1961b. Symbos convexifrons an Invalid Species. Journal of Mammalogy, 42:114-115. Jer6mie, [N.] 1720. Relation du Detroit et de la Baye de Hudson. In Bernard'sRecueilde Voyages au Nord, 6(1720):l-52. [1912 edition. Manitoba: Saint- Bontface, seen.] KahUce,H.-D. 1964. Early Middle Pleistocene (Mindel/Elster) Praeovibos and Ovibos. Socielas ScierUiarum Fennica Commentationes Biologicae, 26(5): 17 pages. 8 figures. 1975. The Macro-faunas of Continenul Europe during the Middle Pleistocene: Stratigraphic Sequence and Problems of Intercorrela- tion. In K.W. Butzer and Glynn U. Isaac, editors. After the Australopithecines: Stratigraphy, Ecology, and Culture Change in the Middle Pleistocene, pages 309-374, 1 figure. The Hague and Paris: Moulon Publishers. Keim. CJ. 1969. Aghvook, White Eskimo: Otto Geist and Alaskan Archaeology, xix + 313 pages. 80 photographs. The University of Alaska Press. Khan. E. 1970. Biostratigraphy and Palaeontology of a Sangamon Deposit at Fort Qu'AppeUe, Saskatchewan. National Museum of Natural Sciences, Publications in Palaeontology (National Museums of Canada), 5: viii -I- 82 pages, 29 plates, 17 tables. Koubek, R, and V.Hrabfi 1983. Craniometrical Characteristics of Rupicapra rupicapra caucasica. Folia Zoologica, 32(3):213-223,2 figures, 2 plates. 4 tables. Kretzoi, M. 1942. Der moschusochs in Ungam. Foldtani Kozlony (Budapest). 72: 357-363. plate 25. Kurten. B. 1968. Pleistocene Mammals of Europe. 317 pages. 111 figures. 15 ubles. London: Weidenfeld and Nicolson. Kurten. B., and E. Anderson 1980. Pleistocene Mammals of North America, xvii + 442 pages, 92 figures. 13 tables. New York: Columbia University Press. Leidy, J. 1852a. [Remarks on Two Crania of Extinct Species of Ox.] Proceedings of the Academy of Natural Sciences of Philadelphia, 6(3):71. 1852b. Memoir on the Extinct Species of American Ox. Smithsonian Contributions to Knowledge, 5(3): 20 pages, 5 plates. 1854. Remarks on die Question of the Identity of Bootherium cavifrons with Ovibos moschatus or O. maximus. Proceedings ofthe Academy of Natural Sciences of Philadelphia. 7(6):209-210. 1869. The Extinct Mammalian Fauna of Dakota and Nebraska. Including an Account of Some AUied Forms from Olher Localities. Togedier with a Synopsis of the Mammalian Remains of North America. Journal ofthe Academy of Natural Sciences of Philadelphia, series 2. 7: 472 pages, 1 figure, 30 plates, ubles. [Presented to Academy meeting, 25 January 1870.] 76 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Lonnberg, E. 1900. On the Stmcture and Anatomy of the Musk-Ox (Ovibos moschatus). Proceedings of the Zoological Society of London, 1900(19 June):686-718,10 figures, 1 uble. Lydekker, R. 1885. Catalogue of the FossU Mammalia in the British Museum (Natural History). Part II: Ccmuining the Order Ungulau. Suborder Artiodac- tyla. xxii-I- 324 pages. 39 figures. 1898. Wild Oxen, Sheep, and Goats of All Lands, Living and Extinct, xiv + 318 pages. 61 figures. 27 plates. Lcmdon: R. Ward. Lyman. R.L.. and S.D. Livingston 1983. Late Quaternary Mammalian Zoogeography of Eastem Washington. Quaternary Research, 20:360-373.1 figure. 5 Ubles. Lyon. M.W, Jr. 1926. A Specimen of the Extinct Musk-ox, Symbos cavifrons (Leidy) from North Liberty, Indiana. Proceedings of the Indiana Academy of Science, 35(for 1925):321-324,1 figure, 1 uble. Lyon. M.W, Jr.. and F.T HaU 1937. SkuU of Musk-ox, Genus Symbos, from Montgomery Couniy, Indiana. American Midland Naturalist. 18(4):608-611, plates 1-2.1 table. Martin. L.D. 1972. The Microtine Rodents of the Mullen Assemblage from the Pleistocene of North Central Nebraska. Bulletin ofthe University of Nebraska State Museum. 9(5):173-182.3 figures. 1 table. McDonald. H.G.. and R.A. Davis In press. Fossil Musk-oxen of Ohio. Canadian Journal of Zoology. McDonald, J.N. 1981. North American Bison: Their Classification and Evolution. xii-(-316 pages, 38 plates, 104 figures, 75 ubles. Berkeley: University of California Press. 