A Review of the Extinct Wolverine, Plesiogulo (Carnivora: Mustelidae), from North America JESSICA A. HARRISON SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY • NUMBER 46 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 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. S. Dillon Ripley Secretary Smithsonian Institution SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY • NUMBER 46 A Review of the Extinct Wolverine, Plesiogulo (Carnivora: Mustelidae), from North America Jessica A. Harrison SMITHSONIAN INSTITUTION PRESS City of Washington 1981 ABSTRACT Harrison, Jessica A. A Review of the Extinct Wolverine, Plesiogulo (Carni¬ vora: Mustelidae), from North America. Smithsonian Contributions to Paleobiology , number 46, 27 pages, 16 figures, 1981.—There are two fossil species of Plesiogulo in North America: Plesiogulo marshalli , to which most of the fossil material is herein referred, and P. lindsayi, new species. Both species are restricted to the late Hemphillian. A formal diagnosis is offered for P. marshalli. Remains of Plesiogulo are relatively rare, possibly due to low densities in extinct populations. The only known juvenile specimens of Plesiogulo are from the Edson Local Fauna, where three juveniles together with a single mature individual probably represent a female with a litter of cubs. Plesiogulo migrated to the New World some time between 7.0 and 6.5 million years ago. This taxon, generally interpreted as an inhabitant of forest or woodland, was probably equally well adapted to the open plains. 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 Harrison, Jessica A. A review of the extinct wolverine, Plesiogulo (Carnivora, Mustelidae) (Smithsonian contributions to paleobiology ; no. 46) Bibliography: p. Supt. of Docs. no. : SI 1.30:46 1. Plesiogulo. 2. Paleontology—Pliocene. 3. Paleontology—North America. I. Title. II. Series QE701.S56 no. 46 [QE882.C15] 560s [569'.74] 81-607075 AACR2 Contents Page Introduction . 1 Generic Relationships . 1 Paleoecology . 2 Geographic Distribution . 3 Temporal Distribution . 4 Abbreviations . 4 Measurements . 5 Acknowledgments . 5 System at ics . 5 Plesiogulo marshalli (Martin, 1928) . 5 Plesiogulo lindsayi, new species . 18 The WaKeeney Specimen . 25 Conclusions . 25 Literature Cited . 26 A Review of the Extinct Wolverine, Plesiogulo (Carnivora: Mustelidae), from North America Jessica A. Harrison Introduction Plesiogulo , an extinct relative of the living wol¬ verine Gulo, is known from 14 Hemphillian local¬ ities in North America. Remains, however, are seldom abundant, and at only a handful of local¬ ities are more than one or two individuals repre¬ sented. Although complete skulls and mandibles have been recovered from Old World localities, most notably those in the Paote area of the People’s Republic of China, the North American material consists, for the most part, of partial dentitions and isolated teeth. Although Plesiogulo has been recorded in the literature from seven localities, several additional localities and the bulk of the fossil material re¬ mained undescribed. In 1928, H. T. Martin des¬ ignated a nearly complete right ramus from the Edson Local Fauna as the type of Brachypsalis marshalli and referred a C 1 and a maxillary frag¬ ment bearing P 4 -M J Shortly thereafter, Hibbard (1934) transferred this material to Plesiogulo in a faunal list. Matthew and Stirton (1930), Reed and Longnecker (1932), and Sellards, Adkins, and Plummer (1932) produced lists of the mam¬ mals from the type Hemphillian Coffee Ranch Jessica A. Harrison, Department of Paleobiology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. Local Fauna of Texas, but none contained Plesio- gulo. Hesse (1936:67) indicated the presence of Plesiogulo at Coffee Ranch, and Savage (1941:705) listed ? Plesiogulo from Optima (=Guymon), Okla¬ homa, but in neither paper was the nature of the fossil material discussed. Shotwell (1955:332) also included Plesiogulo in a faunal list of Coffee Ranch, and he (1956:733) identified as Plesiogulo sp. an associated P 4 -M* and an isolated Mi from McKay Reservoir in Orgeon. Wilson (1968:111) erroneously referred to Plesiogulo an isolated M 1 from the WaKeeney Local Fauna of Kansas. In 1969:11 Dalquest referred to P. marshalli an iso¬ lated M 1 and a partial Mi from Coffee Ranch. Schultz (1977:75) confirmed the presence of Ple¬ siogulo in a list of the Optima Local Fauna. MacFadden (1977:789) noted a palate of Plesi¬ ogulo sp. from San Juan Quarry, New Mexico. McFadden, Johnson, and Opdyke (1979:357) re¬ ferred the material from the Wikieup Local Fauna, Arizona, to P. marshalli ; this material con¬ stitutes the hypodigm of P. lindsayi , new species, described herein. Baskin (1979, fig. 2) illustrated an isolated M 1 of Plesiogulo from the White Cone Local Fauna of Arizona, and Bennett (1979:5) referred an isolated P 4 from Lost Quarry, Kansas, to P. marshalli. Generic Relationships. —The general resem¬ blance of Gulo and Plesiogulo has been noted by many workers (Miller, 1912:433; Zdansky, 1924: 1 2 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY 38; Webb, 1969:65). Viret (1939:15) and Kurten (1970:20) maintain that Plesiogulo , in particular P. monspessulanus , was ancestral to Gulo. However, I agree with Zdansky (1924:44) and Webb (1969: 65) that such a line of descent would require an inordinate number of evolutionary reversals. As early as 1912, Miller (1912:432) advocated a separate subfamily, the Guloninae, for Gulo. This classification was followed by Pocock (1920: 187), who also agreed with Gill (1872:66), that Mellivora should be placed within its own subfam¬ ily, the Mellivorinae. Webb (1969:68) supported the removal of Gulo from the Mustelinae, but placed it instead in the Mellivorinae, where, to¬ gether with Plesiogulo , Hadnctis , and Ischyrictis, it constituted the Gulonini. The subsequent discov¬ ery of new fossil mustelids exhibiting gulonine/ mellivorine affinities (Bryant, 1968:2; Bjork, 1970:19), in addition to new material of existing taxa (Hendey, 1978:330), further complicates the phylogeny of this interesting group of carnivores. Paleoecology.— The diet of Gulo consists chiefly of meat, augmented in the summer months by berries, eggs, and wasp larvae (Myhre and Myrberget, 1975; Krott, 1960). A large per¬ centage of the meat is carrion. Although Gulo possesses both the strength and ferocity requisite for the taking of large prey, smaller animals such as rodents, birds, fish, and frogs are also common dietary items (Grinnell, Dixson, and Linsdale, 1937:268; Jackson, 1961:362). Morphological similarities in the dentition of Plesiogulo suggest that this taxon was also such a broad-based, opportunistic feeder. The size and stoutness of the premolar series argues that Plesiogulo could have masticated bone more efficiently than Gulo , but not, perhaps, with the virtuousity of hyaenids and borophagines. Plesiogulo has traditionally been treated as an indicator of forest or woodland paleohabitat (Shotwell, 1958). This interpretation of Plesiogulo is based in part on the common misconception of Gulo as an animal largely restricted to boreal forest. However, Gulo , far from being an obligate forest-dweller, is very abundant in open tundra (Hall and Kelson, 1959; Krott, 1960; Nowak, 1973; and D. Wilson, pers. comm., 1979). Plantigrady is not a common foot morphology for medium to large, eurytopic mammals; hence, in the minds of many researchers, flat feet have rendered Plesiogulo unfit for duty in treeless re¬ gions. While Gulo superficially appears to be plan¬ tigrade, and indeed leaves a rather flat track (Jackson, 1961), it is really digitigrade (Seton, 1929). Both Plesiogulo and Gulo are actually less “flat-footed” than either Ursus or Thalarctos , gen¬ era that frequent treeless habitats. Moreover, all three extant genera are noteworthy wanderers, Gulo traveling up to 40 miles (64 km) in a day. Most of the fossil localities at which Plesiogulo occurs have not produced a sample of significant size, often no more than one individual. Only four localities have produced samples containing four or more individuals (Table 1); these are Optima, Coffee Ranch, Edson, and Wikieup. These faunas include many hypsodont grazers. Coffee Ranch, Edson, and especially, Optima contain large numbers of Dinohippus interpolate. Teleoceras, although hardly a cursorial rhino, is extremely hypsodont. It occurs in great abun¬ dance at Optima, but is almost absent at Coffee Ranch and Edson. Aphelops is more cursorial than Teleoceras , but much more brachydont and is well represented at Coffee Ranch and Edson, but absent at Optima. Texoceras , a hypsodont, curso¬ rial antilocaprid, is present in the hundreds at Optima, but less than half a dozen individuals are known from Edson or Coffee Ranch. Even after allowance is made for the much greater amount of fossil material recovered from Optima as opposed to Coffee Ranch or Edson, the differ¬ ence in relative abundance of the aforementioned taxa is striking. Savage’s (1941:705) description of Optima dur¬ ing the Hemphillian as “an open plain or plains country with most of the forested areas confined to the river bottoms and stream valleys” is appli¬ cable as well to Coffee Ranch and Edson. A slightly greater representation of brachydont ele¬ ments in the latter two faunas argues for slightly more woodland in their respective paleoenviron- ments. None of these localities, however, could be NUMBER 46 3 interpreted as having been well forested, and, indeed, Plesiogulo is most abundant at Optima, the least forested locality. In conclusion, it may be seen that the habitat, foot morphology, and behavior of Gulo do not support the restriction of Plesiogulo to a woodland habitat. The fossil record indicates that Plesiogulo was most abundant at localities interpreted as open plains with very limited woodlands. It seems probable that Plesiogulo was equally at home on the open plains or in the narrow belts of riparian woodland. Geographic Distribution. — Plesiogulo has been recovered from 14 North American localities (Fig¬ ure 1). The majority of these are in the Southwest and southern Great Plains with outliers in Ore¬ gon, California, and Florida. Detailed knowledge of the geographic distribution in North America, as well as the morphology of Plesiogulo , is ham¬ pered by the rarity of this taxon. Only 5 localities out of 14 contain more than a minimum of one individual. Five localities contain only one speci¬ men. The minimum number of individuals for each locality is listed in Table 1 together with the Figure 1.