JESSICA A. HARRISON Giant Camels from the Cenozoic of North America 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 the 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 Anthropology Smithsonian Contributions to Astrophysics Smithsonian Contributions to Botany Smithsonian Contributions to the Earth Sciences Smithsonian Contributions to the Marine 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 libraries, 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 5 7 Giant Camels from the Cenozoic of North America Jessica A. Harrison SMITHSONIAN INSTITUTION PRESS City of Washington 1985 ABSTRACT Harrison, Jessica A. Giant Camels from the Cenozoic of North America. Smithsonian Contributions to Paleobiology, number 57, 29 pages, 17 figures, 1985.—Seven genera of giant camels occurred in North America during the interval from the late Clarendonian to the early Holocene. Aepycamelus was the first camel to achieve giant size and is the only one not in the subfamily Camelinae. Blancocamelus and Camelops are in the tribe Lamini, and the remaining giant camels Megatylopus, Titanotylopus, Megacamelus, Gigantoca- melus, and Camelus are in the tribe Camelini. Megacamelus is a late Hemphil- lian giant camel most closely related to Gigantocamelus. Titanotylopus is reserved for the brachyodont form from the Irvingtonian of Nebraska, and Giganto camelus is reinstated for the broad-chinned, Blancan form. Official publication date is handstamped in a limited number of initial copies and is recorded in the Institution’sannual report, Smithsonian Year. Series cover design: The trilobite P ha cops ran a Green. Library of Congress Cataloging in Publication Data Harrison, Jessica A. Giant camels from the Cenozoic of North America. (Smithsonian contributions to paleobiology ; no. 57) Bibliography: p. Supt. of Docs, no.: SI 1.30:57 1. Camels, Fossil. 2. Paleontology—Cenozoic. 3. Paleontology—North Amer¬ ica. I. Title. II. Series. QE701.S56 no. 57 [QE882.U3] 560s [599.73'6] 84-600303 Contents Page Introduction. 1 Acknowledgments. 1 Phylogenetic Relationships. 1 Aepycamelus . 4 Megatylopus . 5 Titanotylopus . 7 Gigantocamelus . 8 Megacamelus . 10 Megacamelus merriami, new combination. 11 Camelus . 21 Blancocamelus . 23 Camelops . 24 Summary. 24 Literature Cited. 26 iii Frontispiece.—R econstruction of the head of Megacamelus merriami. Giant Camels from the Cenozoic of North America Jessica A. Harrison Introduction Throughout the later Cenozoic, camels often figure as an abundant and diverse element of any fauna in which they occur. Until the late Pleis¬ tocene, when the group fell on hard times, the Camelidae must be accounted one of the more successful ungulate families. As in many other herbivore families, the earliest members of the Camelidae were of small body size. However, a trend toward gigantism can be observed throughout the later Cenozoic, from the Clar- endonian into the Holocene. Descriptions of very large camels are almost as abundant in the literature as their remains in late Cenozoic faunas. The confusing taxonomic his¬ tory of the giant camels is such that, for every specific identification, there are many more re¬ ferrals to “camelid, large, gen. et sp. indet.” The purpose of this paper is to provide a temporal, geographic, and systematic framework for the large, late Cenozoic camels. Acknowledgments. —I am grateful for the use of specimens from the Frick Collection, De¬ partment of Vertebrate Paleontology, American Museum of Natural History (F:AM), the Univer¬ sity of California Museum of Paleontology Jessica A. Harrison, formerly Department of Paleobiology, Na¬ tional Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560, now Research Associate, Department of Geosciences, University of Arizona, Tucson, Arizona 85721. (UCMP), the University of Nebraska State Mu¬ seum (UNSM), and the University of Kansas Museum of Natural History (KUVP). I very much appreciate careful and constructive reviews by George Corner, Michael Voorhies, John Breyer, and Robert Emry. Drs. Corner and Voorhies were particularly generous in sharing with me their new information on Titanotylopus. The frontispiece was done by Robert Hynes. Phylogenetic Relationships The cladogram in Figure 1 summarizes rela¬ tionships within the Camelinae. It is interesting to note that the trend toward gigantism is far more apparent in the Camelini than in the Lam- ini. All of the genera comprising the Camelini can be called giants, but only two of the Lamini, Camelops and Blancocamelus, achieve a formida¬ ble body size. Aepycamelus, the only noncameline genus, represents the camels’ earliest experimen¬ tation with gigantism. The characters appearing at nodes 1 through 35 in the cladogram are listed below. The com¬ position and apomorphies of the Protolabidini are from Honey and Taylor (1978:419-420), whereas those of the Lamini and Camelini ap¬ peared in part in Harrison (1979:3-8). More detailed discussion of the characters may also be found in those papers. 1 2 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Node 1. Aepycamelus shares with the Camelinae a. metastylid present on the lower molars Node 2. Aepycamelus is distinguished by a. extremely elongate limbs b. extremely elongate cervical vertebrae c. metapodials longer than the basal length of the skull Node 3. The Camelinae are united by a. weak buccinator fossa b. elongate rostrum Node 4. The Protolabidini are united by a. narrow rostrum b. laterally expanded anterior nares Node 5. Tanymykter is distinguished by a. closely appressed P| roots Node 6. Protolabis and Michenia are derived relative to Tanymykter by a. the absence of elongate basioccipital tuberosities b. P' without strong, continuous lingual cingulum c. auditory bulla less inflated with medial plates more compressed d. moderate to strong buccinator fossa Node 7. Protolabis is distinguished by a. hypsodont molars b. anteroposteriorly elongate M3 c. very weak to absent metastylid on lower molars d. ventrally produced mandibular angle with weak