'4P ^ifk New Enaliarctos^ (Pinnipedimorpha) from the Oligocene and Miocene of Oregon and the Role of "Enaliarctids" in Pinniped Phylogeny SERIES PUBUCATIONS 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, Josei;^ Henry outHned a program that irv;luded the following statement: *lt 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 ?dl branches of knowledge." This tt^me of t>ask: research has been adhered to through the years by thousarKte of titles issued in series publk^tkjns under the SmithscKuan imprint, (xynmencing with Smithsonian Contributions to Knowlodge in 1848 arKl contirniing wittt the following active series: Smithsonian Contributions to Anthrc^ogy Smithsonian Contritjutions to Astrophysics Smithsonian Contributions to Botany Smithsonian Contributions to the Earth Sciences Smithsonian Contributions to the Marine Sciences Smithsonian Contributions to Paleobiology Smithsoni^i Contributions to Zoology Smithsonian Folklife Studies Smithsonian Studies in Air and Space Smithsonian Studies in History and Technology In these series, the Institution putilishes small papers arKl full-scale monographs ttiat report the research and collections of its various museums and bureaux or of professkxial colleagues in the worid of science and schdarship. The publications are distrit)uted by mailing lists to NtKaries, universities, and similar institutkins throughout ttie worid. Papers or monographs submitted for series put>licatk>n are received by the Smithsonian tnstitutk>n Press, subject to its own review for format and style, only through departonents of the various SmithscMiian museums or bureaux, where the martuscr^ts are given substantive review. Press recMrements for manuscript and art prepwatton are outlined on the Inside back cover. Robert McC. Adams Secretary Smithsonian InstltutkMi S M I T H S O N I A N C O N T R I B U T I O N S T O P A L E O B I O L O G Y ? N U M B E R 6 9 New Enaliarctos^ (Pinnipedimorpha) from the Oligocene and Miocene of Oregon and the Role of "Enaliarctids" in Pinniped Phylogeny Annalisa Bert a SMITHSONIAN INSTITUTION PRESS Washington, D.C. 1991 A B S T R A C T Berta, Annalisa. New Enaliarctos* (Pinnipedimorpha) from the Oligocene and Miocene of Oregon and the Role of "Enaliarctids" in Pinniped Phylogeny. Smithsonian Contributions to Paleobiology, number 69, 33 pages, 22 figures, 1991.?Three new species of the pinnipedimorph Enaliarctos* are described from the marine late Oligocene and early Miocene (Arikareean and Hemingfordian or early Barstovian correlatives) of coastal Oregon. Enaliarctos tedfordl, new species, is based on a partial cranium from the late Oligocene Yaquina Fonnation. A related new species, Enaliarctos emlongi, is founded on a nearly complete cranium, jaws, and associated skeletal elements from the late Oligocene to early Miocene Nye Mudstone. A third new species, Enaliarctos barnesi, is based on a partial cranium and jaws from late Oligocene or early Miocene rocks near the contact between the Yaquina Formation and the Nye Mudstone. Another skull, from the Nye Mudstone, is referred to a previously described species, Enaliarctos mitchelli Barnes, 1979. Three of these species, E. mitchelli, E. emlongi, and E. tedfordl form a monophyletic clade, united by reduced cheek teeth cingula and short metacone of the upper camassial. The major trend observed in Enaliarctos over 10 million years of history is an intermediate stage in the transformation to homodonty evidenced by premolarization of the upper camassial and molars and reduction and simplification of cusps on the lower camassial. Cladistic analysis of 52 cranial and dental characters suggests the following phylogenetic hypotheses: (1) the subfamily "Enaliarctinae" (= "Enaliarctidae") is paraphyletic, (2) monophyly of the genus Enaliarctos* is questioned although the status of this taxon as sister taxon to other pinnipeds is affirmed, (3) other "enaliarctid" pinnipeds, Pteronarctos and Plnnarctidlon, are assigned to less inclusive pinniped clades {Pteronarctos -I- all other pinnipeds and Pinnarctidion + Desmatophoca, Allodesmus, and the Phocidae). 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 Berta, Annalisa New Enaliarctos* (Pinnipedimorpha) from the Oligocene and Miocene of Oregon and the role of "Enaliarctids" in Pinniped phylogeny / Annalisa Berta. p. cm.?(Smithsonian contributions to paleobiology ; no. 69) Includes bibliographical references. 1. Enaliarctos. 2. Paleontology?Oligocene. 