—Distribution of fossil localities known to contain Plesiogulo (1 = Ordnance (Westend Blowout), Oregon, 2 = McKay Reservoir, Oregon, 3 = Modesto Reservoir (Turlock Lake), California, 4 = Pinole Tuff, California, 5 = Wikieup, Arizona, 6 = White Cone, Arizona, 7 = Old Cabin Quarry, Arizona, 8 = Redington Quarry, Arizona, 9 = San Juan Quarry, New Mexico, 10 = Edson Quarry, Kansas, 11 = Lost Quarry, Kansas, 12 = Optima (Guymon), Oklahoma, 13 = Coffee Ranch, Texas, 14 = Bone Valley, Florida). Table 1.—Population densities of Plesiogulo at 14 North American localities (elements used to calculate minimum number of individuals in Column 2 are listed in column 3; R. = right, L. = left)_ Locality Minimum no. of individuals Element Optima (Guymon) 6 R. P 4 , R. Mi Wikieup 4 R. Mi Coffee Ranch 4 L. M 1 Edson 4 R. ramus Ordnance 2 maxilla (Westend Blowout) R. P.\ R. M 1 , R. Mi McKay Reservoir 1 Old Cabin Quarry 1 palate Redington Quarry 1 R. Ci San Juan Quarry 1 R. P 4 -M' Pinole Tuff 1 L. P 4 Modesto Reservoir 1 L. ramus (Turlock Lake) L. M 1 Bone Valley 1 White Cone 1 R. M 1 Lost Quarry 1 R. P 4 most abundant element upon which each respec¬ tive calculation was based. The paucity of fossil material may be due to low population density which may, in turn, be a function of extremely large individual territories. Large territories are characteristic of Gulo , the closest living relative of Plesiogulo. More impor¬ tant, morphological similarities between these two genera suggest that they were functionally similar. The stout premolars, shearing carnassials, and broad molars together with the low, cylin¬ drical condyloid process, deep masseteric fossa, and wide zygomatic arch suggest that both genera were capable of ingesting an impressive array of dietary items, ranging from berries to bones. The postcrania indicate that both genera were unusu¬ ally large in body and long-limbed for mustelids. Such functional similarity lends credence to eco¬ logical comparisons of Plesiogulo and Gulo. Walker (1964:1204) states that a single male wolverine may share a territory of some 300,000 hectares (1158 sq mi) with two or three females. This would result in an average of one adult individual per 75,000 to 100,000 hectares (about 290-385 sq mi). Krott (1959) suggests even larger 4 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY territories for Gulo: up to 770 sq mi (about 200,000 hectares) for males and 150 to 190 sq mi (about 39,000 to 50,000 hectares) for females. Vast indi¬ vidual territories seem a reasonable explanation for the geographically widespread but poorly rep¬ resented species of Plesiogulo in North America. However, the related Asian species, P. crassa (Teil¬ hard de Chardin, 1945) appears to have attained much higher population densities. In excess of 30 skulls and 20 mandibles have been collected from the Paote localities of Shansi Province, People’s Republic of China. They are housed in the Pa¬ leontological Institute at Uppsala, Sweden, and the American Museum of Natural History, New York. Temporal Distribution.— Only 6 of the 14 faunas containing Plesiogulo are associated with radiometric dates. Three dates have been pub¬ lished for the ash overlying the Coffee Ranch Local Fauna. Boellstorff (1976:64) obtained a fission-track date on glass of 5.3 ± 0.4 m.y. Izett (1975:202) obtained both a younger date, also on glass, of 4.7 ± 0.8 m.y. and an older date, on zircon, of 6.6 ± 0.8 m.y. San Juan Quarry occurs 5 meters above the base of the upper tuffaceous zone of the Chamita Formation. MacFadden and Manley (1976) fission-track (zircon) dated an ash overyling the quarry, 5 meters below the top of the upper tuffaceous zone, at 5.6 ± 0.9 m.y. MacFadden, Johnson, and Opdyke (1979:360) also fission-track (zircon) dated the Big Sandy Formation containing the Wikieup Local Fauna at 5.5 ± 0.2 m.y. The White Cone Local Fauna is from the upper member of the Bidahochi For¬ mation, and a basalt from the middle member has been K-Ar dated at 6.60 ± 1.6 m.y. (Scarbor¬ ough, Damon, and Shafiquallah, 1974:472). Vol¬ canic tuffs in the Quiburis Formation, which contains Redington Quarry and Old Cabin Quarry, have been K-Ar dated from 5.21 to 6.25 m.y. (Jacobs, 1977:505). Wood et al. (1941:12) designated Plesiogulo as an index fossil of the Hemphillian Land Mammal Age, which was “based on the Hemphill Member of the Ogallala.” The Hemphill “member,” al¬ though not a distinct lithologic unit (Schultz, 1977:70), contains two faunas: the Higgins Local Fauna from the early Hemphillian and the Coffee Ranch Local Fauna from the late Hemphillian. Of these two, Plesiogulo occurs only in the younger Coffee Ranch Local Fauna, and moreover, has been found elsewhere only in faunas similar to that of Coffee Ranch. The faunas from Edson, Optima, McKay, Ordnance, Lost Quarry, Mo¬ desto Reservoir, Wikieup, Redington Quarry, and Old Cabin Quarry are all late Hemphillian in age, while those from San Juan Quarry and Upper Bone Valley are latest Hemphillian. Slightly older than Coffee Ranch and its correl¬ atives, but still late Hemphillian in age, is the White Cone Local Fauna. The assessment of these faunas as late Hemphillian is corroborated by the presence of other immigrant genera, among them Agnotherium and Machairodus. Plesiogulo is known from Old World faunas of Turolian (late Miocene) age. It is probable that sometime between 7.0 and 6.5 million years ago that this species migrated to the New World and rapidly extended its range across Canada and the United States. Glacial activity during the Pleis¬ tocene subsequently destroyed much of the more northerly late Miocene deposits; thus, remains of Plesiogulo are restricted largely to the southern half of the United States. The first appearance of Plesiogulo has come to be one of the criteria deter¬ mining the late Hemphillian (R. H. Tedford, pers. comm., 1979). Abbreviations.— The following abbreviations are used in this study: AMNH Department of Vertebrate Paleontology, Amer¬ ican Museum of Natural History, New York, New York F:AM Frick Collection, American Museum of Natural History, New York, New York KUVP University of Kansas Museum of Natural His¬ tory, Lawrence, Kansas LACM Los Angeles County Museum of Natural His¬ tory, Los Angeles, California MU Midwestern University, Wichita Falls, Texas TMM Texas Memorial Museum, University of Texas, Austin, Texas UCMP University of California Museum of Paleontol¬ ogy, Berkeley, California NUMBER 46 5 UF Florida State Museum, University of Florida, Gainesville, Florida UMMP University of Michigan Museum of Paleontol¬ ogy, Ann Arbor, Michigan UOMNH University of Oregon Museum of Natural His¬ tory, Eugene, Oregon USNM Former United States National Museum, collec¬ tions now in National Museum of Natural History, Smithsonian Institution, Washing¬ ton, D.C. Measurements.— Width of P 4 was measured across the protocone. Two anteroposterior dimen¬ sions of M 1 were measured: the lingual length refers to the maximum anteroposterior distance across the expanded inner lobe, and the median length refers to the minimum anteroposterior dis¬ tance across the constriction at midcrown. Stan¬ dard deviation was calculated only for samples of four or more. Acknowledgments.