to strong lateral flare e. fused metapodials f. elongate proximal phalanx with distal articular surface anteriorly extended Node 8. Michenia is derived relative to Protolabis in hav¬ ing a. short braincase b. weak I s -C\ small C| c. shallow symphysis d. inflection of mandibular angle suppressed Node 9. The Camelinae exclusive of the Protolabidini are united by a. metacarpal length exceeds metatarsal length b. metapodials completely fused c. I 1 absent Node 10. Procamelus is the sister taxon to the remaining camelines. It retains several primitive characters but has almost completed the loss of I 2 NUMBER 57 3 Node 11. The Lamini share with the Camelini c. P 3 absent a. I 2 absent d. dorsal surface of the mandibular condyle trans¬ b. Pi absent versely concave c. raised posterolateral edges on the proximal end of the proximal phalanx e. suspensory ligament scar extends to center of proximal phalanx and has a raised center Node 12. The Lamini are united by Node 27. The Camelini are united by a. in cross section the anterior end of the nasals form a high, bilobed arch (Harrison, 1979, fig. a. angular process on mandible enlarged and strongly inflected 3) b. large postglenoid foramen b. anteroexternal style (= llama buttress) present on lower molars c. long postglenoid process on skull with corre¬ spondingly large facet on mandibular condyle Node 13. Pliauchenia, Hemiauchenia, (?)Blancocamelus, Pa- laeolama, Lama, and Vicugna share d. Cl enlarged and rounded in cross section, espe¬ cially in males a. reduced lacrimal vacuity e. ventrally flattened auditory bulla b. shortened rostrum f. diastemal crest on mandible low and rounded Node 14. Pliauchenia is primitive in all known characters g- reduced maxillary fossa to the remaining Lamini. h. thickened, heavy premaxilla Node 15. Hemiauchenia, Q)Blancocamelus, Palaeolama, Node 28. Megatylopus and Titanotylopus share Lama, and Vicugna are united by a. reduced Pi a. small Pi b. reduced P 3 b. small or absent P 3 Node 29. Megatylopus is distinguished by Node 16. Hemiauchenia and Blancocamelus share a. reduced P 3 a. extremely elongated metapodials b. cheek teeth higher crowned than Titanotylopus Node 17. Hemiauchenia is distinguished by Node 30. Titanotylopus is derived relative to Megatylopus in a. extremely elongated cervical vertebrae having Node 18. Blancocamelus is distinguished by a. Pi absent a. great size b. P 3 more reduced than in Megatylopus Node 19. Palaeolama, Lama, and Vicugna share c. larger body size than Megatylopus a. Pi absent Node 31. Megacamelus, Gigantocamelus, and Camelus share b. reduced maxillary fossa a. metapodials shorter in relation to basal length c. moderate to strong anteroexternal style on lower of the skull molars b. cheek teeth more hypsodont than Megatylopus or Node 21. Lama and Vicugna are derived relative to Palaeo¬ Titanotylopus lama in having Node 32. Megacamelus and Gigantocamelus share a. Ps absent a. spatulate lower incisors b. metacarpal length subequal to metatarsal length b. splayed Ci c. strong anteroexternal style on lower molars Node 33. Megacamelus is primitive in all characters relative d. greatly reduced lacrimal vacuity to Gigantocamelus except for e. extremely retracted nasals a. I 3 enlarged and caniniform f. greatly reduced P 4 Node 34. Gigantocamelus is distinguished by Node 22. Lama is distinguished by a. short, blunt chin with a shortened ramal sym¬ a. callosities on the inner foreleg physis Node 23. Vicugna is distinguished by b. greater size than Megacamelus a. hypsodont lower incisors c. lower incisors arrayed almost transversely Node 24. Alforjas and Camelops are derived relative to d. I 3 absent or vestigial other lamines in having Node 35. Camelus is distinguished by a. moderately hypsodont to very hypsodont molars a. reduced paroccipital process b. cheek teeth narrow in relation to length b. metapodials subequal in length and shorter than Node 25. Alforjas is primitive in all known characters rel¬ the basal length of skull ative to Camelops c. maxillary fossa reduced or absent Node 26. Camelops is derived relative to Alforjas in having d. zygomatic arch straight in lateral view a. cheek teeth much more hypsodont e. retracted nasals b. Pj absent f. center of suspensory ligament scar raised 4 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Aepycamelus Macdonald, 1956 Type-Species. — Aepycamelus giraffinus Mac¬ donald, 1956:198 (= Alticamelus giraffinus Mat¬ thew in Matthew and Cook, 1909:402). Aepycamelus is the geologically oldest of the giant camels, ranging from the Barstovian through the early Hemphillian and occurring through the southern and western United States (Figure 2). When Marsh (1894:274) described Procamelus altus from Oregon, he based it solely upon an isolated calcaneum. Cope (1894:869) took almost instant exception to the designation of such an undiagnostic element as a type speci¬ men. Matthew (1901:429), under the impression that the type of P. altus was more extensive, described Alticamelus from northeastern Colo¬ rado and named A. altus the genotypic species. When he became aware of the indeterminate nature of the type of A. altus, Matthew renamed the Colorado material A. giraffinus (Matthew and Cook, 1909:402). Macdonald (1956:198) main¬ tained that Alticamelus was a nomen vanum and proposed Aepycamelus as a replacement name, with A. giraffinus as the new genotypic species. All of the species of Aepycamelus are noted for extremely elongate, slender limbs and cervical vertebrae. The teeth are quite brachyodont, and the dental formula is I^CiPljM 3 . When present, I 1,2 are reduced to stumps as in the type of A. giraffinus. Pi is always present and less reduced than in Procamelus. Matthew and Cook (1909:402) described Alti¬ camelus procerus from the Snake Creek beds of Nebraska, and later Matthew (1924:187) de¬ scribed a second species, Alticamelus priscus , from the Sheep Creek beds. Matthew (1924:187) also referred material from Snake Creek to Alticame¬ lus leptocolon described by him from