3. Paleontology?Miocene 4. Paleontology?Ore? gon. I. Title. II. Series. QE701.S56 no. 69 [Qe882.P51 560 s-dc20 [569'.74] 91-18125 ? The paper used in this publication meets the minimum requirements of the American National Standard for Permanence of Paper for Printed Library Materials Z39.48?1984. Contents Page Introduction 1 Abbreviations 1 Acknowledgments 2 Location, Stratigraphy, and Correlation 3 Previous Work 4 PINNIPEDIMORPHA 4 Enaliarctos* Mitchell and Tedford 5 Enaliarctos emlongi, new species 5 Enaliarctos barnesi, new species 13 Enaliarctos tedfordi, new species 16 Enaliarctos mitchelli Barnes 19 Interspecific Affinities of Enaliarctos* 25 Higher Level Relationships of Enaliarctos* and "Enaliarctids" 27 Conclusions 31 Literature Cited 32 Ul New Enaliarctos'^ (Pinnipedimorpha) from the Oligocene and Miocene of Oregon and the Role of "Enaliarctids" in Pinniped Phylogeny Annalisa Berta Introduction "Enaliarctids" are basal pinnipeds implicated in the higher level phylogeny of all other pinnipeds (including otariids, odobenids, phocids, and their fossil allies). Initial study established "enaliarctids" as an intermediate evolutionary stage between ursoids and "otarioids" (including otariids, odobenids, and the extinct desmatophocids) (Mitchell and Tedford, 1973). This interpretation was part of a broader, diphyletic view of pinnipeds that proposed that "otarioids" derived from ursids whereas phocids originated from mustelids (Mivart, 1885; McLaren, 1960; Tedford, 1976). This notion was widely accepted (Repenning, 1976; Repenning and Tedford, 1977; Barnes, 1979; de Muizon, 1982; King, 1983). Recently, however, Wyss (1987) rejected the "Otarioidea" as based largely on primitive characters and he marshalled osteological, soft anatomical, and biomolecular evidence to support a monophyletic origin of pinnipeds. He concluded that "enaliarc? tids" as traditionally defined are paraphyletic and he recognized Enaliarctos as the sister group to the remaining pinnipeds. Pinniped monophyly was further strengthened by skeletal evidence provided by Wyss (1988a), Berta et al. (1989), and Berta and Ray (1990). Recent reviews of carnivoran phylogeny demonstrate, Annalisa Berta, Department of Biology, San Diego Stale University. San Diego. California, 92182 (Research Associate in the Department of Paleobiology, National Museum of Natural History, Smithsonian Institution). Review Chairman: Clayton E. Ray, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. Reviewers: Lawerence G. Barnes, Natural History Museum of Los Angeles County, Los Angeles; and Andre R. Wyss, University of California, Santa Barbara. however, that there is no consensus regarding pinniped relationships. While expressing some reservations, Flynn et al. (1988) endorsed monophyly whereas Wozencraft (1989) rejected Wyss's arguments and concluded that pinnipeds are diphyletic. Barnes (1989) also contested Wyss's conclusions and supported recognition of the "Otariidae" (= "Otarioidea") comprised of the following subfamilies: Enaliarctinae, Otar- iinae, Desmatophocinae, Allodesminae, Imagotariinae, Dusig- nathinae, and Odobeninae. This latter arrangement follows that originally proposed by Mitchell (1968, 1975). A large, diverse, well-preserved collection of fossil pin? nipeds representing the genus Enaliarctos* found by Douglas Emlong from late Oligocene-early Miocene rocks of coastal Oregon formed the basis for this study. This report provides description of Enaliarctos* species new to the published record and an analysis of their interrelationships. Also considered is the role of "enaliarctids" in pinniped phylogeny. This study suggests that Enaliarctos should be regarded as a metataxon, so indicated by an asterisk (*). This convention, an asterisk beside the taxonomic name (formulated by Gauthier, 1986; see also Gauthier et al., 1988; Donoghue, 1985) is used to designate taxa for which there is no unambiguous character evidence supporting either monophyly or paraphyly. ABBREVIATIONS.?The following institutional abbreviations are used: UCMP Museum of Paleontology, University of California, Berkeley, CA USNM Collections of the former United States National Museum, now in the National Museum of Natural History, Smithsonian Institution, Washington, D.C. SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY Type Locality Enaliarctos emlongi uSNM 250345 Portland N c O o o CB CL Enaliarctos mitchelli USNM 175637 ml. Type Locality Enaliarctos tedfordl USNM 206273 Type Locality USNM 314295 Enaliarctos barnesi Seal Rocks FIGURE I.?^Detail of part of the Yaquina 15-minute quadrangle, USGS, showing the localities discussed in the text. Arrow indicates location of map enlargement. ACKNOWLEDGMENTS.?Specimens were skillfully prepared by Ed Pederson (American Museum of Natural History), Charles A. Repenning (U.S. Geological Survey, Denver), and Gladwyn Sullivan and Michael Tiffany (National Museum of Natural History). Line drawings were made by Christi Endres. Michael McCaffery and Nicki Watson (San Diego State University) greatly assisted with specimen photography. David Bohaska and Robert Purdy (National Museum of Natural History) provided locality information and assistance with the measurement of specimens. Lawrence G. Barnes (Natural History Museum of Los Angeles County) provided casts of comparative specimens. For permission to study specimens under their care I thank L.G. Barnes, Clayton E. Ray (National Museum of Natural History), and Richard H. Tedford (American Museum of Natural History). NUMBER 69 Time (mybp) 2 0 " 25 30 - 3 5 - Epochs LU Z LU 8 LU ?o OJ LU CO LU Benthic Foraminif. Stages SAUCESIAN ZEMORRIAN Molluscan Stages NEWPORTIAN 9 PILLARIAN JUANIAN MATLOCKIAN North American Land Mammal Ages BARSTOVIAN HEMINGFORDIAN ARIKAREEAN WHITNEYAN Central Coast Range Oregon ASTORIA FM. NYE MUDSTONE YAQUINA FM. ALSEA FM. \ r ? Enaliarctos emlongi Enaliarctos mitchelli Enaliarctos