— I would like to thank Earl E. Manning and Henry Galiano, American Museum of Natural History, for their helpful comments and creative criticism. The following people did much to facilitate the loan of fossil specimens or comparative material: Walter W. Dalquest, Midwestern University; Eric Gustav- sen, University of Oregon Museum of Natural History; Everett H. Lindsay, University of Ari¬ zona Laboratory of Paleontology; S. David Webb, Florida State Museum; Richard H. Ted- ford, American Museum of Natural History; J. Howard Hutchison, University of California Mu¬ seum of Paleontology; and James E. Martin, South Dakota School of Mines and Technology. Additional material was borrowed from the Uni¬ versity of Kansas Museum of'Natural History and the Texas Memorial Museum. John Lance, National Science Foundation, and Hugh Wag¬ ner, University of California Riverside, provided elusive bits of information concerning particular specimens and localities. Donald E. Wilson, U. S. Department of Fish and Wildlife, generously al¬ lowed me access to unpublished data on Gulo. George Krochak, Registrar, American Museum of Natural History, was particularly adept at the art of the invoice. Chip Clark, Office of Exhibits, Smithsonian Institution, was a great source of photographic advice. Childs Frick initially iden¬ tified most of the Plesiogulo material in his collec¬ tion. I would also like to thank the many mem¬ bers, past and present, of the American Museum of Natural History Department of Vertebrate Paleontology and of the Frick Laboratory, who have contributed to the identification of the larg¬ est sample of Plesiogulo in the Western Hemi¬ sphere. Such well curated collections greatly en¬ hance the productivity and pleasure of visiting researchers. This paper is an outgrowth of re¬ search undertaken in partial fulfillment of the doctorate degree at the University of Kansas. The preliminary research was supported in part by National Science Foundation Grant DEB77- 15869. The manuscript was critically read and considerably improved by Earl E. Manning, Henry Galiano, Robert J. Emry, and Gary Mor¬ gan. Systematics The following hierarchic classification is used in this report: Order Carnivora Bowdich, 1821 Family Mustelidae Swainson, 1835 Subfamily Mellivorinae Gill, 1872 Tribe Gulonini Webb, 1969 Genus Plesiogulo Zdansky, 1924 Plesiogulo marshalli (Martin, 1928) Figures 2, 3, 7-12 Brachypsalis marshalli Martin, 1928:233, pi. 20. Plesiogulo marshalli. —Hibbard, 1934:247. Holotype.— KUVP 3464, right ramus with P 34 , Mi, and alveoli of I 2 -P 2 and M 2 and skull fragments bearing glenoid fossae (Figure 2 a-c). Type Locality.— Edson Quarry, NW 1/4 SE 1/4 Sec. 25, T10S, R38W, Sherman County, Kansas. Referred Specimens. —From the type locality: KUVP 3465, right P 4 -M 4 ; KUVP 3467, right juvenile ramus with dP 3 _ 4 , M 1 - 2 ; KUVP 3606, right juvenile ramus with Ci, dP 4 , Mi; F:AM 49479, associated right and left juvenile rami with SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Figure 2. — Plesiogulo marshalti: a-c , KUVP 3464, holotype, right ramus, lateral, medial, and occlusal views; o', KUVP 3465, right P 4 -M', occlusal view. (Specimens from Edson Quarry; X 1.) right C 1 -P 2 , dP 4 , P 4 (germ), Mi and left C 1 -P 2 , dP 3 ; F:AM 104724, left Ci; F:AM 67650A, partial right humerus. From Optima (Guymon), Texas County, Okla¬ homa: F:AM 49497, crushed skull with both rami and partial right radius; F:AM 49490, left maxilla with I 2 ’ 3 , P 3 -M 1 ; F:AM 49492, right P 4 ; F:AM 49491, right ramus with Ci-Mp F:AM 49494, right ramus with P 3 -M 2 ; F:AM 49493, right ra¬ mus with P 3 -M 1 ; F:AM 49499, right ramal frag¬ ment with P 2 - 4 ; F:AM 49495, right ramal frag¬ ment with P4-M1; F:AM 49498, left ramal frag¬ ment with P4-M1; F:AM 49496, right ramal frag¬ ment with Mi; F:AM 67918N, partial left radius. From Coffee Ranch, Hemphill County, Texas: TMM 41261-12, left M 1 ; MU 11523, right M 1 ; MU 5132, partial right Mg UCMP 31944, left C 1 ; UCMP 31942, right P 4 ; UCMP 31833, right ramal fragment with P 4 -Mp UCMP 30180, right Mi; F:AM 108062, crushed skull; F:AM 23386, left maxilla with l 1 ' 2 , C 1 , P 2 -M 4 ; F:AM 23387, right maxilla with P 2 -M ! ; F:AM 23378, right NUMBER 46 7 Figure 3. — Plesiogulo marshalli : a , b, F:AM 49479, right and left juvenile rami, occlusal and lateral views, teeth in “6” are tip of Ci, P2, roots of dP.3, dP4 atop P4, and Mi; c, KUVP 3467, right juvenile ramus, medial view, teeth are dP.3-4, Mi_ 2 . (Specimens from Edson Quarry; X 1.) partial maxilla with P 3 -M x ; F:AM 23379, left partial maxilla with P 3 -M x ; F:AM 23378A, left P 4 ; F:AM 23378B, left M 1 ; F:AM 22270, right ramal fragment with P 2 - 4 ; F:AM 23388, right ramal fragment with P 3 -M 1 ; F:AM 108052, pos¬ sibly associated left humerus and ulna. From Modesto Reservoir, Stanislaus County, California: LACM 61696, left ramus with C 1 -M 1 . From McKay Reservoir, Umatilla County, Or¬ egon: UOMNH F-3656, right maxilla fragment with P 4 -M l ; UOMNH F-2441, right partial Mi. From Ordance (Westend Blowout), Umatilla County, Oregon: two partial maxillae and an isolated Mi. From San Juan Quarry, Rio Arriba County, New Mexico: F:AM 49230, right maxilla frag¬ ment with P 4 -M 1 . From Upper Bone Valley, Polk County, Flor¬ ida: UF 19253, right maxilla fragment with P 4 - M 1 ; UF 19295, left M 1 . From White Cone, Navajo County, Arizona: USNM 244492, right M 1 . From Lost Quarry, Wallace County, Kansas: KUVP 12433, right P 4 Revised Diagnosis.— Martin (1928) did not diagnose his new taxon, Brachypsalis marshalli , nor did Hibbard (1934) discuss his own rationale for transferring marshalli to Plesiogulo. However, such characters in the type specimen as a relatively large Mi with a broadened talonid, lack of a parastyle on P 4 , absence of M 2 , and presence of an auditory meatal spout clearly remove the spe¬ cies from Brachypsalis (Figure 2). The following diagnosis is offered for Plesiogulo marshalli : A North American species far larger than the Old World genotypic species, P. brachygnathus, and generally exceeding in size the Old World species, P. crassa; Plesiogulo marshalli differs from both Old World species in having a heavier, more massive dentition with a greater ratio of width to length for individual teeth, a larger protocone on P 4 , and more pronounced cingula. Description. —The following section pertains to undescribed material from Edson (for addi- 8 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY tional data on the type ramus and P 4 -M\ see Martin, 1928). Martin maintained that all of the Plesiogulo specimens that he described from Edson probably belonged to the same individual. Cor¬ respondence in size and occlusal surfaces supports the association of the maxilla fragment bearing P 4 -M* (KUVP 3465) with the type ramus; how¬ ever, the smooth, noncrenulated enamel of the right C 1 (KUVP 3469) precludes its referral to Plesiogulo. This specimen should instead be re¬ ferred to Osteoborus cyonoides , a borophagine canid, also present in the Edson Local Fauna. Based on the broken alveolus in the type ramus, Martin described Ci of Plesiogulo to be of great size and out of proportion to the rest of the teeth. However, the alveolus is large in order to accom¬ modate the posteriorly expanded area at the base of the crown. The canine itself is formidable, but not grossly disproportionate for a carnivore of this size. It falls far short of the astonishing canine development observed in many of the Felidae. As Martin (1928:234) stated, the Mi talonid is in¬ deed well developed; however, its antero-internal surface is not “deeply basined,” but only slightly concave and inclined lingually. In unworn speci¬ mens the Mi talonid is almost flat, although still inclined. Three lower jaws (KUVP 3467 and 3606 and F:AM 49479) from Edson represent the only known deciduous dentitions of Plesiogulo from either Old or New World localities (Figure 3). These specimens considerably enhance our knowledge of the genus. Each of the three jaws has an alveolus indicating that a very small, single-rooted tooth was closely appressed and sit¬ uated slightly lingual to the anterior end of dP 2 in juveniles and P 2 in adults. There is no indica¬ tion of a tooth germ beneath any of these alveoli. Slaughter, Pine, and Pine (1974:115) state that only a single tooth ever develops in the premolar one position except in Tapirus and possibly certain hyraxes. Hence, the tooth associated with the aforementioned alveolus is herein referred to as dPi. The condition of this tooth in the type of P. marshalli is not known because the appropriate portion of the ramus is missing; however, both dPi and dP 1 alveoli are present in all of the referred specimens in which this portion of the ramus or maxilla is preserved. Several specimens of P. crassa from the Paote localities contain dPi in situ. Ci, P 2 , and M 2 are also missing from the type but just emerging in juveniles F:AM 49479 and KUVP 3467. Ci is large, although, as stated previously, not unusually so. The enamel is cren- ulated and a ridge extends from the basal cin¬ gulum to the tip of the crown on the antero- internal surface. P 2 is oval and double-rooted with a single blunt cusp situated towards the anterior end of the tooth. A low crest connects this cusp with the anterior and posterior borders of the crown. M 2 is a small, rounded button placed well up on the ascending ramus. A low transverse crest connects the two small lingual and labial cingular cusps. The lower cheekteeth erupt in the following order: dPi, dP 2 , dP 3 , dP 4 , Mi, M 2 , P 2 , P3, P 4 . In the Edson juveniles, DP 2 is not preserved, but alveoli indicate a double-rooted tooth smaller than dP 3 . DP 3 is a slender tooth with a high central cusp, a small accessory cusp on the antero- internal cingulum, and a posterior cingular shelf. Predictably, DP 4 foreshadows the morphology of Mi; however, the paraconid-protoconid notch is more open, the metaconid is larger, and the tal¬ onid is shorter and narrower than in Mi. The enamel of the deciduous dentition is not so heav¬ ily crenulated as that of the permanent dentition. The three juveniles from Edson are very close in age. All have dP 4 in place and slightly worn with Mi just emerging and unworn. In KUVP 3467 and 3606 the tip of P 2 is just visible through the alveoli of dP 2 . In the third ramus, F:AM 49479, the thin bone has been chipped away to expose the unworn P 2 . A slightly worn dP 3 is present in both KUVP 3467 and F:AM 49479. The canine is just emerging in F:AM 49479 and KUVP 3606, and in the latter specimen the terminus of the root of dCi remains closely ap¬ pressed to Ci. It seems likely that these three juveniles were litter mates; moreover, there is no evidence to NUMBER 46 9 cm WIDTH Figure 4. — Scatter diagram of P 4 ’s of Plesiogulo from North America. cm Figure 6.