1984a. The Saltville, Virginia, Locality: A Summary of Research and Field Trip Guide. 45 pages, 5 figures, 1 table. CharlottesviUe: Virginia Division of Mineral Resources. 1984b. An Extinct Muskox Mummy from near Fairbanks, Alaska: A Progress Report. In D.R. Klein, R.G. White, and S. Keller, editors. Proceedings ofthe First International Muskox Symposium, Biologi- cal Papers of the University of Alaska, Special Report. 4:148-152, 2 figures, 1 Uble. 1985a. Symbos cavifrons (Artiodactyla: Bovidae) from Delu County, Colorado. Great Basin Naturalist, 45(3):455-46l,6 figures, 1 Uble. 1985b. A Record of Symbos (Artiodactyla: Bovidae) from Kaufman County, Texas. Texas Journal of Science, 37(4):311-320,5 figures, 1 uble. 1986. Classification and Zoogeography of the North American Ovibovines [Abstract]. AAG '86—Twin Cities, Abstracts, 50. Washington. D.C: Association of American Geographers. McDonald. J.N.. and C.S. Bartlett, Jr. 1983. An Associated Musk Ox Skeleton from SaltviUe, Virginia. Journal of Vertebrate Paleontology. 2(4):453-470,10 figures, 6 ubles. McDonald. J.N., and K.C. Corkum 1987. A Woodland Musk Ox. Symbos cavifrons (Artiodactyla: Bov- idae), from Bayou Sara. Louisiana. Southwestern Naturalist. 32(1): 139-143.2 figures. 1 Uble. McDonald, J.N.. S.W. Neusius. and V.L. Clay 1987. An Associated Partial Skeleton of Symbos cavifrons (Artiodaayla: Bovidae) from Montezuma County. Colorado. Journal of Paleontol- ogy, 61(4):831-843,8 figures. 5 tables. McDonald, J.N.. and C.E. Ray 1987. Bootherium bombifrons, the Autochthonous Low-Homed Musk Ox of Pleistocene North America [Abstract]. Second International Muskox Symposium, Abstracts, P22. Saskatoon: Westem CoUege of Veterinary Medicine and University of Saskatchewan. In press. Bootherium bombifrons, the Autochthonous Musk Ox of Pleistocene North America. Proceedings ofthe Second Inter national Muskox Symposium. Moigne, A.-M. 1984. The Giant Muskox of the Arago Cave, Tauuvel, France. In D.R. Klein, R.G. White, and S. KeUer, editors. Proceedings of the First International Muskox Symposium. Biological Papers ofthe Univer- sity of Alaska. Special Report. 4:145-147,2 figures. Moya-Sola. S.. J. Agusti.J. Gibert. and J. Pcms-Moya 1981. El yacimiento cuatenario de Venta Micena (Espana) y su importan- cia dentro de las asociaciones faunisticas del Pleistoceno inferior europeo. Paleontologia i Evolucio, 16:39-53.2 plates, 5 Ubles. Neas, J.F. 1986. A Systematic Review of Symbos and Bootherium (Artiodactyla. Bovidae), with Notes ertis complementum. Novi commerUarii Academiae scientiarum imperi- alis Petropolitanae, 17:576-606, plates 15-17. [Not seen.] Panis. D.C. 1983. New and Revised Records of Pleistocene Mammals of New Jersey. The Mosasaur, 1:1—21.4 figures, 1 table. Pennant. T. 1781. History of Quadrupeds. Volume I, xxiv -(- 284 pages. 32 plates. London: B. White. [Third edition. 1793, seen.] 1784. Arctic Zoology. Volume I, Introduction, Class I: Quadmpeds. [12] -I- CC -(- [6] + 186 pages. 8 plates, frontispiece. London: Henry Hughs. Peterson, O.A. 1926. The FossUs of the Frankstown Cave. Blair County, Pennsylvania. Annals ofthe Carnegie Museum. 16(2):249-314, plates 17-25. 10 figures, 3 Ubles. P6w6.TL. 1975. Quaternary Geology of Alaska. United States Geological Survey Professional Paper. 835: v -^ 145 pages, 45 figures. 1 plate, 14 ubles. Pewe. T.L.. and D.M. Hopkins 1967. Mammal Remains of Pre-Wisconsin Age in Alaska. In D.M. NUMBER 66 77 Hopkins, editor, The Bering Land Bridge, pages 266-270, 1 Uble. Stanford: Stanford University Press. Pinsof, J.D. 1986. The Pleistocene Vertebrate Fauna of South Dakou. xvii -t- 298 pages, 16 figures, 27 tables. Master's thesis, Department of Geology and Engineering. South Dakou School of Mines and Technology. Rarick, R.D., and WJ. Wayne 1969. The WolcottvUle SkuU. Outdoor Indiana. 