barnesi Enaliarctos tedfordi FIGURE 2.?Correlation of rock units containing Enaliarctos species (after Berggren et al., 1985 (time scale and correlation of NALMA) and Armentrout et al., 1983). L.G. Barnes, R.H. Tedford, A.R. Wyss, C.E. Ray, and CR. Crumly have provided valuable comments on the manuscript. Emlong's field work was supported by the Smithsonian Institution through the Smithsonian Research Foundation and the Walcott and Kellogg Funds. I gratefully acknowledge financial support for this research provided by NSF grant BSR 8607061. Location, Stratigraphy, and Correlation Marine sedimentary rocks crop out in structural basins along the Oregon coast at Astoria, Tillamook, Newport, and Coos Bay (Snavely et al., 1964). Fossil specimens referred to Enaliarctos* described in this report were collected by Douglas Emlong from rock units exposed in the Yaquina Bay area between Newport and Waldport in northwestern Lincoln County, Oregon (Figure 1). This region has been mapped (Snavely et al., 1976) and more detailed descriptions of the regional geology are given in Snavely et al., 1980. The fossils described below come from three successive formations of late Oligocene through early Miocene age. The Yaquina Formation, composed of sandstone and siltstone of late Oligocene age, overlies siltstone of the Alsea Formation (Snavely et al., 1980). This formation has been assigned to the Juanian Molluscan Stage (Addicott, 1976; Armentrout, 1981) and Zemorrian Foraminiferal Stage (Armentrout et al, 1983, chart). The Yaquina Formation has produced, in addition to pinnipeds, desmostylians (Domning et al., 1986), cetaceans (Emlong, 1966; Barnes, 1987b), and a single land mammal, an anthracothere. The stage of evoludon of the anthracothere (see Domning et al., 1986) suggests correlation of this formation with the early part of the Arikareean land mammal age, approximately 28-25 Ma (Figure 2). Marine siltsone and fine-grained sandstone of the Nye Mudstone overlie by gradational contact the Yaquina Forma? tion (Snavely et al., 1980). The Nye Mudstone is late Oligocene to early Miocene in age and has been assigned to the Pillarian Molluscan Stage (Addicott, 1976; Armentrout, 1981), which has been correlated with the Saucesian Foraminiferal Stage (Armentrout et al, 1983, chart). The Nye Mudstone can be correlated with the middle and late Arikareean and early Hemingfordian, approximately 25-18 Ma (Figure 2). Sandstone and siltstone of the Astoria Formation un- SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY conformably overlie the Nye Mudstone (Snavely et al., 1980). The holotype of Enaliarctos emlongi, according to Emlong's field notes, was collected "north of the Nye-Astoria contact in the Astoria Formation." Repenning (pers. comm.) has sug? gested that this specimen may actually be from the upper part of the Nye Mudstone. A sample of the enclosing matrix of USNM 250345, processed for foraminifera by Kristin MacDougall of the U.S. Geological Survey, Menlo Park, yielded taxa that were not age diagnostic. The Astoria Formation spans the late part of the early Miocene through the early part of the middle Miocene (Barnes, 1987a; Armentrout, 1981) and has been assigned to the Newportian Molluscan Stage (Addicott, 1976; Armentrout, 1981), which has been correlated with the Saucesian Foraminiferal Stage (Addicott, 1976; Armenurout et al., 1983, chart). Two other fossil pinnipeds have been recovered from the Astoria Formation. Desmatophoca oregonensis was collected from the Astoria Formation, at an unspecified horizon (Barnes, 1987a) and Pteronarctos goedertae was collected from a horizon just above the base of the Astoria Formation where it contacts the underlying Nye Mudstone (Barnes, 1989). A chalicothere collected from the Iron Mountain horizon within the Astoria Formation together with other biostratigraphic (molluscan and foraminiferal) and radiometric evidence provided a correlation of the Astoria Formation with the Hemingfordian or early part of the Barstovian between 19 and 15 Ma (Barnes, 1987a; Munthe and Coombs, 1979) (Figure 2). Previous Work Table 1 summarizes previous arrangements of "enaliarctids" discussed below. Mitchell and Tedford (1973) described a new genus and species of the pinnipedimorph, Enaliarctos mealsi, from early Miocene rocks widiin the Pyramid. Hill Sandstone Member of the Jewett Sand at Pyramid Hill, Kern County, California. They erected the new subfamily of the Otariidae, Enaliarctinae, to accommodate this taxon. As noted previously, Enaliarctos* was viewed as a transitional form, intermediate between terrestrial carnivores, especially ursoids, and the aquatic Otariidae. Tedford (1976) elevated the Enaliarctinae to familial level and placed die group in the superfamily Otarioidea. Barnes (1979) described a second species of Enaliarctos*, E. mitchelli, and proposed a new genus and species, Pinnarctidion bishopi; both were collected from Pyramid Hill, California, from rocks at a stratigraphically higher level within the Pyramid Hill Sandstone Member than die deposits that produced E. mealsi. A third new genus and species of "enaliarctid" was recently described by Barnes (1989), Pter? onarctos goedertae, from the late early Miocene Astoria Formation near Newport, Lincoln County, Oregon. In this same paper, Barnes reviewed the status of anodier taxon, Kamt- schatarcios sinelnikovae, from die Miocene of die USSR proposed by Dubrovo (1981, 1984) as a new genus and species of "enaliarctid" for which she established a new subfamily, die TABLE 1.?Previous classifications of "enaliarctids" and selected groups. 