— Scatter diagram of M u s of Plesiogulo from North America. WIDTH Figure 5. — Scatter diagram of Mi’s of Plesiogulo from North America. SYMBOLS FOR FIGURES 4.5. & 6 ♦ Edson ■ Bone Valley * Optima A San Juan Quarry o Coffee Ranch ▲ Lost Quarry □ MCKay Reservoir • Wikieup OD Old Cabin Quarry 0 Modesto Reservoir 10 Table 2.— Measurements (cm) of Plesiogulo marshalli from Edson Quarry (O.R. = observed range, X = sample mean) Element No. O.R. X p4 length 1 1.94 width at protocone 1 1.35 M 1 length, lingual 1 1.44 length, median 1 0.99 width 1 1.85 P 2 length 1 0.73 width 1 0.57 Pa length 1 1.10 width 1 0.76 P 4 length 1 1.36 width 1 0.85 Mi length 3 2.36-2.41 2.39 width at talonid 3 0.83-0.96 0.90 m 2 length 1 0.66 width 1 0.60 Ps-Mi length 1 4.77 Condylar width 1 3.29 length 2 0.83-0.90 dP 3 width 2 0.45-0.50 length 3 1.42-1.43 1.43 dP 4 width 3 0.60-0.67 0.63 indicate that the only adult Plesiogulo at Edson was not a female. The young of Gulo, usually numbering two or three, remain with their mother for about two years (Walker, 1964). A female with her litter of cubs seems the most logical explanation of the unusual sample of Ple¬ siogulo found in the Edson Local Fauna (Table 2). Discussion.— Specimens of Plesiogulo marshalli from nine localities are discussed in relation to the holotypic and referred material of P. marshalli from Edson. The problems inherent in comparing small samples and, in many cases, isolated teeth are all too familiar. Too often is individual vari¬ ation interpretated as variation at the specific level. Fortunately, the very large sample of P. crassa from Asia (Zdansky, 1924; Kurten, 1970) provides considerable insight into the amount of variation present in a related species. Scatter diagrams (Figures 4-6) illustrate the size variation in the three most generically dis¬ tinct teeth of Plesiogulo: P 4 , M 1 , and Mi. Fortu¬ nately, these teeth were also among the most numerous in the total North American sample (P 4 = 11, M 1 = 16, Mi = 17). The upper and SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Table 3. —Measurements (cm) of Plesiogulo marshalli from Optima (O.R. = observed range, X = sample mean, s.d. = standard deviation) _ Element No. O.R. X s.d. I 3 length 1 0.91 width 1 0.84 P 3 length 1 1.16 width 1 0.80 P 4 length 2 2.07-2.11 width 1 1.43 M 1 length, 1 1.59 lingual length. 1 0.83 median width 1 1.80 c, length 1 1.52 width 1 1.05 P 2 length 3 0.72-0.77 0.74 width 3 0.51-0.55 0.53 Pa length 5 0.91-1.01 0.95 0.05 width 5 0.61-0.72 0.65 0.04 P 4 length 7 1.23-1.44 1.31 0.08 width 7 0.73-0.87 0.79 0.05 Mi length 6 2.21-2.48 2.40 0.10 width 7 0.82-0.90 0.86 0.03 at talonid m 2 length 2 0.72-0.77 width 2 0.62-0.65 Ci-M, length 1 7.04 Radius length 1 12.32 width, 1 1.79 proximal lower carnassials (Figures 4, 5) proved to be the most diagnostic at the specific level. M 1 , espe¬ cially the length of the lingual lobe, was the most variable of all the cheek teeth (Figure 6). The sample from Optima contains only one partial maxilla in addition to a badly crushed partial skull; the remaining specimens are lower dentitions (Figure 7, Table 3). There are no upper incisors or P 3 from the type locality to compare with those from Optima; however, these teeth, and the rest of the Optima sample do compare well with dentitions from Coffee Ranch. The P 4 from Optima is slightly larger with the anterior notch slightly deeper than the P 4 from Edson. In the Optima M 1 the inner lobe is slightly longer than in the Edson M 1 . None of the several partial rami and ramal fragments from Optima differ Figure 7 .—Plesiogulo marshalli: a , b, F:AM 49490, left maxilla, lateral and occlusal views; c-e, F:AM 49491, right ramus, medial, occlusal, and lateral views. (Specimens from Optima; X 1.) 12 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Figure 8 .—Plesiogulo marshalli, F:AM 108602, partial skull: a, lateral view; b, occlusal view. (Specimen from Coffee Ranch; X 1.) significantly from the type ramus. The Ci from Optima is slightly smaller than the isolated Ci from Edson. The P 3-4 and Mi from Optima com¬ pare well with those from Edson. The Optima sample contains the only other M 2 ’s (F:AM 49494 and 49497) from a New World locality; they are slightly larger than the M 2 from Edson. The Coffee Ranch sample bears an even closer resemblance to the type material than that of Optima. The ratio of upper to lower dentitions is the reverse of the Optima sample; Coffee Ranch contains nine partial upper dentitions, but only two lowers. The partial skull, F:AM 108062 (Fig¬ ure 8 , Table 4) from Coffee Ranch is far from complete, although not so badly damaged as the one from Optima. These two specimens, in ad¬ dition to the fragments from Edson, constitute the only cranial material known for P. marshalli. The skull is least damaged on the left side; the remaining teeth are worn left P 3-4 . The left in¬ fraorbital canal is intact; it is larger and more rounded than in Gulo and situated above the NUMBER 46 13 Table 4.— Measurements (cm) of Plesiogulo marshalli from Coffee Ranch (O.R. = observed range, X = sample mean, s.d. = standard deviation) Element No. O.R. X s.d. C 1 length 2 1.22-1.35 width 1 0.97 p2 length 2 0.80-0.82 width 2 0.57-0.59 P 3 length 4 1.11-1.29 1.17 0.08 width 4 0.66-0.80 0.72 0.06 p4 J length 5 1.85-2.17 1.95 0.13 width 4 1.21-1.50 1.32 0.14 M 1 length, 7 1.30-1.63 1.47 0.16 lingual length, 7 0.84-1.03 0.92 0.08 median width 7 1.67-1.99 1.81 0.14 C'-M 1 length 1 6.54 P.3 length 2 1.03-1.04 width 2 0.73-0.75 P 4 length 4 1.28-1.39 1.34 0.05 width 4 0.76-0.88 0.81 0.05 M, length 3 2.35-2.54 2.42 0.10 width 3 0.95-1.02 0.98 0.04 at talonid Hum- length 1 15.11 erus width, 1 4.02 distal Ulna length 1 14.96 antero- 1 2.53 posterior at coro- noid middle of P 4 rather than the front of P 4 as in Gulo. Both postorbital processes are present and a low temporal crest from each extends posteriorly to form the sagittal crest. All of the muzzle anterior to P 3 is missing, as well as most of the occipital region. Part of the left parietal, mastoid process, and glenoid fossa are still present. As in Gulo , the optic foramen is situated 1 to 2 cm anterior to the closely appressed orbital fissure and foramen ovale. The foramen rotundum is located medial to and on a level with the glenoid fossa. The common aperature of the sphenopalatine fora¬ men and the caudal end of the posterior palatine canal lies, as in Gulo , in the medial wall of the orbit at a level with the posterior end of P 4 . The P 4, s from Coffee Ranch bracket the Edson Table 5. —Measurements (cm) of Plesiogulo marshalli from several localities (O.R. = observed range)_ Element No. O.R. MODESTO RESERVOIR Ci length 1 1.53 width 1 1.18 P 2 length 1 0.78 width 1 0.59 P 3 length 1 1.05 width 1 0.68 P 4 length 1 1.48 width 1 0.90 Mi length 1 2.51 width at talonid 1 0.94 C 1 -M 1 length 1 7.98 MCKAY RESERVOIR P 4 length 1 1.97 width 1 1.27 M 1 length, lingual 1 1.39 length, median 1 0.88 width 1 1.70 BONE VALLEY P 4 length 1 2.18 width 1 1.42 M 1 length, lingual 2 1.21-1.25 length, median 2 0.87-0.98 width 2 1.75-1.76 SAN JUAN QUARRY P 4 length 1 1.85 width 1 1.24 M 1 length, lingual 1 1.32 length, median 1 0.93 width 1 1.62 LOST QUARRY P 4 length 1 1.82 width 1 1.24 WHITE CONE M 1 length, median 1 0.83 P 4 in size, and one (F:AM 23386) is almost iden¬ tical (Figure 9d, e). The Coffee Ranch M x ’s show a wide range of variation, some approaching in size those of P. lindsayi, new species, and some almost as small as P. marshalli from Bone Valley. The lower premolars compare well with those from Edson. The single Mi is very close in length to the Edson Mi’s, although very slightly wider across the talonid (Figure 9 a-c). The single specimen of P. marshalli from Mo¬ desto Reservoir (LACM 61696) is a left ramus 14 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Figure 9 .