the Pawnee Creek area of Colorado. Davidson (1923:399) described Alticamelus alexandrae from Barstow, California, but Macdonald (1949:190) referred this form to Hesperocamelus. Henshaw (1942:153) named a species from Tonopah, Ne¬ vada, Alticamelus? stocki. Macdonald (1956:199) described Aepycamelus bradyi from the Nightin¬ gale Road fauna, Truckee Formation, Nevada. Leidy (1886:12) described Auchenia major from Mixson, Florida; Leidy and Lucas (1896:53) later changed the name to Procamelus major. Although Simpson (1930:196) referred this material to Me- gatylopus major, it is more likely an advanced species of Aepycamelus (pers. comm., Beryl E. Taylor, 1973). Aepycamelus bradyi is the largest of the species, followed by A. giraffinus. Aepycamelus procerus is smaller than A. giraffinus and has completely lost I 1 ' 2 . Aepycamelus stocki is smaller than A. procerus but larger than A. leptocolon and retains I 1,2 . Aepycamelus priscus is the smallest of the lot. In A. bradyi and A. stocki the premolars are a bit more reduced than in the other species of Aepy¬ camelus but are not so reduced as in Procamelus. Aepycamelus bradyi has as well an almost complete internal crescent on P 3 . Although such a project is not within the scope of this paper, it may be seen that the genus Aepycamelus, often difficult to distinguish from Procamelus, would benefit considerably from a revision. A number of specimens have been referred simply to Aepycamelus. Hesse (1936:66) referred two partial jaws from Beaver Quarry, Oklahoma, to ?Alticamelus. Savage (1941:701) referred a series of metapodials and phalanges from the Optima fauna of Oklahoma, but the dimensions of these specimens are more characteristic of Hemiauchenia. Macdonald (1966:12) described an associated partial skeleton and jaws from the Camp Creek fauna of Nevada. Skinner, Skinner, and Gooris (1968:432) reported a partial radius- ulna of Aepycamelus from Turtle Butte, South Dakota. Webb (1969:147) referred two partial metapodials from Burge Quarry, Nebraska, to Aepycamelus sp. Patton (1969:149) referred limb elements from the Cold Spring fauna and the La para Creek fauna of Texas to Aepycamelus sp. Forsten (1970:48) referred an astragalus and some teeth from the Trail Creek fauna of Wyo¬ ming to ? Alticamelus', Cassiliano (1980:55) changed the reference to Aepycamelus sp. Galusha (1975:54) listed Aepycamelus cf. A. priscus in a preliminary faunal list from the Box Butte For¬ mation of Nebraska. NUMBER 57 5 0 A. bradyi Key 5 Sheep Creek/Snake Creek, ♦ A. stocki Nebraska □ A. leptocolon 6 Box Butte, Nebraska ■ A. giraffmus 7 Pawnee Creek, Colorado A A. procerus 8 Turtle Butte, South Dakota ▲ A. priscus 9 Burge, Nebraska O A. major 10 Beaver Quarry, Oklahoma • A. sp. 11 Clarendon, Texas 1 Nightingale Road, Nevada 12 Cold Spring, Texas 2 Tonopah, Nevada 13 Lapara Creek, Texas 3 Camp Creek, Nevada 14 Mixson, Florida 4 Trail Creek, Wyoming Figure 2.—Geographic distribution of Aepycamelus. Megatylopus Matthew and Cook, 1909 Type-Species. — Pliauchenia gigas Matthew and Cook, 1909:396. Megatylopus is the geologically oldest of the giant Camelini. It ranges from the late Claren- donian into early Blancan throughout the west¬ ern United States (Figure 3). Megatylopus was originally proposed as a subgenus of Pliauchenia and was later elevated to generic rank. The type provenience of M. gigas, the genotypic species, is the late Hemphillian ZX Bar fauna, Snake Creek Formation, Nebraska (Skinner, Skinner, and Gooris, 1977:360). Megatylopus shares with 77- tanotylopus a tendency to reduce the third and fourth premolars. Its teeth are higher crowned than those of Titanotylopus, but both genera are more brachyodont than Megacamelus, Gigantoca- melus, and Camelus. The limbs of Megatylopus, particularly the metapodials, are not shortened in relation to the basal length of the skull. Additional species of Megatylopus are M. coch- rani (Hibbard and Riggs, 1949:854) from Keefe Canyon, Kansas, and M. matthewi (Webb, 6 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY ♦ Key M. gigas 6 Chamita Formation A M. matthewi 7 Snake Creek, Nebraska • M. sp. 8 Wray area, Colorado 1 Little Valley, Juniper Creek, 9 Edson Quarry, Kansas and Black Butte, Oregon 10 Found Quarry, Kansas 2 Smiths Valley, Nevada 11 Optima, Oklahoma 3 Kinsey Ranch, California 12 Coffee Ranch, Texas 4 Wikieup, Arizona 13 Axtel, Currie Ranch, and 5 Redington and Camel Canyon, Christian Ranch, Arizona Texas Figure 3.—Geographic distribution of Megatylopus. 1965:42) from the Coffee Ranch fauna of Texas. Megatylopus cochrani was originally described as Pliauchema cochrani but was transferred to Me¬ gatylopus by Webb (1965:42). Although these workers have commented on the similarity of M. cochrani to Camelops, Corner and Voorhies (pers. comm., 1984) believe that M. cochrani may be more closely related to Titanotylopus. Megatylopus matthewi is distinguished from M. gigas by the complete internal crescent on P\ the greater reduction of P 3 , and a deeper maxillary fossa. In addition to other specimens of M. matthewi, Dalquest (1980:110) described five thoracic ver¬ tebrae preserved in articulation. Based on their structure, he proposed a large dorsal hump, py¬ ramidal in profile rather than rounded. Table 1 lists additional occurrences of Megatylopus. A fourth species of Megatylopus, M. major, was originally described by Leidy (1886:12) as Au- chenia major and based upon an isolated astraga¬ lus from Mixson, Florida. Leidy and Lucas (1896:53) transferred the species to Procamelus and assigned to it a composite dentition, also from Mixson, which they believed had come from a single individual. Simpson (1930:196) changed the identification to ?Megatylopus major, noting that the species was nearly as large as M. gigas. He also distinguished M. major as having broader cheek teeth than M. gigas and a complete internal crescent on P\ Subsequent excavation NUMBER 57 7 Table 1.