1. MitcheU and Tedford, 1973 Family OTARIIDAE SubfamUy ENALIARCTINAE Enaliarctos mealsi 2. Tedford, 1976 Superfamny OTARIOIDEA Family ENALIARCTIDAE Enaliarctos mealsi Family OTARIIDAE (including Odobeninae) 3. Bames, 1979; 1989 Family OTARIIDAE Subfamily ENALDVRCTINAE Enaliarctos mealsi, E. mitchelli Pinnarctidion bishopi Pteronarctos goedertae Kamtschatarctinae. Bames reassigned Kamtschatarctos to the Imagotariinae, which he recognized as a subfamily of the Otariidae. Thus, Bames recognized the Enaliarctinae including Enaliarctos*, Pinnarctidion, and Pteronarctos as a basal group of otariid pinnipeds involved in the ancestry of various otariid lineages. Aldiough Wyss (1987) proposed no formal classification of pinnipeds, he regarded die "Enaliarctidae" as paraphyletic. Enaliarctos was recognized as the sister group of all odier pinnipeds. Pinnarctidion and Allodesmus were regarded as members of a group including odobenids and phocids. My analysis agrees widi the latter arrangement, and I provide evidence for recognition of Enaliarctos as a metataxon and Pteronarctos as die sister taxon of a clade that includes otariids, odobenids, desmatophocids, and phocids. Additional skeletal material referrable to Pteronarctos is presentiy under study (Berta, in prep.). PINNIPEDIMORPHA This taxon includes Enaliarctos*, Pteronarctos, Otariidae, Odobenidae, Allodesmus, Desmatophoca, Pinnarctidion, and Phocidae. DEFINITION.?The most recent common ancestor of En? aliarctos and all of its descendants. DIAGNOSIS.?^Pinnipedimorphs are distinguished by die following synapomorphies: posterior extension of the palatine process of the maxilla; large infraorbital foramen; anterior palatine foramina positioned anterior to the maxillary-palatine suture; gready reduced or absent lacrimal that does not contact die jugal; supraorbital processes absent or large and shelf-like; foramen rotundum merged with the anterior lacerate foramen; large epitympanic recess; postglenoid foramen vestigial- absent; jugular foramen gready enlarged; enlarged auditory ossicles; pseudosylvian sulcus strongly developed; M '^^ small relative to premolars; M, entoconid and hypoconid reduced- absent; M3 absent; short, robust humerus with strongly developed deltopectoral crest and enlarged tuberosities; loss of NUMBER 69 entepicondylar foramen on humerus; elongation of digit I in the manus and digits I and V in the pes; short ilium; extemely short, anteroposteriorly flattened femur and medially inclined con? dyles; large, broadly developed greater frochanter on the femur; conical patella. COMMENTS.?Berta et al. (1989) originally proposed die name Pinnipedimorpha to include Enaliarctos and all odier pinnipeds and used postcranial synapomorphies (listed above) to diagnose this monophyletic group. Barnes' (1989) arrange? ment differs in his incorporation of Enaliarctos in a more exclusive group of pinnipeds (Otariidae sensu lato) in which die Phocidae were not included. Of the 19 characters used by Barnes to support the monophyly of the "Otariidae" most apply to more inclusive groups. Enaliarctos* Mitchell and Tedford TYPE SPECIES.?Enaliarctos mealsi Mitchell and Tedford, 1973. INCLUDED SPECIES.?Enaliarctos mealsi Mitchell and Ted? ford, 1973, E. mitchelli Barnes, 1979, and three new species: E. tedfordi, E. emlongi, and E. barnesi. DEFINITION.?Enaliarctos* is a metataxon composed of the common ancestor of Enaliarctos* and all the species listed above. Enaliarctos* possesses the diagnostic synapomorphies of the Pinnipedimorpha but lacks all of die synapomorphies of die next most exclusive clade, the Unnamed Taxon that includes Pteronarctos and all other pinnipeds. REVISED DIAGNOSIS OF GENUS.?The following attributes retained by Enaliarctos are primitive relative to those of odier pinnipeds: deep embrasure pit between P* and M ;^ P^-M^ with multiple roots; P^ with shelf-like protocone; Mj widi well- developed cusps on the trigonid and a hypoconid present on die talonid; six lumbar vertebrae; ulna lacking die posteriorly expanded olecranon process; radius lacking a strongly flattened and expanded distal end; fifdi intermediate phalanx of die manus unreduced; femur with well-developed pit on the head for the teres femoris ligament; metapodial shafts rounded in cross section with keeled heads and strongly trochleated phalangeal articulations. COMMENTS.?Because diere are no unequivocal synapomor? phies that diagnose Enaliarctos* I have recognized this taxon as a metataxon. Enaliarctos* was originally diagnosed by characters that now appear to be apomorphic at a more general level than was evident when Mitchell and Tedford's (1973) original description appeared. Supplemental descriptions by Bames (1979) do not provide any odier characters that can be interpreted as synapomorphies. Enaliarctos emlongi, new species RGURES 3-10 DIAGNOSIS.?Distinguished from other species of En? aliarctos by having P^ with broadly developed posterolingual shelf, P2-^ with large crest-like metacone, F^_4 metaconid small, Mj metaconid very reduced and close to protoconid. Distinguished from all other species except E. mitchelli in having posterior portion of zygoma that joins palate anterior to M .