—Plesiogulo marshalli: u-c, F:AM 23388, right ramus, lateral, medial, and occlusal views; d, e, F:AM 23386, left maxilla, lateral and occlusal views. (Specimens from Coffee Ranch; X 1.) bearing Ci, P 2 - 4 , and Mi (Figure 10^-i). The cheek teeth exhibit very little wear and indicate a mature individual somewhat younger than the type specimen. Ci and P 2 compare well with those in an Edson juvenile (F:AM 49479) (Figure 3a, b ). Although broken, Ci still exhibits part of the antero-internal ridge as well as crenulated enamel. The enamel of the cheek teeth is not so strongly crenulated as that of Ci. An alveolus for dP 2 separates Ci and P 2 . The longitudinal crests of P 2-4 are more pronounced than in the Edson specimens, and P 3 -M 1 resemble very closely those of the type ramus. Two very small depressions indicate the position of the M 2 alveolus, which had commenced closure sometime during the life of this individual. The right maxilla fragment bearing P 4 -M : and a possibly associated partial Mi of P. marshalli from McKay Reservoir (Figure 10a, Table 5) are very close in size and morphology to those from Edson (Table 2). There is a high degree of cor¬ respondence even in such features as the internal cingulum and the two anterior ridges on P 4 and the occlusal outline of M 1 , an extremely variable tooth. Two undescribed partial maxillae from nearby Ordnance are probably referable to P. marshalli (J. E. Martin, pers. comm.). An isolated right M 1 is the only specimen of Plesiogulo from White Cone (Figure 106, Table 5). The external end of the tooth, including most of the paracone and metacone, is missing, as are fragments of enamel all along the posterior edge of the crown. The tooth compares well with the Edson M 1 in size, placement of the paracone, metacone, and protocone, and extent of the me¬ dian constriction. The hypocone is not quite so widely expanded as in the Edson M 1 , and the internal border is more strongly concave. How- 15 Figure 10. — Plesiogulo marshalli: a , UOMNH F-3656, right P 4 -M\ occlusal view, specimen from McKay Reservoir; b, USNM 244492, right M 1 , occlusal view, specimen from White Cone; c, F:AM 49230, right P 4 -M\ occlusal view, specimen from San Juan Quarry; d , KUVP 12433, right P 4 , occlusal view, specimen from Lost Quarry; e, UF 19295, left M 1 , occlusal view, specimen from Bone Valley; f UF 19253, right P 4 -M\ occlusal view, specimen from Bone Valley; g-i , LACM 61696, cast of left ramus, occlusal, medial, and lateral views, specimen from Modesto Reservoir. (X 1.) ever, since these discrepancies are most probably isolated left M 1 have been recovered from Bone due to individual variation, I have herein referred Valley. This P 4 is one of the largest in the total the White Cone specimen to P. marshalli. sample; the M x, s, however, are the smallest such Three localities have produced a few isolated teeth. The two Bone Valley M x, s compare well teeth of P. marshalli, which are somewhat smaller with the San Juan Quarry M 1 ; moreover, three and more slender than material from the preced- M l5 s from Coffee Ranch (not illustrated) closely ing localities (Figure 10, Table 5). The sole spec- approach these teeth in size. There is more than imen from Lost Quarry is the smallest P 4 in the sufficient overlap in both size and morphology total P. marshalli sample. The San Juan Quarry between the material from San Juan Quarry, material consists of a right maxilla fragment bear- Bone Valley, and Lost Quarry and the material ing P 4 -M x . This P 4 is almost as small as that from from Edson, Optima, and Coffee Ranch to war- Lost Quarry and, like it, bears a slightly reduced rant its referral to P. marshalli. protocone. An associated right P 4 -M x and an To date, no postcranial elements of Plesiogulo zso8 o f i*i Figure 11 .—Plesiogulo marshalli : a-c, F:AM 67650A, right humerus, anterior, lateral, and posterior views, specimen from Edson Quarry; d, e, F:AM 108052, left ulna, lateral and medial views, specimen from Coffee Ranch; f g, F:AM 108052, left humerus, anterior and posterior views, specimen from Coffee Ranch. (X 1.) NUMBER 46 17 Figure 12 .—Plesiogulo marshalli: a, b, F:AM 49497, right radius, anterior and posterior views; c, d, F:AM 67918N, left radius, anterior and posterior views. (Specimens from Op¬ tima; X 1.) have been described in the New World literature; however, Hendey (1978) briefly described some postcrania from Langebaanweg, South Africa, and referred them to P. monspessulanus Viret. I have herein referred five specimens to P. marshalli, a partial left humerus from Edson, a possibly associated left humerus and ulna from Coffee Ranch, and two partial radii from Optima. These elements were not found in direct association with dentitions, but their size is compatible with esti¬ mates of body size in Plesiogulo and in many characters their morphology is suggestive of Gulo. Moreover, these elements clearly do not belong to any other genera in their respective faunas. Osteoborus cyonoides is the only carnivore present in these faunas with postcrania similar in size to those of Plesiogulo , although most adult Osteoborus skeletal elements are somewhat larger. The Edson humerus is complete distal to the tip of the deltopectoral rugosity, except for frag¬ ments from the lateral and medial epicondyles (Figure 11 a-c). The Coffee Ranch humerus is complete but for a fragment of the medial epi- condyle (Figure 11 f, g). The two humeri are very close in size, and both bear a large entepicondylar (supracondyloid) foramen, which is not present in Osteoborus. The humeral shaft, especially the proximal half, is straighter in Plesiogulo than in Osteoborus. The medial epicondyle in Plesiogulo is much smaller than in Gulo or Osteoborus, resulting in a proportionately narrower distal end. The olecranon fossa is larger and more deeply exca¬ vated in Plesiogulo than in Gulo. The only ulna is from Coffee Ranch and is missing only a few small chips of bone (Figure 1 Id, e). Its shaft is not so long and slender as in Osteoborus, but considerably heavier than in Gulo. The olecranon process is longer and more concave medially in Plesiogulo than in Gulo and bears an elongate groove in its top. The roughened area for the attachment of the interosseous ligament is much larger in Plesiogulo than in Gulo and extends farther up the shaft than in Osteoborus. The styloid process of the ulna is heavier and more rounded in Plesiogulo than in Osteoborus. This structure is also heavy in Gulo, but tends to be flattened distally rather than rounded as in Plesiogulo. The crest on the posteromedial face of the distal end is much sharper in Gulo than in Plesiogulo. Two radii, one right and one left, are present in the Optima sample (Figure 12). The right radius is missing half of the proximal articular surface, and its distal end is badly crushed. The left radius is missing only the distal articular surface. Radii of adult Osteoborus are almost al¬ ways longer and heavier than those of Plesiogulo. The outline of the proximal articular surface is pointed in Plesiogulo as opposed to rounded in Osteoborus. The bicipital (radial) tuberosity for the 18 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY insertion of m. biceps is a raised, rough area bordered by a curving lateral crest in Plesiogulo', an additional medial crest is present in Gulo. In Osteoborus this feature is a large, knob-like process. Plesiogulo has a deep groove on the lateral face of the distal end. This groove is shallow in Osteoborus and very shallow in Gulo. The crest forming the posterior border of this groove is sharper in Ple¬ siogulo than in Osteoborus. The limbs of Plesiogulo are generally compara¬ ble to those of Gulo. The ratio of the lengths of proximal to distal elements is approximately the same and does not suggest a marked difference in degree of cursoriality. Both wolverines possess distal elements that are far shorter than those of Osteoborus , which conforms to the long-legged canid norm. When compared to corresponding bones of compariable size in Gulo , the greater size and rugosity of muscle attachments on the radius and ulna of Plesiogulo suggest that the latter genus had the more powerful antebrachium. Plesiogulo lindsayi, new species Figures 13-16 Holotype.— F:AM 49384, muzzle bearing right I^M 1 and left I 2-3 , P 2-3 , and broken P 4 -M 1 (Figure 13). Type Locality.— Wikieup area, Mojave County, Arizona; Big Sandy Formation; late Hemphillian Land Mammal Age. Referred Specimens.— From the type locality: F:AM 49386, crushed palate with right I 2 -C\ P 2-3 and left P 2 " 4 ; F:AM 49385, crushed palate with right I 3 -C\ P 2 -M x and left P 2 -M : ; F:AM 108053, right C 1 ; F:AM 108049, right P 4 ; F:AM 49391, right P 3 ; F:AM 49374, left M 1 ; F:AM 49392A, right M 1 ; F:AM 49392B, left M 1 ; F:AM 108054, partial left M 1 ; F:AM 108055, right Ci; F:AM 108056, left Ci; F:AM 108057, right C x ; F:AM 49387, left ramus with P 2 -M 1 ; F:AM 49369, right ramus with P2-M1; F:AM 49389C, left P2-4; F: AM 49370, left P4-M1; F:AM 49388, right P4-M1; F:AM 49390, right P4-M1; F:AM 108048, right P 4 ; F:AM 49389D, left P 4 ; F:AM 49373, right Mi; F:AM 49389A, partial left Mi; F:AM 49389B, left Mi; F:AM 67957, right ulna and left ischium with partial acetabulum; F:AM 67958, right radius; F:AM 67958A, right radius; F:AM 67958B, partial left radius; F:AM 105372, partial right ulna. From Old Cabin Quarry, Pima County, Ari¬ zona: F:AM 50690 , crushed palate with right P 2 - M 1 and left P 3 -M\ right ramus with I3-C1, P2-M1, and left ramus with I3-C1, P3-M1. From Redington Quarry, Pima County, Ari¬ zona: F:AM 108058, right C 1 ; F:AM 108059, left Ci; F:AM 108060, left ulna. From Pinole Tuff, Contra Costa County, Cal¬ ifornia: UCMP 57522, left P 4 . Etymology.