—Additional occurrences of Megatylopus. Species Locality Reference M. gigas Wray area, Colorado Cook, 1922:1 1 Edson Quarry, Kansas Harrison, 1983:8 M. matthewi Chamita Formation, New MacFadden, 1977:791 Mexico Optima, Oklahoma Schultz, 1977:75 Wikieup, Arizona MacFadden, Johnson, and Opdyke, 1979:357 Ocote, Mexico Dalquest and Mooser, 1980:18 M. cochrani White Bluffs, Washington Gustafson, 1978:48 M. sp. or ?M. Smiths Valley, Nevada Macdonald, 1959:885 Little Valley, Oregon Shotwell, 1970:98 Juniper Creek, Oregon Shotwell, 1970:98 Black Butte, Oregon Shotwell, 1970:96 Kinsey Ranch, California Miller and Downs, 1974:1 1 Axtel, Texas Schultz, 1977:89 Currie Ranch, Texas Schultz, 1977:89 Christian Ranch, Texas Schultz, 1977:89 Redington, Arizona Lindsay, 1978:270 Camel Canyon, Arizona Lindsay, 1978:270 Found Quarry, Kansas Bennett, 1979:12 of the Mixson bone bed produced a much larger sample of this camel, currently housed in the Frick Collection of the Department of Verte¬ brate Paleontology, American Museum of Nat¬ ural History. Skulls and mandibles bearing com¬ plete dentitions as well as diagnostically elongate metapodials and cervical vertebrae indicate that M. major should be transferred to Aepycamelus major (pers. comm., Beryl E. Taylor, 1973). Titanotylopus Barbour and Schultz, 1934 Type-Species. — Titanotylopus nebraskensis Barbour and Schultz, 1934:291. Barbour and Schultz (1934:291) described Ti¬ tanotylopus nebraskensis based on a single mandi¬ ble from a Pleistocene gravel pit near Red Cloud, Nebraska (Figure 4). Other than a proximal pha¬ lanx from another Pleistocene gravel pit, no more specimens have been referred to T. nebras¬ kensis. Based upon this material, Webb (1965) and Breyer (1976) felt that the differences be¬ tween Gigantocamelus and Titanotylopus war¬ ranted distinction only at the specific level. Cor¬ ner and Voorhies (pers. comm., 1984) have since identified additional material, which is referrable to T. nebraskensis and which they believe validates the generic independence of these two taxa. I agree with Corner and Voorhies that the name Titanotylopus should be applied to a single spe¬ cies, T. nebraskensis, as yet known only from early Irvingtonian localities in Nebraska. The type mandible of Titanotylopus is very large, 662 mm long, with a dental formula of I 3 C 1 P 0 M 3 . The mandibular symphysis is long and extends well beyond the large canines. The non- spatulate incisors are arrayed in an arc. There is no indication of Pj. P 3 is broken, but its position is indicated by alveoli for its two roots. P 4 , like the molars, has sustained some breakage. All of the cheek teeth are quite brachyodont. In their study of Gigantocamelus spatulus from Keefe Canyon, Hibbard and Riggs (1949) classi¬ fied as female those jaws with small Ci and no P], whereas jaws classified as male had large Cj, and P) was present. Webb (1965:36) felt that the type of T. nebraskensis fell within the range of variation assigned to females and hence attached 8 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Figure 4.—Type-localities of Titanotylopus nebraskansis and Blancocamelus meadei. little significance to the missing Pj. His synonymy of Titanotylopus and Gigantocamelus has been fol¬ lowed by most authors except Hibbard (Skinner et al., 1972:114). Considerations of P] aside, the highly derived chin and greater degree of hyp- sodonty in G. spatulus as well as the much shorter distance between Cj and P 3 in T. nebraskensis must weigh heavily against the congenerity of these two species. Gigantocamelus Barbour and Schultz, 1939 Type-Species.— Gigantocamelus fricki Barbour and Schultz, 1939:17 (= Pliauchenia spatula Cope, 1893:70; = Gigantocamelus spatulus Meade, 1945:531; = Titanotylopus spatulus (Meade) fide Webb, 1965:36). Gigantocamelus is known from Blancan locali¬ ties throughout the central and western United States (Figure 5). The giant camel from the Blanco beds of Texas was first described as Pliauchenia spatula by Cope (1893:70). Barbour and Schultz (1939:17) subsequently described a large sample of giant camel from Lisco, Ne¬ braska, as Gigantocamelus fricki. Meade (1945:531) recognized these camels as the same species and united them under the name Gigan¬ tocamelus spatulus. Hibbard and Riggs (1949:844) followed Meade when they described a third large sample of this giant camel from Keefe Canyon, Kansas. Gigantocamelus is a much more hypsodont camel than Titanotylopus. Its chin is blunt with the ramal symphysis extending only a few centi¬ meters beyond the large, splayed canines. The spatulate lower incisors are arrayed almost trans¬ versely. In Titanotylopus the chin is long, Ci is large but not greatly splayed, and the nonspatu- late incisors are arrayed in an arc. Pj is present in Gigantocamelus and absent in Titanotylopus. As indicated in the preceding section, many workers have followed Webb (1965) and Breyer (1976) NUMBER 57 9 • Key G. spatulus 6 White Rock, Kansas ■ G. sp. 7 Kentuck, Kansas 1 Grand View, Idaho 8 Keefe Canyon, Kansas 2 Delmont, South Dakota 9 Holloman, Oklahoma 3 Sand Draw, Nebraska 10 Mt. Blanco, Texas 4 Lisco, Nebraska 11 Gilliland, Texas 5 Donnelly Ranch, Colorado 12 Hudspeth, Texas Figure 5. —Geographic distribution of Gigantocamelus. in synonymizing Gigantocamelus and Titanotylo- pus. I believe that the above characters, in addi¬ tion to the new data from Corner and Voorhies, support the generic validity of Gigantocamelus. Several other workers have made reference to Gigantocamelus in one or more of its previous taxonomic incarnations. Dumble (1894:559), Wortman (1898:128), Gidley (1903:627), Mat¬ thew (1901:423; 1909:120), and Merriam (1917:435) noted the presence of Pliauchenia spatula at Mt. Blanco. Matthew (1899:75) listed material from Goodnight, Texas, as P. spatula, but this is more likely Megatylopus. From the Saw Rock fauna of Kansas, Hibbard (1953:407) re¬ ferred a toe bone to Gigantocamelus cf. G. spatu- lus. Although I have not seen this specimen, the measurements fall within the range of Megaca- melus. If indeed the toe can be identified