^ Distinguished from E. barnesi and E. mealsi in having reduced cheek teeth cingula, P* with a short metacone. TYPE MATERIAL.?Holotype: USNM 250345 (Emlong field no. E-53), a nearly complete cranium and jaws. Probably from the same individual, there is a left femur and right patella, tibia, and fibula (Emlong field no. E-70). Collected by Douglas Emlong, fall, 1961. ETYMOLOGY.?The species is named in honor of Douglas Emlong for a lifetime devoted to the collection of fossd marine mammals from Oregon and his recognition of their value to science. TYPE LOCALITY.?Cranium and jaws found in place in concretionary bank about 2 ft [60 cm] above beach level and 50 ft [15 m] south of highest part of concretionary bluff, and postcranial elements found 5 ft [1.5 m] from skull, approxi? mately 74 mi [400 m] south of Big Creek, Lincoln County, Oregon. SEV4, SWV4, Sec. 32, T 10 S, R. 11 W, Yaquina Quadrangle, 15 minute series, USGS: 44?39'17"N, 124?04'42"W. HORIZON.?^Just north of Astoria Formation-Nye Mudstone contact. According to Emlong's field notes this locality is in the Astoria Formation; Repenning (pers. comm.) suggests it may in fact be from the Nye Mudstone. REFERRED SPECIMENS.?USNM 314540 (E-76-38), crushed snout and palatal portion of cranium with right P, Cs , left P ,^ and M^ collected by Douglas Emlong on 22 May 1976. According to Emlong's field notes this specimen was collected in place in bedrock (Nye Mudstone) at floor of beach, 75 ft [23 m] out from bank, ^3 mi [1.07 km] south of Big Creek, Lincoln County, Oregon. NWV4, NWV4, Sec. 5, T 11 S, R. 11 W, Yaquina Quadrangle, 15 minute series, USGS: 44?38'58"N, 124?03'36"W. USNM 314290 (E-71-58), cmshed skull widi left P^, P ,^ and M^ collected by Douglas Emlong on 30 November 1971. According to Emlong's field notes this specimen was collected in a fine-grained sandstone layer that altemates widi coarser grey sandstone containing large concretions between typical Nye Mudstone and Yaquina Formation, approximately 74 mi [400 m] soudi of Beaver Creek, Lincoln County, Oregon. NW74, Sec. 19, T 12 S, R. 11 W, Yaquina Quadrangle, 15 minute series, USGS: 44?31'00"N, 124?,04'42'^. DESCRIPTION.?^The pattern of suture closure (see Sivertson, 1954) in the holotype indicates that it is an adult. The rostrum is broad and low and the narial opening is ovoid (Figure 3; Tables 2-4 ). The nasals are relatively long and narrow. The fusion of sutures has obscured the frontal-nasal contact The nasolabialis fossa is well developed and positioned on the maxilla above the antorbital rim (Figure 3). The posteroventral margin of this fossa forms a strong antorbital ridge whereas the anterodorsal margin is produced into a thickened protuberance. The narrow interorbital region is marked by small, rounded supraorbital ridges (Figure 3). The sagittal crest begins just SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY FIGURE 3.?Enaliarctos emlongi, new species, USNM 250345, holotype, skuU: A, dorsal view; B, ventral view; C, lateral view. (Scale = 2 cm.) NUMBER 69 TABLE 2.?Cranial measurements (mm) of Enaliarctos from coastal Oregon (e = estimated). Measurements Cranium length (CBL) Greatest width at anterior nares Width across antorbital processes Width across greatest interorbital constriction Width across supraorbital processes Width of braincase al anterior edge of glenoid fossa Postpalatal length (postpalatal notch to basion) Width of rostrum across canines Zygomatic width Auditory width Mastoid width Width of zygomatic root of maxilla Basion to anterior edge of zygomatic root Width of palate across anterior root of P" Width between infraorbital foramen Greatest width across occipital condyles Greatest width of foramen magnum Greatest height of foramen magnum Transverse diamenter of infraorbital foramen Length of toothrow, C-M^ E. emlongi USNM 250345 228.00 32.30 42.40 22.26 28.97 66.59 100.82 55.43 126.00= 87.79 103.14 16.36 152.61 56.36 59.31 51.12 21.08 20.36 10.86? 78.18 E. barnesi USNM 314295 37.18' 56.30 28.99 40.13 58.39 14.92 49.75= 56.95= 9.83 69.27= E. tedfordi USNM 206273 43.16 21.06 28.16 63.82 99.99 123.70 91.85 105.84 15.32 142.90 51.97 58.95 53.72 25.67 16.03 9.70 E. mitchelli USNM 175637 172.06 30.08 38.24 18.83 22.86 68.59= 76.64= 39.63= 91.37= 68.00= 13.47 117.49 43.22 44.81 24.63 18.62 10.23 63.56= TABLE 3.?^Measurements (mm) of the upper dentition of Enaliarctos from coastal Oregon (a - alveolus only; e - estimated). E. emlongi USNM 250345 5.42 2.69 5.82 3.33 7.57 5.40 10.65 7.65 8.35 5.74 11.72 5.80 12.10 7.13 12.15 9.87 6.86" 9.04" 4.15 3.92 E. barnesi USNM 314295 5.01 3.09 6.90 5.00 10.74 10.74 8.52 6.18 11.05 6.12 10.61 7.06 11.94 9.01 7.04 8.62 3.33" 2.45' E. tedfordi USNM 206273 7.36" 5.25' 11.51 8.64 5.20" 7.90" E. mitchelli USNM 175637 7.50= 7.50= 6.25" 3.88' 8.74 4.50 8.90 5.26 9.90 7.08 7.31 4.83 3.39' 2.00' length width length width length width length width length width p2 1^"S^^ width length width p4 length width length M width length ^ width SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY TABLE 4.?