34(l):l0-l 1.2 figures. Ray, C.E. 1966a. The Identity of Bison appalachicolus. Notulae Naturae, 384: 7 pages, 2 figures. 1966b. The Sutus of Bootherium brazosis. Texas Memorial Museum, The Pearce-Sellards Series, 5: 7 pages, 2 figures. 1983. Outerbank Ovibovinc Adds Link lo Musk Ox Research. Whale- bones, 15:3,1 figure. Ray, C.E.,B.N. Cooper, and W.S. Benninghoff 1967. Fossil Mammals and PoUen in a Late Pleistocene Deposit at SaltviUe, Virginia. Journal of Paleontology, 41(3):608-622. plates 65-66,4 figures. 6 ubles. Ray. C.E., D.L. WUls, and J.C. Pabnquist 1968. Fossil Musk Oxen of Ulinois. Illinois Stale Academy of Science Transactions, 61(3):282-292.5 figures. Rhoads, S.N. 1895. Distribution of the American Bison in Pennsylvania, with Remarks on a New FossU Sfjecies. Proceedings of the Academy of Natural Sciences of Philadelphia, 47:244—248. 1897. Notes on Living and Extinct Species of North American Bovidae. Proceedings of the Academy of Natural Sciences of Philadelphia, 49:483-502. plate 12. Rice. H.C. Jr. 1951. Jefferson's Gift of Fossils to the Museum of Natural History in Paris. Proceedings ofthe American Philosophical Society. 95(6):597- 627. 15 figures. 1 table. Richards, R.L., and W.R. Wepler 1985. Extinct Woodland Muskox. Symbos cavifrons. Cranium from Miami County. North Central Indiana. Proceedings ofthe Indiana Academy of Science. 94:667-671,1 figure, 1 Uble. Richardson, J. 1852. The Zoology ofthe Voyage of H M.S. Herald, under the Command of Captain Henry KelleU, RM., C£., during the Years 1845-51: Fossil Mammals. 140 pages. 24 plates. Ubles. London: Reeve and Co. [Complete work, including Uving vertebrates, and three additional pages on fossU mammals, pubUshed 1854.] Romer, A.S. 1966. Vertebrate Paleontology. Third edition, ix + 468 pages, 443 figures, 4 ubles. Chicago: University of Chicago Press. Riitimeyer, L 1865. Beitrage zu einer Palaeontologischen Geschichte der Wiederkauer. Zunachst an Linne's Genus Bos. Verhandlungen der Naturforschen- den Gesellschaft in Basel, 4(2):299-354. [Received at Smidisonian Instituuon 2 April 1866.] Ryziewicz. Z. 1933. Ovibos recticornis n. sp. ein Beitrag zur Systematik der UnterfamUie Ovibovinae. Bulletin de I'Academie Polonaise des Sciences et des Lettres. Classe des Sciences Mathematiques el Naturelles. Series B (Sciences NaiureUes) (2). 1-5:71-87, plate 5,5 ubles. 1955. Systematic Place of the FossU Musk-ox from the Eurasian DUuvium. Prace Wroclawskiego Towarzystwa Naukowego. series B, 49: 74 pages, 55 figures. 51 tables. Schultz. C.B. 1934. The Pleistocene Mammals of Nebraska. Bulletin of the Nebraska State Museum. l(41):357-393.2 ubles. Semken, H.A.. B.B. MiUer. and J.B. Stevens 1964. Late Wisconsin Woodland Musk Oxen in Association with PoUen and Invertebrates from Michigan. Journal of Paleontology, 38(5):823-835.2 figures.plates 129-132.4 tables. Sher. A.V. 1974. Pleistocene Mammals and Stratigraphy of the Far Northeast USSR and North America. International Geology Review. 16(7): 1-89; (8):90-121; (9):122-206; (10):207-284. i-vi. 43 figures, plates 1-30.40 ubles. Skwara. T, and E.G. Walker In press. Extinct Muskox and Other Additions to the Late Pleistocene RiddeU Fauna (RiddeU Member, Floral Formation; Late Ranchola- brean). Saskatoon. Canada. Canadian Journal of Earth Sciences. Smidi.CH. 1827. Supplement to die Order Ruminanlia. In Edward Griffith. Charles HamUton Smith, and Edward Pidgeon. The Class Mammalia Arranged by the Baron Cuvier, with Specific Descriptions, 4:33- 428.51 plates. [Reprint 1978. New York: Amo Press.] Suudinger. W 1908. Praeovibos priscus, nov. gen. et nov. sp., ein Vertreter einer Ovibos nahestehenden Gattung aus dem Pleistocan Thiiringens. Centralblatt fiir Mineralogie, Geologie und Palaontologie, 1908(16):481-502,4 figures.2 ubles. Stokes. W.L., and G.H. Hansen 1937. Two Pleistocene Musk-Oxen from Utah. Utah Academy of Sciences, Arts and Letters, 14:63-65,1 figure. Tener, J.S. 1965. Muskoxen in Canada: A Biological and Taxonomic Review. Canadian Wildlife Service Monograph, 2: 166 pages, 7 photographs. 