as Gigantocamelus , it would be the earliest occur¬ rence of this genus. From the Gilliland fauna of Texas, Hibbard and Dalquest (1962:86) referred a distal radius, some distal metapodials, an as¬ tragalus, and three phalanges to ?Gigantocamelus , and later added a cervical vertebra and changed the identification to fTitanotylopus, reflecting Webb’s synonymy. Strain (1966:50) referred two distal metapodials from the Hudspeth fauna of Texas to Gigantocamelus sp. Semken (1966:164) referred a partial calcaneum from the Kentuck fauna of Kansas to Gigantocamelus sp. A number of foot and limb bones from the Grand View fauna of Idaho were referred to Titanotylopus sp. by Shotwell (1970:96). Hibbard (Skinner et ah, 1972:114) referred some material from the Sand 10 Draw fauna of Nebraska to G. spatulus. Martin and Harksen (1974:14) referred a partial man¬ dible from the Delmont fauna of South Dakota to Titanotylopus. Dalquest (1975:42) described additional specimens of T. spatulus from Mt. Blanco. Eshelman (1975:47) referred a proximal ulna from the White Rock fauna of Kansas to Gigantocamelus sp. Hager (1975:14) referred some tooth fragments and foot bones from the Donnelly Ranch fauna of Colorado to Giganto¬ camelus sp. Dalquest (1977:260) described a ra¬ dius of T. spatulus from the Holloman fauna of Oklahoma. Corner and Voorhies (pers. comm., 1984) regard the specimens from the Gilliland and Holloman faunas as generically indetermi¬ nate between Gigantocamelus and Titanotylopus. Ferrusquia-Villafranca (1978:255) listed Gigan¬ tocamelus mexicanus and G. magnus from the Mex¬ ican faunal assemblage. Dalquest (1974:196- 197) noted that these two species were based SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY upon the same type material, G. mexicanus being the senior synonym, and identified them as Ca- melops. Megacamelus Frick, 1929 Type-Species. — Pliauchenia merriami Frick, 1921:358 (= Megacamelus merriami, new combi¬ nation). Megacamelus is presently known only from the late Hemphillian of the southwestern United States (Figure 6). Frick (1921:358) described Pliauchenia merriami from Mt. Eden, California, and later (1929:107) named Megacamelus blicki from Kearns Canyon, Arizona. Study indicates that these camels are conspecific. The Mt. Eden camel is clearly not Pliauchenia and although Webb (1965:36) referred it to Titanotylopus, it does not belong to that genus for reasons dis¬ cussed below. Therefore, a new taxonomic com- Key • M. merriami, new combina¬ tion 1 Mt. Eden, California 2 Reams Canyon, Arizona Figure 6 . —Geographic distribution of Megacamelus. NUMBER 57 11 bination, Megacamelus merriami, is proposed for the giant camels from Mt. Eden and Kearns Can¬ yon. Megacamelus merriami (Frick, 1921), new combination Figures 7-16 Pliauchenia merriami Frick, 1921:358. Megacamelus blicki Frick, 1929:107. Holotype.— UCMP 23483, anterior portion of the upper jaws bearing right and left I 3 , C 1 , P'; anterior portion of the lower jaws bearing right 12.3, Ci, Pi, partial P 3 and left I 2i3 , Ci, Pu distal humerus, proximal radius-ulna, distal ra¬ dius-ulna, scaphoid, lunar, cuneiform, pisiform, distal tibia, astragalus, calcaneum, navicular, cu¬ boid, ectocuneiform, proximal metatarsus, 2 dis¬ tal metapodials, 6 proximal phalanges, 5 medial phalanges, 7 distal phalanges, and 8 sesamoids. Referred Specimens.— From Mt. Eden, Cal¬ ifornia: UCMP 23416, left P 3 ; UCMP 23433, right partial upper molar; UCMP 23783, left I 2 ; UCMP 23790, right I 3 ; UCMP 23791, left I 3 ; UCMP 23435, left M 2(?) ; UCMP 23789, left M, (?) . From Kearns Canyon, Arizona (all of the fol¬ lowing are F:AM numbers): skulls, 23201, 23202, 23202A, 23203, 23207; partial skulls, 23203A, 23203B, 23203C, 23204, 23205, 23205A, 23208, 23209, 104395; partial maxil¬ lae, 23206, 23210; mandibles, 23216, 23218, 23220, 23230, 23231, 23232, 23233, 23235, 23239A; rami, 23217A, 23217B, 23219, 23221, 23222, 23223A, 23223B, 23225,23226, 23227, 23228, 23229, 23233A, 23234, 23240B; partial rami, 23224, 23317, 23320, 104274; incisors, 104329,104330,104333,104334,104335;ca¬ nine, 104383; associated postcrania, 23312 (as¬ tragalus, calcaneum, tarsal fragments, and meta¬ tarsus); atlas, 104291,104292,104293,104294, 104295; axis, 104284, 104299; cervical verte¬ brae, 104281, 104282, 104283, 104285, 104286, 104287, 104288, 104289, 104290, 104297, 104298; thoracic vertebrae, 104286, 104296, 104305; lumbar vertebrae, 104277, 104300, 104301, 104304; sacra, 104280, 104302; scapulae, 23245, 23246, 23247, 23248, 23249, 23250, 23251, 23275, 23276, 104278, 104279, 104408, 104607; humeri, 23254, 23255,23256,23257,23258,104405,104406, 104407; radius-ulnae, 23260, 23261, 23261A, 23262, 23263, 23264, 23265, 23266, 23267, 23268, 23269, 23270, 104400, 104600;scaph- oids, 104311, 104351, 104352, 104376, 104377, 104603; lunars, 104309, 104323, 104324, 104349, 104350, 104361; cuneiforms, 104325, 104332, 104389, 104605; pisiforms, 104312, 104345, 104356, 104370, 104371, 104372; trapezoids, 104316, 104341, 104359; magna, 104339, 104360, 104390; unciforms, 104307, 104308, 104342, 104343, 104344, 104346, 104364; metacarpi, 23277, 23278, 23279, 23280, 23281, 23282, 23283, 23284, 23285,23301, 104401; pelves, 104303, 104396; femora, 23290, 23291, 23292, 23293, 23294, 23295, 104403, 104404; patellae, 104358, 104394; tibiae, 23271, 23296, 23297, 104399; distal fibula, 104347, 104375; astragali, 104229, 104230, 104231, 104232, 104233, 104234, 104235, 104236,104272, 104273, 104602; cal¬ canea, 104237, 104238, 104239, 104240, 104601; naviculars, 104310, 104331, 104373, 104374; cuboids, 104306, 104336, 104353, 104354, 104355, 104362, 104363; ectocunei- forms, 104313, 104322, 104378, 104379, 104380, 104381, 104382, 104384, 104385, 104386, 104392, 104393; metatarsi, 23302, 23303, 23304, 23305, 23306, 23307, 23308, 23313; partial metapodials, 23314, 23316, 104326, 104327, 104328, 104397, 104398, 104402; sesamoids, 104315, 104317, 104318, 104319, 104320, 104321, 104337, 104338, 104348, 104357, 104365, 104366, 104367, 104368, 104369, 104387, 104388, 104606; proximal phalanges, 104241, 104242, 104243, 104244, 104245, 104246, 104247, 104248, 104249, 104250, 104251, 104252, 104253, 104254, 104255, 104256, 104257, 104258, 104259, 104260, 104261, 104262, 104263, 104265,104266,104267, 104268, 104269; me¬ dial phalanges, 104210, 104211, 104212, 104213, 104214, 104215, 104216, 104217, 12 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Table 2.