^Measurements (mm) of the mandible and lower dentition of Enaliarctos from coastal Oregon. Length of ramus Length of toothrow C-M2 Depth of ramus below P3 length 1^ width length ^ width length 3^ width length width length ^1 width length ^2 width length ^3 width length ^4 width length H width length ?^ width E. emlongi USNM 250345 165.58 80.09 24.09 4.55 5.07 11.49 7.33 8.44 4.72 11.25 5.05 12.50 5.85 12.60 5.73 12.53 5.98 3.60 2.95 E. barnesi USNM 314295 70.55 26.03 9.53 4.63 11.19 5.46 11.91 5.13 12.12 5.70 behind the supraorbital ridges and extends as a low ridge back to the lambdoidal crest. The posterior surfaces of the braincase are elevated above the anterior surface and the parietals are raised into a prominent lambdoidal crest. A pair of deep pseudosylvian sulci are obliquely positioned on die lateral surface of the braincase above the'squamosal (Figure 3). The exoccipital portion of the occipital shield is broad and low. A short median supraoccipital crest separates the dorsal portion of the shield. Beneath die lambdoidal crest the shield is concave becoming convex above the nearly circular foramen magnum. The zygomatic arch is less bowed laterally and less curved dorsally dian in E. mealsi (Mitchell and Tedford, 1973, fig. 6). The vend-al portion of die arch is slightly curved anterodorsally where it rises from the side of the skull. The posterior border of die zygomatic arch joins die palate between M^ and M .^ The maxillary-jugal suture is fully closed and cannot be traced. The narrow tip of the zygomatic process of the squamosal contacts die base of die postorbital process of die jugal in a splint-like arrangement. Large postglenoid processes are present. A small postglenoid foramen is positioned at die lateral termination of a transverse groove across the postglenoid process. The entire orbital region is mediolaterally compressed. The orbital wall is solid with no vacuities. Sutures in this region are not preserved. There is no indication of a lacrimal foramen or bone nor a pit for the inferior oblique muscles. The large infraorbital canal is nearly circular in outiine. The spheno? palatine foramen is relatively large and ovoid. The smaller posterior palatine foramen is positioned anterior and slightly ventral to it. The palate is slightiy arched dorsally for its entire length (Figure 3). The incisive foramina have large, paired palatal openings separated by a narrow crest of bone. The anterior palatine foramina and dieir associated sulci originate medial to P^ and terminate medial to P .^ Several smaller posterior foramina are on either side of the palate. Small, rounded palatine processes of the maxilla extend a short distance behind M .^ The posterior border of the palate is U-shaped and a very small median tuberosity projects slightiy posteriorly. The palatine, alisphenoid, pterygoid strut between die palate and die braincase is inflated with a rounded, convex lateral margin (Figure 3). Most of the pterygoid hamulus is broken off. The basioccipital is rectangular and characterized by having a horizontal flange that abuts the entotympanic. Located posteri? orly and separated by a thick median crest are a pair of very deep depressions for the rectus capitis muscles. Positioned at die anterior end of these depressions are a pair of small tuberosities. The presence of similar tuberosities in E. mealsi, Pinnarctidion bishopi, and Pteronarctos goedertae were noted by Barnes (1979:26, 1989:9), who discovered that dorsal to these tuberosities are matrix filled cavities in the basioccipital. Bames suggested that these cavities may be homologous with the embayment of the inferior pefrosal venous sinus for a loop of the intemal carotid artery seen in ursids and amphicyonids (see Hunt, 1974a,b, 1977). A more detailed evaluation of this character and its distribution is currently in progress (Berta, in prep.). A pair of ellipsoidal hypoglossal foramina are posi? tioned along the posterolateral margin of die basioccipital. The tympanic bullae are large, flask-shaped, and medially inflated (Figure 3). A steep crest of the bulla forms a groove that joins the bulla to die postglenoid process. Posterolaterally, the bulla is fused to the mastoid and paroccipital processes. The mastoid process is considerably larger than the paroccipital process and connected to it by a very low, broad ridge (Figures 3, 4). The paroccipital process is short, blunt, and directed posterovenfrally. Several small foramina pierce the surface of diat portion of the bulla diat forms die tubular external auditory meatus. A flange of the bulla appears to have separated die median lacerate foramen from the eustachian tube aldiough much of die flange is broken away. The posterior opening of die carotid canal is confluent widi the enlarged jugular foramen. The fossa for the tympanohyal is located medial and slighdy posterior to the large, deep, circular stylomastoid