1 figure. 72 tables. 1 map. TroxeU.E.L. 1915. A FossU Ruminant from Rock Creek, Texas, Preptoceras mayfieldi sp. nov. The American Journal of Science, 40:479-482,3 figures, 1 Uble. WaUcer.D.N. 1982. A Late Pleistocene Ovibos from Souiheastem Wyoming. Journal of Paleontology. 56(2):486-491.4 figures. 1 table. White, J.A. 1985. Late Pleistocene Musk Oxen from Southem Idaho. Tebiwa, 22:64-71,4 figures, 2 ubles. Wisur.C 1818. An Account of Two Heads Found in die Morass, CaUed the Big Bone Lick, and Presented to the Society, by Mr. Jefferson. Transactions of the American Philosophical Society, new series. 1(31):375-380, plates 10-11. TiU.S. GOVERNMENT PRINTING OFFICE: 1989-241-726/60050 REQUIREMENTS FOR SMITHSONIAN SERIES PUBLICATION Manuscripts intended for series publication receive substantive review (conducted by their originating Smithsonian museums or offices) and are submitted to the Smithsonian Institution Press with Form SI-36, which must show the approval of the appropriate authority designated by the sponsoring organizational unit. Re- quests for special treatment—use of color, foldouts, case-bound covers, etc.—require, on the same form, the added approval of the sponsoring authority. Review of manuscripts and art by the Press for requirements of series format and style, completeness and clarity of copy, and arrangement of all material, as outlined t>elow. will govern, within the judgment of the Press, acceptance or rejection of manuscripts and art. Copy must be prepared on typewriter or word processor, double-spaced, on one side of standard white bond paper (not erasable), with I'A* margins, submitted as ribbon copy (not carbon or xerox), in loose sheets (not stapled or bound), and accompanied by original art. Minimum acceptable length is 30 pages. Front matter (preceding the text) should include: title page with only title and author and no other information, abstract page with author, title, series, etc., following the established fomfiat; table of contents with indents reflecting the hierarchy of heads in the paper; also, foreword and/or preface, if appropriate. First page of text should carry the title and author at the top of the page; second page should have only the author's name and professional mailing address, to be used as an unnumbered footnote on the first page of printed text. Center heads of whatever level should be typed with initial caps of major words, with extra space above and below the head, but no other preparation (such as all caps or undertine, except for the undertine necessary for generic and specific epithets). Run-in paragraph heads should use period/dashes or colons as neces- sary. Tabulations within text (lists of data, often in parallel columns) can be typed on the text page where they occur, but they should not contain mies or numbered table captions. Formal tables (numbered, with captions, boxheads, stubs, rules) should be submitted as carefully typed, double-spaced copy separate from the text; they will be typeset unless otherwise requested. If camera-copy use is anticipated, do not draw rules on manuscript copy. Taxonomic keys in natural history papers should use the aligned-couplet form for zoology and may use the multi-level indent form for botany. If cross referencing is required between key and text, do not include page references within the key, but number the keyed-out taxa. using the same numt)ers with their corresponding heads in the text. Synonymy in zoology must use the short form (taxon, author, yearpage), with full reference at the end of the paper under "Literature Cited." For botany, the long form (taxon, author, abbreviated journal or txwk title, volume, page, year, with no