— Measurements (cm) of the skull and upper dentition of Megacamelus merriami, new combination, from Kearns Canyon (O.R. = observed range, X = sample mean, s.d. = standard deviation). Element No. O.R. X s.d. Length from premaxilla 3 69.30-82.05 73.65 to occipital crest Length from premaxilla 3 61.67-71.35 65.54 to occipital condyles Width l 3 Width of muzzle across 3 6.68-8.25 7.68 of muzzle across 3 7.69-8.59 8.16 C 1 Minimum width at post- 4 6.71-9.65 8.45 orbital constriction Maximum width across 2 26.40-27.34 26.87 zygomatic arches Width of occipital con- 7 9.25-11.55 10.14 0.76 dyl es Diastema I 3 -C‘ 2 2.10-2.30 2.20 Diastema C'-P 1 2 3.08-3.27 3.17 Diastema P'-P 3 3 7.10-8.83 7.81 I 3 length 2 2.20-3.19 2.69 width 2 1.22-1.73 1.47 C 1 length 3 2.88-3.46 3.15 width 3 1.82-2.33 2.11 p 1 length 4 1.94-2.52 2.20 width 4 1.38-1.74 1.54 p3 length 6 1.84-3.05 2.62 0.44 width 7 1.88-2.03 1.95 0.06 P 4 length 7 2.85-3.25 3.03 0.18 width 6 2.75-3.31 3.04 0.22 M 1 length 6 4.30-4.75 4.54 0.15 width 6 3.80-4.45 4.11 0.27 M 2 length 8 4.89-5.66 5.42 0.35 width 6 3.82-4.75 4.46 0.34 M 3 length 7 5.40-6.13 5.69 0.28 width 6 3.86-4.39 4.20 0.19 I 3 -M 3 length 3 37.77-41.10 39.06 p 3 -m 3 length 5 18.96-20.03 19.41 dP 2 length 2 1.39-1.50 1.44 width 2 0.42-0.82 0.62 dP 3 length 3 3.29-3.60 3.43 width 3 2.45-2.65 2.55 dP 4 length 3 3.72-4.15 3.86 width 3 3.00-3.20 3.07 dP 2 -M 2 length 1 16.00 dP 3 -M 2 length 2 14.88-15.96 15.42 104218, 104219, 104220, 104221, 104222, 104223, 104224, 104225, 104270, 104604; dis¬ tal phalanges, 104226, 104227, 104228, 104271, 104314, 104340, 104391. Table 3.—Measurements (cm) of the mandible and lower dentition of Megacamelus merriami, new combination, from Kearns Canyon (O.R. = observed range, X = sample mean, s.d. = standard deviation). Element No. O.R. X s.d. Maximum length of ra- 5 12.81-17.10 14.89 mal symphysis width of manidbular 8 4.35-5.93 5.36 0.51 condyle Diastema I 3 -C 1 4 0.63-1.06 0.88 Diastema C]-P] 5 3.36-4.38 3.90 Diastema P 1 -P 3 6 7.26-8.23 7.55 0.41 Ci length 6 3.19-3.88 3.48 0.25 width 6 2.19-3.17 2.58 0.41 Pi length 6 2.23-2.63 2.36 0.18 width 6 1.23-1.89 1.49 0.29 P 3 length 8 1.87-2.38 2.08 0.16 width 7 1.16-1.35 1.24 0.06 P 4 length 10 2.72-3.26 2.96 0.18 width 9 1.72-2.07 1.88 0.12 M| length 12 4.05-5.05 4.41 0.33 width 12 2.32-2.90 2.64 0.19 M 2 length 12 4.86-5.44 5.09 0.28 width 14 2.27-3.53 2.96 0.38 M 3 length 11 5.98-6.84 6.49 0.26 width 9 2.61-3.43 3.04 0.29 C|-M 3 length 5 36.05-39.41 38.09 P 3 -M a length 6 20.34-21.74 20.71 0.51 dP 2 length 3 1.06-1.31 1.18 width 3 0.12-0.13 0.13 dP 3 length 3 2.05-2.29 2.18 width 3 1.20-1.27 1.23 dP 4 length 2 4.21-5.30 4.75 width at pos- 2 2.06-2.16 2.11 terior cusp Description.— The extensive sample from Kearns Canyon contains several fine skulls and mandibles as well as a wealth of postcrania (Fig¬ ures 7-11). With one exception (F:AM 23312), none of the material is associated. The Kearns Canyon camel compares well with that described by Frick (1921) from Mt. Eden. Both exhibit very large, caniniform I 3 , C 1 , and P\ a massive premaxilla, and heavy anterior ramus with large Ci and large, caniniform P]. The size and pro¬ portions of the postcrania from both localities are comparable. The skull of M. merriami is long with a flat¬ tened dorsal profile and a deep, massive rostrum. The occipital crest is a broad fan that extends 14 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Figure 8. —Megacamelus merriami, new combination, from Kearns Canyon, Arizona: 1, F:AM 23201, skull, lateral view; 2, 3, F:AM 23232, mandible, occlusal and lateral views. (X ‘ A .) NUMBER 57 15 Figure 9. —Megacamelus merriami, new combination, from Kearns Canyon, Arizona, F:AM 23202, skull: 1, dorsal view; 2, occlusal view. (X %.) well posterior to the occipital condyles. The sag¬ ittal crest is likewise well developed. The orbit is circular in outline followed by a strong postor¬ bital bar. A triangular lacrimal vacuity is present in most of the Kearns Canyon M. merriami , but it is a highly variable feature. In one specimen. F:AM 23202, it is present on the left side of the skull, but reduced to a slit on the right side (Figure 9). Another skull, F:AM 23202A, bears well-developed lacrimal vacuities on both sides (Figure 10). Although Meade (1945:531) re¬ ported the presence of a lacrimal vacuity in M. spatulus from Mt. Blanco, Hibbard and Riggs (1949:846) reported its absence and, moreover, suggested that the opening in the Mt. Blanco skulls could represent an artifact of preservation. The upper dentition consists of l\ C 1 , P 1 3 4 , M 1-3 . I s is always present and, as mentioned above, is large and caniniform. C 1 is large and deviates slightly from a vertical orientation. Pre¬ sumably, when Barbour and Schultz (1939:24) stated that the canines of the Kearns Canyon Figure 10 .—Megacamelus mernami, new combination, from Kearns Canyon, Arizona, F:AM 23202A, skull: 1, dorsal view; 2, occlusal view. (X 'A.) Note the large lacrimal vacuities. camel were not enlarged, it was in comparison to the extreme development observed in Giganto- camelus spatulus and Titanotylopus nebraskensis. P 1 is caniniform and only slightly smaller than I \ P is large with an incomplete internal crescent. P 4 is not much longer than P' but much wider due to its complete internal crescent. Only the parastyle shows much development on P 4 , but a strong parastyle and mesostyle are present on each molar. The teeth are higher crowned than those of Megatylopus gigas or Titanotylopus ne¬ braskensis. The mandible is long and massive but still smaller than that of Gigantocamelus spatulus or Titanotylopus nebraskensis. Idle mandibular pro¬ portions of M. merriami agree well with those of G. spatulus; however, M. merriami does not dis¬ play the degree of variation in the symphyseal NUMBER 57 Table 4. —Measurements (cm) of the postcrania of Megacamelus merriami, new combination, from Kearns Canyon (O.R.