foramen. A very thin flange of bone separates diem. The venfral wall of die bulla on die right side of the holotype was removed to expose the middle ear cavity (Figure 4). The venfrolateral wall of the carotid canal creates a shelf in the medial wall of the bulla. The tympanic crest is considerably NUMBER 69 FIGURE 4.?Enaliarctos emlongi, new species, USXM 250345, holotype, ventral view of right side of basicranium with ventral wall of bulla removed. (Abbreviations: ac = posterior opening of ahsphenoid canal; cc = ventral wall of carotid canal; er = epitympanic recess; et = groove for eustachian tube; fo = foramen ovale; hf = hypoglossal foramen; mc = septum of musculotubarius canal; p = promontorium; pc = posterior op)enmg of carotid canal; pf = postglenoid foramen; pif = posterior lacerate foramen; rw = round window; sf = stylomastoid foramen; tc = tympanic crest; if = tympanohyal fossa; llf = tensor tympani fossa. Scale is approximately 5 mm.) 10 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY larger than in E. mealsi and it projects farther into the tympanic cavity. In addition, the tympanic cavity differs from that of E. mealsi in its farther posterior extension. The pear-shaped promontory is broader mediolaterally than in E. mealsi. The round window is posteriorly oriented and the smaller oval window faces laterally. Lateral to die promontory is a deep, elliptical fossa for the tensor tympani. Posterolateral to the fossa for the tensor tympani is the deep, nearly circular epitympanic recess. A shallow groove marks the anterior margin of this fossa and leads to the groove for the eustachian tube; its lateral margin defined by the septum of die musculotubarius canal. Several smaller grooves are developed on the anterior face of the promontory. The lateral groove, the largest, probably housed the intemal carotid artery. A small opening, presumably the promontory foramen is visible in die posteromedial wall of die bulla. The malleus and incus were recovered from die holotype in excellent condition (Figure 5). The muscular process is absent from the malleus. The anterior portion of the mallear head is relatively large and it bears a deep concavity. Although incomplete, it is apparent that a thin anterior lamina extends from the anterior process to the base of die neck. The manubrium is thin and relatively short. The incus is large and slightiy inflated with a short, broad stapedial process and a long, thin posterior cms. I^ "3 are arranged in a broad transverse arc. I^ "^ are distinguished by having a fransverse groove that extends across die crown creating die effect of a double cusp (Figure 6). I^ is larger with a nearly cfrcular crown. Deep vertical wear facets are developed on the lateral margins of the crown. RGURE 5.?Enaliarctos emlongi, new species, USNM 250345, holotype, malleus and incus. (Scale = 1 mm.) The canine is relatively large and slighdy recurved. On die posterior side of the crown is a vertical crest diat extends medially from the vertical crest to its termination midway around the lingual face of the tooth. r w 'etween P^-M', shaUow-absent 35. Pit for tensor tympani reduced-absent 36. Processus gracUis and anterior lamina of malleus reduced 37. P^ enlarged protocone shelf absent 38. Narrow contact between nasals and premaxUla 39. Ascending process of premaxUla-maxiUa dips into nasal aperture 40. Canals for CN VII and VIII widely separate 41. Auditory bulla underlaps basioccipital 42. Petrosal visible through posterior lacerate foramen 43. Caudal entotympanic inflated 44. Basioccipital short, broad, widened posteriorly 45. Reduced coronoid process 46. P caniniform with circular cross-section 47. P lingual cingulum absent 48. Nasals terminate posterior to frontal-maxUlary contact 49. Posteriorly wide palate 50. Squamosal-jugal, mortised contaa 51. Flange below ascending ramus of mandible 52. Lower premolars with large paraconids u OJ .y i; c: ^ '-5 -S .S 2 --D i i i O C L , O a , a , Q - < ; 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 7 1 1 7 1 1 0 1 7 1 1 1 1 7 1 7 1 1 1 1 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 3 1 0 1 1 1 1 1 1 1 0 1 1 2 2 1 1 0 0 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 2 1 1 1 1 1 1 1 1 1 0 1 1 2 2 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 2 1 1 2 1 1 0 1 1 1 2 1 0 2 0 0 0 0 0 7 1 0 1 1 1 1 1 1 0 1 7 1 I 1 1 7 1 7 7 ? ? 1 1 7 1 1 1 1 ? 7 0 0 0 1 0 1 0 0 0 7 0 0 0 0 0 7 7 0 0 0 0 7 2 2 0 1 1 1 1 1 1 1 1 1 1 1 0 c 1 1 7 1 ? 1 1 1 1 1 1 1 7 7 ? 1 0 c 7 1 7 ? 7 7 2 ? ] 7 ( ? " 1 ] 7 ] 0 : 7 7 7 7 1 0 ( 1 1 ? ' 0 7 7 2 : 1 1 1 7 7 7 7 7 1 1 ? 7 2 2 2 1 2 1 1 1 0 NUMBER 69 29 PINNIPEDIMORPHA PINNIPEDIA rs ida e =) to s na lia rc LU ro ta s te ra na 0. a> ta riid a O ida e do be n O CO Ha de s '^ op ho c es m at Q tid ion inn ar c CL Q> ho cid a a. (2Z.43') (1',7',S,27',46') (1,9,23,25,39,40',41, 45,48,49,50',51) 4 (24,38.40,42-44,47) 3 (2,7',17,31,32,33,35,37) 2 (20'.21',23,26,27,29,30,34,36) 1 (1,3-8,10-22,28) FIGURE 22.?Higher level relationships of Enaliarctos and "enaUarctids." Numbers refer to derived characters listed in Table 7 and discussed in text. Characters marked with a tick (') designate a multistate character. Postulated reversals lo ancestral character states are underlined. larger than in terresfrial camivorans (Wyss, 1987, table 1). 