— The species is named for Everett H. Lindsay. Diagnosis. —Plesiogulo lindsayi compares to Old World species as follows: larger than P. brachy- gnathus, P. minor, P. crassa, and P. praecocidens, smaller than P. major, and about the same size as P. monspessulanus. Plesiogulo lindsayi differs from P. monspessulanus in the presence of a strong meta- conid on Mi. P. lindsayi is larger than the only other New World species, P. marshalli , and differs from it in the following: I 1-2 not as reduced; I 3 and C 1 relatively larger; superior premolars larger in relation to M and more crowded; P bears a lingual bulge, which is small or absent in P. marshalli ; P 4 proportionately larger with more posteriorly placed protocone and a weak labial cingulum not present in P. marshalli ; lingual lobe of M 1 not as expanded and with a more angular posterior corner; greater ratio of width to length in M 1 ; protocone of M 1 stronger and more cen¬ trally placed; ramus deeper and more massive; masseteric fossa more deeply excavated and ex¬ tending beneath the Mi talonid as opposed to beneath Mi posterior border in P. marshalli ; infe¬ rior premolars larger in relation to Mi; stronger lingual bulge on P 4 ; greater ratio of width to length in Mi; talonid broader with a higher hypoconid and mesoconid; metaconid not as tightly appressed to protoconid; M 2 alveolus smaller and not as high on the ascending ramus. Description.— The type specimen, collected by Robert J. Emry, consists of the anterior portion Figure 13 .—Plesiogulo lindsayi, new species, holotype, F:AM 49384, partial skull: a, lateral view; b, frontal view; c occlusal view; specimen from Wikieup. (X 1.) of a skull; everything posterior to M 1 is missing (Figure 13, Table 6). The nasals and most of the premaxilla are crushed, although the ventral bor¬ der of the external nares, both infraorbital foram¬ ina, and both anterior roots of the zygomatic arches are intact. The section of the premaxilla bearing I 1 " 3 is undamaged. The palate is com¬ plete, and the only teeth missing are left I 1 , left C 1 , both P^s, and portions of left P 4 -M\ The premaxilla is robust to accommodate the large incisors. The incisive foramina (palatine Fissures) are elongate and separated by a narrow bridge of bone. The large, round infraorbital foramen is dorsal to the anterior end of P 4 The 20 Table 6.— Measurements (cm) of Plesiogulo lindsayi, new species, from Wikieup (O.R. = observed range, X = sample mean, s.d. = standard deviation) Element No. O.R. X s.d. I 1 length 1 0.85 width 1 0.46 I 2 length 1 0.92 width 1 0.53 I 3 length 2 1.16-1.24 width 2 0.78-1.07 c 1 length 1 1.75 width 1 1.39 p2 length 3 0.93-0.97 0.95 width 3 0.66-0.70 0.69 p3 length 4 1.21-1.39 1.30 0.08 width 4 0.85-1.02 0.93 0.07 p4 length 1 2.35 width 1 1.73 M 1 length, 3 1.33-1.55 1.46 lingual length, 4 0.94-1.02 0.97 0.15 median width 3 2.00-2.15 2.06 C^M 1 length 1 7.82 p 2 length 3 0.79-.81 0.80 width 3 0.64-.65 0.65 P.3 length 3 1.04-1.13 1.09 width 3 0.70-0.77 0.74 P 4 length 6 1.40-1.65 1.51 0.08 width 6 0.87-1.01 0.93 0.05 Mi length 7 2.49-2.92 2.71 0.13 width 6 0.98-1.18 1.06 0.07 at talonid Ulna antero- 2 2.47-2.56 posterior at coro- noid Radius width, 2 1.73-1.75 proximal antero- 3 1.00-1.13 1.08 posterior, proximal first and second incisors are subequal in size, but both are greatly exceeded by I 3 , which is very large and caniniform. The incisiors are closely appressed, but still in alignment; their anterior surfaces form a shallow arc, rather than a straight line as in Gulo. A very short diastema separates I from C 1 . The canine is quite impressive, and its enamel is heavily crenulated; two ridges extend SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY from the base of the crown to the tip, one on the antero-internal surface and one on the posterior surface. Alveoli indicate a small, single-rooted dP 1 . P 2 is double-rooted with a cingular shelf surrounding the blunt central cusp. The lingual root of P is large, distinct from the posterior root, and pro¬ duces a strong bulge in the lingual border of the crown. The tooth is surrounded by a cingulum, which lengthens into a shelf posteriorly. The up¬ per carnassial (P 4 ) is very large. A ridge extends from the basal cingulum to the tip of the paracone on the anterior surface. A second ridge connecting the large protocone with the paracone is less distinct. A weak cingulum is present on the labial surface and a stronger one on the lingual surface. The width of M 1 exceeds its length, and the tooth exhibits the characteristic median constriction and expanded inner lobe. The paracone is larger and higher than the metacone, and both cusps are situated well in from the labial border. The lingual lobe of the tooth is long with an angular rather than a rounded posterior corner. The pro¬ tocone is low and crescentic. Twenty-three dentitions and isolated teeth con¬ stitute the hypodigm of P. lindsayi (Figure 14, Table 6). The upper dentitions closely resemble that of the type. The most complete ramus (F:AM 49382) is missing most of the coronoid process, about half of the condyle, and everything anterior to the P3 alveolus. The body of the ramus is deep and heavy, and the deep masseteric fossa extends beneath the talonid of Mi. Neither dPi nor any of the lower incisors have been recovered. The lower canine, like the upper, is very large with heavily crenulated enamel. P2-4 are double- rooted, single-cusped teeth. P 3 and P 4 each have prominent anterior and posterior ridges extend¬ ing from the basal cingulum to the tip of the single central cusp. P 4 , much like P 3 , has a strong bulge on the lingual side of the crown; the pos¬ terior root is quite broad. The lower carnassial is broad for its length with a strong metaconid. The talonid is wide and its flat, non-basined surface dips lingually. The only two discernible talonid cusps are the hypoconid and mesoconid. A small, NUMBER 46 Figure 14 —Plesiogulo lindsayi, new species: a-c, F:AM 49388, right ramus, lateral, medial, and occlusal views; d,e, F:AM 49370, left P 4 -M\ occlusal and medial views ;/g, F:AM 49387, partial left ramus, occlusal and lateral views. (Specimens from Wikieup; X 1.) 22 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Figure 15 —Plesiogulo lindsayi, new species, F:AM 50690: a , palate, occlusal view; b,c rami, lateral and occlusal views. (Specimens from Old Cabin Quarry; X 1.) single-rooted M 2 is situated at the base of the ascending ramus. The single specimen of Plesiogulo from Old Cabin Quarry is an associated palate and jaws, which are extensively crushed and deformed (Fig¬ ure 15, Table 7). The ramus contains I 3 , a tooth not present in the Wikieup sample, but the state of preservation is such that little can be said of the tooth. The size of the dentition, the depth of the ramus, the heavily crenulted enamel on Ci, and the large premolars support the referral of these specimens to P. lindsayi. Two isolated canine teeth, C 1 and Ci, from a third Arizona locality, Redington Quarry, match well those from the type locality in size and degree of crenulation. An isolated left P 4 from Pinole Tuff, California, is very close in size to the mean of the P 4 ’s from Wikieup and bears the same strong lingual bulge. Three partial radii, two partial ulnae, and a NUMBER 46 23 Table 7. —Measurements (cm) of Plesiogulo lindsayi, new species (O.R. = observed range) Element No. O.R. OLD CABIN QUARRY P 2 width 1 0.66 P 3 length 1 1.45 width 1 0.87 M 1 length, lingual 1 1.51 length, median 1 1.12 width 1 1.97 Ci length 1 1.81 P ‘2 length 1 0.87 P 3 length 1 0.73 REDINGTON QUARRY C 1 length 1 1.70 width 1 1.38 Ci length 1 1.79 width 1 1.47 Ulna length PINOLE TUFF 1 15.81 P 4 length 1 1.52 width 1 0.95 partial ischium from Wikieup (Figure 16) and a partial ulna from Redington Quarry are referred to P. lindsayi (Table 7). The Redington Quarry ulna is the largest of the three; its distal end is complete, but several fragments are missing from the coronoid process and trochlear notch. One ulna from Wikieup (Figure 16 c,d) has an un¬ damaged proximal end while the other has an undamaged distal end (Figure 16 e,f). The Red¬ ington Quarry and Wikieup ulnae bear the same long, concave olecranon process as the ulna of P. marshalli. The styloid process, present in two of the specimens, is heavy and rounded as in P. marshalli. The groove in the top of the olecranon process and the sharp posteromedial crest on the distal end are also present in the ulna of P. lindsayi. The partial radii from Wikieup are very similar to those of P. marshalli. The two right radii are missing only the distal articular surface, but the left radius consists only of the proximal quarter. The bicipital tuberosity, the shape of the proxi¬ mal articular surface, and the deep lateral groove on the distal end are all as in the radius of P. marshalli. A pelvic fragment, consisting of the left ischium and the posterodorsal portion of the acetabulum, is questionably associated with an ulna (F:AM 67957) from Wikieup (Figure 16 ej). The ischium is longer in Osteoborus than in Plesiogulo. The ace¬ tabulum is deeper in both Plesiogulo and Gulo than in Osteoborus. The postcrania referred to P. lindsayi are larger than those referred to P. marshalli ; however, this size differential is not so great as that between their respective crania and dentitions. It would appear that P. lindsayi had a somewhat larger body, but a much larger head than P. marshalli. Discussion. —Schlosser (1903:26) described as Lutra brachygnathus a ramus from an unknown locality in China, “allegedly from Tientsin.” Zdansky (1924:38) erected the genus