= observed range, X = sample mean, s.d. = standard deviation). 17 Element No. O.R. X s.d. Element No. OR. X s.d. Atlas Tibia length of centrum 2 6.75-6.80 6.77 length 1 67.00 posterior height 1 9.36 distal width 2 10.95-11.45 11.20 Scapula Astragalus maximum length 2 61.53-64.06 62.79 length (tibial to maximum width 1 38.35 tarsal surface) glenoid fossa medial 7 8.19-8.98 8.57 0.34 antero- 3 9.37-10.56 9.87 lateral 7 8.85-9.80 9.47 0.30 posterior distal width 7 6.03-6.65 6.39 0.21 transverse 3 9.17-10.61 9.66 Calcaneum Humerus length 4 17.68-19.56 18.64 length 1 53.27 maximum antero- 4 7.21-8.30 7.90 distal width across posterior trochlea 5 10.60-11.70 11.14 Metatarsus Radius-ulna length 5 48.37-50.58 49.44 maximum length 4 72.65-85.70 80.33 proximal width 2 8.06-9.37 8.71 articular length 4 66.03-75.50 71.22 distal width 4 10.01-10.75 10.35 proximal width 6 10.19-11.52 10.87 0.43 Proximal phalanx distal width 5 10.40-12.62 11.39 length 21 11.05-13.80 12.41 0.84 Metacarpus proximal width 21 4.55-5.73 5.08 0.35 length 7 48.39-54.29 51.08 2.07 distal width 22 3.82-5.14 4.42 0.41 proximal width 8 8.11-11.64 9.12 1.10 Medial phalanx distal width 7 11.38-12.30 11.88 0.33 length 18 7.22-8.47 7.76 0.34 Femur proximal width 18 3.74-4.99 4.19 0.27 length 1 60.89 distal width 17 3.26-4.21 3.79 0.25 proximal width 1 16.07 Distal phalanx distal width 2 13.98-15.34 14.66 length 6 3.16-4.16 3.71 0.34 width of patellar surface 3 5.07-5.79 5.46 width of articular 6 2.45-2.80 2.60 0.13 surface region observed in G. spatulus by Meade (1945:532) or Hibbard and Riggs (1949:847). Meade, in the Mt. Blanco sample, and Hibbard and Riggs, in the Keefe Canyon sample, found jaws with widely splayed and canines and trans¬ versely arrayed incisors as well as jaws with more vertically oriented canines and more convention¬ ally arrayed incisors. Hibbard and Riggs (1949), Webb (1965), and Breyer (1976) have attributed such variation to sexual dimorphism. In the Kearns Canyon sample the incisors are procum¬ bent and arrayed in a more shallow arc than in Titanotylopus. The canines are very large, but only slightly splayed. No specimen displays the degree of canine flare and incisor-row bluntness characteristic of G. spatulus (Cope, 1893, pi. 21; Barbour and Schultz, 1939, fig. 9; Meade, 1945, pi. 54; Hibbard and Riggs, 1949, fig. 8). A groove is present between the median incisors on the ventral symphyseal surface, as noted by Cope (1893:71) and Meade (1945:532). The lower dentition consists of 1 1 _ 3 , Ci, Pi, 3,4, Mj_ 3 . The incisors have spatulate crowns that with much wear assume a more rounded, peg- 18 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Figure 11.— Megacamelus merriami, new combination, from Kearns Canyon, Arizona: 1, 4, F:AM 23216, mandible, occlusal and lateral views; 2, 3, F:AM 23239A, partial ramus bearing deciduous P 2 _4 and M, emerging, occlusal and lateral views. (X 'A.) like appearance. The canines bear the strong anterointernal and posterior enamel ridges typi¬ cal of G. spatulus. Pi is present in all specimens and usually well developed. An exception is F:AM 23218, one of the smallest individuals, in which Pj is correspondingly small. The cheek tooth series is quite similar to that of M. spatulus. Meade (1945:533) reports an anteroexternal style or “llama buttress” on M 2 . 3 of one specimen and notes the presence of this feature in figures of M. spatulus from Lisco. No indication of a “llama buttress” is present in M. merriami. The limbs and feet of M. merriami, especially the metapodials and phalanges, do not exhibit the shortening in relation to the basal length of the skull seen in G. spatulus. Hence, Barbour and 19 Figure 12 .—Megacamelus merriami, new combination, from Kearns Canyon, Arizona: 1, 2, F:AM 104293, atlas, dorsal and ventral views; 3, F:AM 104281, sixth cervical vertebra, lateral view; 4, 5, F:AM 104284, axis, dorsal and ventral views; 6, F:AM 23245, left scapula, lateral view. (X l A.) Schultz (1939:24) remarked that “the skeletal elements appear to be more massive in the Ne¬ braska form” (= G. spatulus). The limbs of M. merriami are, however, shorter and stockier than those of Megatylopus. Discussion.— Megacamelus merriami is most closely related to Gigantocamelus spatulus but dif¬ fers from it in the presence of the large, canini- form I 3 , smaller size, less shortened limbs, and lower-crowned teeth. Breyer (1983:305) re- 20 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Figure 13.— Megacamelus merriami, new combination, from Kearns Canyon, Arizona: 1, 2, F:AM 23256, humerus, anterior and posterior views; 3, 4, F:AM 23262, radius-ulna, anterior and posterior views. (X ' A .) ferred the Kearns Canyon material to Titanotylo- pus nebraskensis on the basis of the projection of the mandibular symphysis beyond the canines. This condition is primitive for camelines and hence not a valid criterion. Moreover, several characters such as the presence of Pi in all spec¬ imens, greater degree of hypsodonty, and the greater distance between Ci and P 3 preclude the referral of the Kearns Canyon camel to Titanoty- lopus. Camelus Linnaeus, 1758 Type-Species. — Camelus dromedarius Lin¬ naeus, 1758:65. Camelus is the smallest of the giant camels. The two extant species, G. dromedarius (mono- gibbose) and C. bactrianus (digibbose), range throughout most of the arid and semi-arid re¬ gions of the Old World. Camelus bactrianus is native to Chinese Turkestan and Mongolia, where small wild populations still exist (Walker, 1964:1374). Both C. bactrianus and C. dromedar¬ ius have been domesticated for several thousand years, and the original native range of the latter species can no longer be determined. Camelus has an extensive fossil record in the Pleistocene of the Old World and has been found in association with human artifacts and remains (Gauthier-Pilters and Dagg, 1981:5). Camels mi¬ grated from North America via Beringea near the end of the Tertiary, probably during the late Ruscinian. Camelus (Paracamelus ) Schlosser (1903) occurs in several late Pliocene localities in the People’s Republic of China. As yet no fossil material of Camelus or Paracamelus has been recovered from North America. Figure 14.