reference in "Literature Cited") is optional. Text-reference system (author, yeanpage used within the text, with full citation in "Literature Cited" at the end of the text) must be used in place of bibliographic footnotes in all Contributions Series and is strongly recommended in the Studies Series: "(Jones, 1910:122)" or "...Jones (1910:122)." If bibliographic footnotes are required, use the short form (author, brief title, page) with the full citation in the bibliography. Footnotes, when few in number, whether annotative or biblio- graphic, should be typed on separate sheets and inserted imme- diately after the text pages on which the references occur. Exten- sive notes must be gathered together and placed at the end of the text in a notes section. Bibliography, depending upon use, is termed "Literature Cited," "References," or "Bibliography." Spell out titles of books, articles, journals, and monographic series. For txx)k and article titles use sentence-style capitalization according to the rules of the lan- guage employed (exception: capitalize all major words in English). For journal and series titles, capitalize the initial word and all subsequent words except articles, conjuncttons, and prepositions. Transliterate languages that use a non-Roman alphabet according to the Library of Congress system. Underline (for iteUics) titles of journals and series and titles of books that are not part of a series. Use the parentheses/colon system for volume (number): pagination: ''10(2):5-9." For alignment and arrangement of ele- ments, follow the format of recent publications in the series for which the manuscript is intended. Guidelines for preparing twbli- ography may be secured from Series Section, SI Press. Legends for illustrations must be submitted at the end of the manuscript, with as many legends typed, double-spaced, to a page as convenient. Illustrations must be submitted as original art (not copies) accompanying, but separate from, the manuscript. Guidelines for preparing art may t»e secured from Series Section, SI Press. All types of illustrations (photographs, line drawings, maps, etc.) may be intermixed throughout the printed text. They should be temned Figures and should be numbered consecutively as they will appear in the monograph. If several illustrations are treated as components of a single composite figure, they should be desig- nated by lowercase italic letters on the illustration; also, in the legend and in text references the italic letters (underiined in copy) should be used: "Figure 9b." Illustrations that are intended to follow the printed text may be termed Plates, and any components should be similarly lettered and referenced: "Plate 9b." Keys to any symbols within an illustration should -appear on the art rather than in the legend. Some points of style: Do not use periods after such abbrevi- ations as "mm, ft, USNM, NNE." Spell out numbers "one" through "nine" in expository text, but use digits in all other cases if possible. Use of the metric system of measurement is preferable; where use of the English system is unavoidable, supply metric equiva- lents In parentheses. Use the decimal system for precise meas- urements and relationships, common fractions for approximations. Use day/month/year sequence for dates: "9 April 1976." For months in tabular listings or data sections, use three-letter abbre- viations with no periods: "Jan, Mar, Jun," etc. Omit space between initials of a personal name: "J.B. Jones." Arrange and paginate sequentially every sheet of manu- script in the following order: (1) titie page, (2) abstract, (3) con- tents, (4) foreword and/or preface, (5) text. (6) appendixes. (7) notes section, (8) glossary, (9) bibliography, (10) legends, (11) tables. Index copy may be submitted at page proof stage, but plans for an index should be indicated when manuscript is sub- mitted. ■wmO-