Plesiogulo , designating L. brachygnathus as the genotypic spe¬ cies, and went on to refer and describe additional material from the Paote area of Shansi Province, People’s Republic of China. This material, con¬ sisting of about a dozen whole and partial crania, some with associated inferior dentitions, is housed in the Lagrelius collection at the Paleontological Institute at Uppsala. Subsequent referral of spec¬ imens to P. brachygnathus include a ramus from the Bhandar bone bed in the Siwaliks of Pakistan (Lewis, 1933) and a skull and three mandibles from Pavlodar, Siberia (Orlov, 1941). Teilhard de Chardin (1945) recognized three subspecies or “formes” from two new localities in China; these were P. b. minor from K’ingyang, Kansu, and P. b. crassa and P. b. major from Yushe, Shansi. The first two subspecies were each based on a partial skull and ramus, while the third was based on a right ramus. Kurten (1970) elevated Teilhard de Chardin’s subspecies to specific rank and described P. prae- cocidens based on a partial ramus from Paote, Shansi. He advocated restricting the name P. brachygnathus to the type ramus alone and trans¬ ferred to P. crassa the specimens from Siberia and Pakistan and all but two of the Paote specimens. Study of the much larger sample of P. crassa from Paote located in the America Museum of Natural History indicates that P. minor falls within the 24 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Figure 16 — Plesiogulo lindsayi, new species: a,b, F:AM 67958, right radius, anterior and posterior views; c,d, F:AM 105372, partial right ulna, lateral and medial views; ej\ F:AM 67957, right ulna, lateral and medial views. (Specimens from Wikieup; X 1.) range of variation of P. crassa. Viret (1939:12) described P. monspessulanus, a very large species lacking a metaconid on Mi, from Montpellier, France. Hendey (1978) re¬ ferred two partial skulls, three mandibles, and assorted postcrania from Langebaanweg, South Africa. Both of the Langebaanweg Mi’s were too damaged to permit an unqualified statement re¬ garding the metaconid. Hendey (1978:333) judges that it was “either very small or absent.” Plesiogulo lindsayi most closely resembles in size P. major and P. monspessulanus. Plesiogulo major dif¬ fers from P. monspessulanus in slightly larger size and a strong Mi metaconid. Kurten (1970:18) sugests that the presence or absence of this cusp is subject to individual variation, and Hendey (1978:343) advocates disregarding the character entirely and lumping the two taxa under P. mon- NUMBER 46 25 spessulanus. The metaconid, however, is consis¬ tently well developed in populations of P. lindsayi, as well as in P. marshalli and P. crassa , and would appear to be a valid character at the specific level. The phylogenetic relationships of the Old and New World species of Plesiogulo are currently under study by the author and will be treated in a subsequent publication. The WaKeeney Specimen An isolated left M 1 (UMMP V55756) from the late Clarendonian WaKeeney Local Fauna, Trego County, Kansas, was referred to Plesiogulo by Wilson (1968:111). The enamel of this tooth is not crenulated and the inner lobe is not so expanded anteroposteriorly as in either P. mar¬ shalli or P. lindsayi. The paracone and metacone are more widely separated and their smoothly curved external border contrasts with the bilobate curve seen in P. marshalli and P. lindsayi. A small cusp is present slightly lingual and posterior to the metacone, which is not present in Plesiogulo. The WaKeeney specimen is much smaller than the smallest M 1 of P. marshalli', dimensions are: lingual length 1.08 cm; median length 0.81 cm, and width 1.29 cm. The WaKeeney specimen is not Plesiogulo, but an undescribed species of Mionictis Matthew, 1924. The tooth corresponds closely to the M l5 s in an undescribed partial skull (F:AM 63296) of Mionictis from MacAdams Quarry, Clarendon area, Texas (H. Galiano, pers. comm., 1979). Conclusions Remains of Plesiogulo from 14 North American localities are referred to two species: P. marshalli from Edson, Optima (Guymon), Coffee Ranch, Modesto Reservoir (Turlock Lake), McKay Res¬ ervoir, San Juan Quarry, Upper Bone Valley, White Cone, Lost Quarry, and Ordance (Wes- tend Blowout) and P. lindsayi, new species, from Wikieup, Redington Quarry, Old Cabin Quarry, and Pinole Tuff. Plesiogulo lindsayi, new species, is distinguished from P. marshalli by its greater size, the placement of the P 4 protocone, the presence of a lateral cingulum on P 4 , the presence of a strong lingual bulge on P 3 and P 4 , and the shape of the inner lobe of M 1 . The heretofore undes¬ cribed postcrania of both species are, as might be expected, reminiscent of Gulo in several charac¬ ters. Plesiogulo is widely distributed across the south¬ ern United States, but is nowhere abundant. Low population density may be the reason for the scarcity of fossil material. A preference for large individual territories in Gulo results in low popu¬ lation densities of that species and it is possible that a similar cause and effect was operative in populations of Plesiogulo. Plesiogulo has been commonly interpreted as a forest-dweller, and thus its presence is considered indicative of woodland in the vicinity of deposi¬ tion. This interpretation of Plesiogulo is based primarily on analogy to the extant wolverine, Gulo, an animal often misinterpreted as an obli¬ gate forest-dweller; Gulo is actually well adapted to the open tundra. Plesiogulo is most abundant in faunas with a large percentage of hypsodont, cursorial ungulates, suggestive of open, grassy plains. Hence, this genus need not be interpreted as a strictly woodland inhabitant. Plesiogulo originated in Asia and migrated to North America between 7.0 and 6.5 million years ago. This taxon is known in North America only from the late Hemphillian. Literature Cited Baskin, J. A. 1979. Small Mammals of the Hemphill Age White Cone Local Fauna, Northeastern Arizona. Journal of Pa¬ leontology , 53(3):695-708, 2 figures. Bennett, D. K. 1979. The Fossil Fauna from Lost and Found Quarries (Hemphillian: Latest Miocene), Wallace County, Kansas. University of Kansas Museum of Natural His¬ tory Occassional Paper , 79:1-24. Bjork, P. R. 1970. The Carnivora of the Hagerman Local Fauna (Late Pliocene) of Southwestern Idaho. Transac¬ tions of the American Philosophical Society , new series, 60(7): 1-54. Boellstorff, J. D. 1976. The Succession of Late Cenozoic Volcanic Ashes in the Great Plains: A Progress Report. Kansas Geological Survey, Guidebook for the 24th Annual Meeting of the Mid-western Friends of the Pleistocene, series 1, 1976:37-71. Bryant, L. J. 1968. A New Genus of Mustelid from the Ellensburg Formation, Washington. Los Angeles County Museum Contributions to Science, 139:1-6. Dalquest, W. W. 1969. Pliocene Carnivores of the Coffee Ranch (Type Hemphill) Local Fauna. Bulletin of the Texas Memorial Museum, 15:1-43. Gill, T. 1872. Arrangement of the Families of Mammals with Analytical Tables. Smithsonian Miscellaneous Collec¬ tions, 11(1): 1-98. Grinell, J., J. S. Dixson, and J. M. Linsdale 1937. Furbearing Mammals of California. Volume 1, 375 pages. Berkeley: University of California Press. Hall, E. R., and K. R. Kelson 1959. The Mammals of North America. Volume 2, 536 pages. New York: Ronald Press Company. Hendey, Q. B. 1978. Late Tertiary Mustelidae (Mammalia, Carnivora) from Langebaanweg, South Africa. Annals of the South Africa Museum, 76(10): 329—357. Hesse, C. J. 1936. A Pliocene Vertebrate Fauna from Optima, Okla¬ homa. University of California Publication, Bulletin of the Department of Geological Sciences, 24(3):57-70. Hibbard, C. W. 1934. Two New Genera of Felidae from the Middle Pliocene of Kansas. Transactions of the Kansas Acad¬ emy of Science, 37:239-255. Izett, G. J. 1975. Late Cenozoic Sedimentation and Deformation in Northern Colorado and Adjoining Areas. In B. F. Curtis, editor, Cenozoic History of the Southern Rocky Mountains. Geological Society of America Memoir, 144:179-209. Jackson, H. H. T. 1961. Mammals of Wisconsin. 504 pages. Madison: Uni¬ versity of Wisconsin Press. Jacobs, L. L. 1977. Rodents of the Hemphillian Age Redington Local Fauna, San Pedro Valley, Arizona. Journal of Paleontology, 51(3): 505-519. Krott, P. 1959. Demon of the North. 260 pages. New York: Knopf. 1960. Ways of the Wolverine. Natural History, 69:16-29. Kurten, B. 1970. The Neogene Wolverine Plesiogulo and the Origin of Gulo (Carnivora, Mammalia). Acta Zoologica Fenmca, 131:1-22. Lewis, G. E. 1933. Notice of the Discovery of Plesiogulo brachygnathus in the Siwalik Measures of India. American Journal of Science, 4:161. MacFadden, B. J. 1977. Magnetic Polarity Stratigraphy of the Chamita Formation Stratotype (Mio-Pliocene) of North- Central New Mexico. American Journal of Science, 277:769-800. MacFadden, B. J., and K. Manley 1976. Magnetic Stratigraphy, Tephrochronology, and Mammalian Biostratigraphy of the Type Chamita Formation, North-Central New Mexico. Geological Society of America Abstracts with Programs, 8(5):605. MacFadden, B. J., N. M. Johnson, and N. D. Opdyke 1979. Magnetic Polarity Stratigraphy of the Mio-Pli¬ ocene Mammal-Bearing Big Sandy Formation of Western Arizona. Earth and Planetary Science Letters, 44:349-364. Martin, H. T. 1928. Two New Carnivores from the Pliocene of Kansas. Journal of Mammalogy, 9:233-236, plates 20-21. Matthew, W. D. 1924. Third Contribution to the Snake Creek Fauna. Bulletin of the American Museum of Natural History, 50(2):59-210. 