— Megacamelus mernami, new combination, from Kearns Canyon, Arizona: 1, 2, F:AM 10431 1, scaphoid, medial and lateral views; 3, 4, F:AM 104323, lunar, medial and lateral views; 5, 6, F:AM 104325, cuneiform, medial and lateral views; 7, 8, F:AM 104312, pisiform, medial and lateral views; 9, 10, F:AM 104359, trapezoid, posteromedial and anterolateral views; 11, 12, F:AM 104360, magnum, proximal and distal views; 13, 14, F:AM 104343, unciform, proximal and distal views; 15, 16, F:AM 23279, metacarpus, anterior and posterior views. (X 'A.) Figure 15.— Megacamelus merriami, new combination, from Reams Canyon, Arizona: 1, 2, F:AM 23293, femur, anterior and posterior views; 5, 4, F:AM 23296, tibia, anterior and posterior views. (X ' A .) Figure 16 .—Megacamelus mernami, new combination, from Kearns Canyon, Arizona: 1-3, F:AM 104239, calcaneum, lateral, medial, and anterior views; 4, 3, F:AM 104255, proximal phalanx, anterior and posterior views; 6, 7, F:AM 104222, medial phalanx, anterior and posterior views; 8-10, F:AM 104231, astragalus, anterior, posterior and lateral views; 11, 12, F:AM 104226, distal phalanx, anterior and posterior views; 13, 14, F:AM 104379, ectocunei- form, proximal and distal views; 13, 16, F:AM 104366, cuboid, proximal and distal views; 17, 18, F:AM 23307, metatarsus, anterior and posterior views; 19, 20, F:AM 104310, navicular, proximal and distal views; 21, 22, F:AM 104375, distal Fibula, medial and lateral views. (X ' A .) Blancocamelus Dalquest, 1975 Type-Species. — Blancocamelus meadei Dal¬ quest, 1975:37. This genus is represented solely by B. meadei, described by Dalquest (1975:37) from Mt. Blanco, Texas. Dalquest noted that although Meade (1945:538) was aware of the uniqueness of this camel, he mistakenly applied to it an unpublished name, Leptotylopus percelsus, from a 1924 manuscript of W.D. Matthew. As used by Meade, the name was a nomen nudum. The taxon to which Matthew had applied the name in manuscript was subsequently identified as Ta- nupolama (= Hemiauchenia) blancoensis. Thus, the genus was left without a valid name until Dalquest (1975) proposed Blancocamelus meadei for it. 24 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Blancocamelus is known only from postcranial elements. Its limbs are exceedingly long, but quite slender, evoking mental images of a giant Hemiauchenia. Indeed, the posterior surface of the proximal phalanx presents an asymmetrical, W-shaped scar for the attachment of the suspen¬ sory ligament that is quite like that of Hemiauch¬ enia (Breyer, 1976, fig. 2). Although Meade (1945) and Kurten and Anderson (1980:302) have speculated upon the possible affinities of Blancocamelus and the aepycamelines, I prefer for the present to group it with the lamines. With the exception of a possible occurrence in the Blancan Red Light fauna of Texas (Akersten, 1972:29), Blancocamelus is restricted to the type- locality (Figure 4). rounded in cross section as in the giant camelines. P{’ 1,3 are lost and P| are reduced. The molars are relatively narrow with external styles less strongly developed than in the camelines. The limbs of Camelops are sturdy and the metapodials less slender than those of the other lamines. The area of attachment for the suspen¬ sory ligament on the posterior surface of the proximal phalanx is distinctive (Breyer, 1974, fig. 2b). Camelops hesternus, C. traviswhitei, and C. huerfanensis are the only species considered within the scope of giant camels. More detailed descriptions of Camelops are given in Savage (1951) and Webb (1965). Summary Camelops Leidy, 1854 Type-Species. — Camelops kansanus Leidy, 1854:172. Camelops is by far the best known of the lam¬ ines, giant or otherwise. It occurs from the late Blancan into the early Holocene in localities throughout the western United States (Kurten and Anderson, 1980, fig. 15.4). Since its descrip¬ tion by Leidy (1854:172), the genus Camelops has undergone a bewildering series of synonymies, referrals, and revisions. Much of this morass was clarified by Webb (1965), who followed Savage (1951) in recognizing five species: C. kansanus Leidy (1854:172), C. hesternus (Leidy, 1873:255), C. huerfanensis (Cragin, 1892: 258), C. sulcatus (Cope, 1893:84), and C. minidokae Hay (1927:93). These five species plus C. travis¬ whitei Mooser and Dalquest (1975:341) were rec¬ ognized by Kurten and Anderson (1980). Camelops, especially the later species, is very hypsodont with large lacrimal vacuities and marked maxillary fossae. The skull is long and does not display the rostral shortening character¬ istic of other lamines such as Hemiauchenia, Lama, and Vicugna. The mandible is long with a sharp diastemal crest and uninflected angular proc¬ esses. The dental formula is I 3 C} P? M3. In Ca¬ melops U and Cj are reduced, laterally com¬ pressed, and recurved rather than enlarged and The trend toward gigantism in camelids is first evident in Aepycamelus in the late Clarendonian and continued throughout the rest of the Ceno- zoic (Figure 17). Eight camelid genera are treated as giant camels in this paper. Megacamelus merriami, new combination, is proposed for the large, late Hemphillian camel from Mt. Eden and Kearns Canyon. The giant camels from Mt. Blanco, Lisco, and Keefe Can¬ yon are referred to Gigantocamelus spatulus. Ti- tanotylopus is applied only to T. nebraskensis. Me- gatylopus major is transferred to Aepycamelus. The Camelini were all very large camels, but ////// / ^