26 NUMBER 46 27 Matthew, W. D., and R. A. Stirton 1930. Equidae from the Pliocene of Texas. University of California Publication, Department of Geological Sciences, 19(17): 349-396. Miller, G. S., Jr. 1912. Catalogue of the Mammals of Western Europe (Europe Exclusive of Russia) in the British Museum. 1019 pages. London: British Museum (Natural History). Myhre, R., and S. Myrberget 1975. Diet of Wolverines (Gulo gulo) in Norway .Journal of Mammalogy , 56:752-757. Nowak, R. M. 1973. Return of the Wolverine. National Parks and Conser¬ vation Magazine, 47(2):20-23. Orlov, J. A. 1941. Tertiary Carnivora of West Siberia, III: Musteli- nae. In J. A. Orlov, Tertiary Mammalia and the Localities of Their Remains. Travaux de I’Institut Paleontologique, Academie des Sciences de I’Union des Republiques Sovietiques Socialistes, 8(3):30-39. Pocock, R. I. 1920. On the External Characters of the Ratel ( Mellivora ) and the Wolverine (Gulo). Proceedings of the Zoolog¬ ical Society of London , 1920:179-187. Reed, L. C., and O. M. Longnecker, Jr. 1932. The Geology of Hemphill County, Texas. Univer¬ sity of Texas Bulletin, 3231:1-98. Savage, D. E. 1941. Two New Middle Pliocene Carnivores from Okla¬ homa with Notes on the Optima Fauna. American Midland Naturalist , 25(3):692-710. Scarborough, R. B., P. E. Damon, and M. Shafiqullah 1974. K-Ar Age for a Basalt from the Volcanic Member (Unit 5) of the Bidahochi Formation. Geological Society of America Abstracts with Programs, 6(5):472. Schlosser, M. 1903. Die fossilen Saugetheire Chinas nebst einer Odon- tographie der recenten Antilopen. Abhandlungen der Bayenschen Akademie der Wissenschaften, 22(1): 1-22. Schultz, G. E. 1977. Guidebook for the Field Conference on Late Cen- ozoic Biostratigraphy of the Texas Panhandle and Adjacent Oklahoma. Kilgore Research Center Special Publication, 1:1 -160. Sellards, E. H., W. S. Adkins, and F. B. Plummer 1932. The Geology of Texas, Volume 1: Stratigraphy. University of Texas Bulletin, 3232:1-1770. Seton, E. T. 1929. Lives of Game Animals. Volume 2, part 2, 746 pages. Garden City: Doubleday, Doran, and Company, Inc. Shotwell, J. A. 1955. An Approach to the Paleoecology of Mammals. Ecology, 36(2):327-337. 1956. Hemphillian Mammalian Assemblage from Northeastern Oregon. Geological Society of America Bulletin, 67:717-738. 1958. Inter-community Relationships in Hemphillian (Mid-Pliocene) Mammals. Ecology , 39(2):271-282. Slaughter, B. H., R. H. Pine, and N. E. Pine 1974. Eruption of Cheek Teeth in Insectivora and Car¬ nivora. Journal of Mammalogy, 55(1): 115-125. Teilhard de Chardin, P. 1945. Les Mustelides de Chine. Institut de Geobiologie, Pekin, 12:1-56. Viret, J. 1939. Monographic Paleontologique de la Faune de Ver- tebres des Sable de Montpellier, III: Carnivora, Fissipedia. Travaux du Laboratoire de Geologie de la Faculte Sciences de Lyon, 37:5-26. Walker, E. P. 1964. Mammals of the World. 2 volumes, 1500 pages. Bal¬ timore: Johns Hopkins University Press. Webb, S. D. 1969. The Burge and Minnechaduza Clarendonian Mammalian Faunas of North-Central Nebraska. University of California Publications in Geological Sciences, 78:1-191. Wilson, R. L. 1968. Systematics and Faunal Analysis of a Lower Pli¬ ocene Vertebrate Assemblage from Trego County, Kansas. Contributions from the Museum of Paleontology, University of Michigan, 22(7):75-126. Wood, H. E., R. W. Chaney, J. Clark, E. H. Colbert, G. L. Jepsen, J. B. Reeside, Jr., and C. Stock 1941. Nomenclature and Correlation of the North Amer¬ ican Continental Tertiary. Geological Society of Amer¬ ica Bulletin, 52:1-48. Zdansky, O. 1924. Jungtertiare Carnivoren Chinas. Paleontologica Sin- ica, series C, 2(1): 1-155. REQUIREMENTS FOR SMITHSONIAN SERIES PUBLICATION Manuscripts intended for series publication receive substantive review within their originating Smithsonian museums or offices and are submitted to the Smithsonian Institution Press with approval of the appropriate museum authority on Form SI-36. Requests for special treatment—use of color, foldouts, casebound covers, etc.—require, on the same form, the added approval of designated committees or museum directors. 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 below, will govern, within the judgment of the Press, acceptance or rejection of the manuscripts and art. Copy must be typewritten, double-spaced, on one side of standard white bond paper, with 1 %" 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 establish¬ ed format, table of contents with indents reflecting the heads and structure of the paper. First page of text should carry the title and author at the top of the page and an unnum¬ bered footnote at the bottom consisting of author’s name and professional mailing address. Center heads of whatever level should be typed with initial caps of major words, with extra space above and below the head, but with no other preparation (such as all caps or underline). Run-in paragraph heads should use period/dashes or colons as necessary. 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 rules or formal, numbered table heads. Formal tables (numbered, with table heads, boxheads, stubs, rules) should be sub¬ mitted as camera copy, but the author must contact the series section of the Press for edito¬ rial attention and preparation assistance before final typing of this matter. Taxonomic Keys in natural history papers should use the alined-couplet form in the zoology and paleobiology series and the multi-level indent form in the botany series. If cross-referencing is required between key and text, do not include page references within the key, but number the keyed-out taxa with their corresponding heads in the text. Synonymy in the zoology and paleobiology series must use the short form (taxon, author, year:page), with a full reference at the end of the paper under "Literature Cited.” For the botany series, the long form (taxon, author, abbreviated journal or book title, volume, page, year, with no reference in the "Literature Cited”) is optional. Footnotes, when few in number, whether annotative or bibliographic, should be typed at the bottom of the text page on which the reference occurs. Extensive notes must appear at the end of the text in a notes section. If bibliographic footnotes are required, use the short form (author/brief title/page) with the full reference in the bibliography. Text-reference system (author/year/page within the text, with the full reference in a "Literature Cited” at the end of the text) must be used in place of bibliographic footnotes in all scientific series and is strongly recommended in the history and technology series: “(Jones, 1910:122)” or . . Jones (1910:122).” Bibliography, depending upon use, is termed "References,” "Selected References,” or “Literature Cited.” Spell out book, journal, and article titles, using initial caps in all major words. For capitalization of titles in foreign languages, follow the national practice of each language. Underline (for italics) book and journal titles. Use the colon-parentheses system for volume/number/page citations: "10(2):5-9.” For alinement and arrangement of elements, follow the format of the series for which the manuscript is intended. Legends for illustrations must not be attached to the art nor included within the text but must be submitted at the end of the manuscript—with as many legends typed, double¬ spaced, to a page as convenient. Illustrations must not be included within the manuscript but must be submitted sepa¬ rately as original art (not copies). All illustrations (photographs, line drawings, maps, etc.) can be intermixed throughout the printed text. They should be termed Figures and should be numbered consecutively. If several "figures” are treated as components of a single larger figure, they should be designated by lowercase italic letters (underlined in copy) on the illus¬ tration, in the legend, and in text references: “Figure 9b," If illustrations are intended to be printed separately on coated stock following the text, they should be termed Plates and any components should be lettered as in figures: “Plate 9b,” Keys to any symbols within an illustration should appear on the art and not in the legend. A few points of style: (1) Do not use periods after such abbreviations as "mm, ft, yds, USNM, NNE, AM, BC.” (2) Use hyphens in spelled-out fractions: "two-thirds.” (3) Spell out numbers “one” through "nine” in expository text, but use numerals in all other cases if possible. (4) Use the metric system of measurement, where possible, instead of the English system. (5) Use the decimal system, where possible, in place of fractions. (6) Use day/month/year sequence for dates: "9 April 1976.” (7) For months in tabular list¬ ings or data sections, use three-letter abbreviations with no periods: “Jan, Mar, Jun,” etc. Arrange and paginate sequentially EVERY sheet of manuscript —including ALL front matter and ALL legends, etc., at the back of the text—in the following order: (1) title page, (2) abstract, (3) table of contents, (4) foreword and/or preface, (5) text, (6) appendixes, (7) notes, (8) glossary, (9) bibliography, (10) index, (11) legends.