17. Pseudosylvian sulcus strongly developed. As described by Mitchell and Tedford (1973:237) in Enaliarctos mealsi die pseudosylvian (= sylvian) fossa is enlarged to a broad and deep crease down the side of the brain. Judging from the skull and endocranial cast of Cephalogale figured by de Beaumont (1965) the pseudosylvain sulcus does not appear to be sfrongly developed. According to Bames (1989, fig. 9) Pteronarctos can be distinguished from Enaliarctos by its shallower pseu? dosylvian sulcus. Comparisons with additional specimens of Pteronarctos from die Emlong collection (Berta, in prep.) indicate that this is not die case. The presence of a strongly developed sulcus is recognized here as having evolved in pinnipedimorphs ancesfrally and having subsequendy been lost in pinnipeds. 19. M^'^ small relative to premolars. The primitive pin? niped dentition as exemplified by Enaliarctos is characterized by having M "^^ reduced relative to the premolars. As commented by Mitchell and Tedford (1973:251), "the degree of reduction of the upper molars in Enaliarctos compared with die premolars is greater dian diat of any known early arctoid." In all later pinnipeds the molars and premolars are subequal. 20. MJ entoconid reduced-absent. An entoconid is present in primitive ursids. This cusp is reduced to a crest in Enaliarctos as noted by Mitchell and Tedford (1973) and suppressed in all later pinnipeds. 21. MJ hypoconid reduced-absent. According to Mitchell and Tedford (1973) Enaliarctos can be distinguished from Cephalogale in having the hypoconid centrally positioned and separated from the trigonid by a broad groove. This cusp is suppressed in all later pinnipeds. 22. MJ absent. Among primitive lu-sids, Amphicynodon, Pachycynodon, Allocyon, and Cephalogale, M3 is present This tooth is absent in Enaliarctos and later pinnipeds. 28. M^'^ cingulum reduced-absent. In the primitive condi? tion, seen in Cephalogale and amphicyonids, an enlarged cingulum encircles the upper molars. This cingulum is reduced or absent in Enaliarctos and all later pinnipeds (Mitchell and Tedford, 1973). The polarity of five other features, although unknown in Enaliarctos, have been assigned a most parsimonious distribu? tion at this node. I treat three of these features, enlarged basal whorl of scala tympani (SYNAPOMORPHY 11), middle ear cavity and external auditory meatus presumably lined with distensible cavernous tissue (SYNAPOMORPHY 16), and deciduous dentition reduced (SYNAPOMORPHY 18), as apomorphies of the Pinnipe- dia clade, die minimum level at which observation confirms their distribution. Two additional features, maxilla makes a significant conu-ibution to die orbital wall (SYNAPOMORPHY 3), and lacrimal fuses early to maxilla, greatly reduced or absent; does not contact jugal (SYNAPOMORPHY 6), are assigned as apomorphies at the level of the Unnamed Clade that includes Pteronarctos and all other pinnipeds. One additional character hsted by Wyss (1987) as a potential pinniped synapomorphy 30 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY (large cochlear aqueduct) requires further comparative study and measurement before it can be definitively assigned at this or any other level (Wyss, pers. comm.). In addition to cranial and dental features the pinnipedi? morphs are distinguished by a large number of synapomorphies of the postcranial skeleton reviewed elsewhere (Berta et al., 1989; Berta and Ray, 1990). Derived features of die forelimb include short, robust humems widi sfrongly developed del? topectoral crest and enlarged tuberosities; elongation of digit I in the manus and digits I and V in the pes; loss of supracondylar foramen on humems ( a reversal in phocines). Shared derived features of the hindlimb include short ilium; extremely short anteroposteriorly flattened femur and medially inclined con? dyles; large, broadly developed greater frochanter on the femur and conical patella. In the axial skeleton, pinnipedimorphs share with all arctoids except ursids possession of an alar notch radier dian a foramen on die adas. The next more restrictive clade is an unnamed group that includes the fossd taxon Pteronarctos and all other pinnipeds and is defined by nine cranial characters (Figure 22, node 2; Table 7). Because relationships among more inclusive pinniped clades were not die principal focus of diis study, descriptions of characters as were provided in the preceding section for diagnosis of Pinnipedimorpha are not detailed here. Pter? onarctos and all odier pinnipeds are characterized dentally by a reduction in the number of roots on the upper molars (SYNAPOMORPHY 27). The upper molars of Enaliarctos and ursids have three separate roots. This number is reduced to two in Pteronarctos and phocids and one in all other pinnipeds. The Pinnipedia clade is defined by 8 cranial characters (Figure 22, node 3). The fossa muscularis (SYNAPOMORPHY 32), prominent in Pteronarctos, Enaliarctos, and ursids, is absent in members of this group. This character and its proposed fransformation warrant a few additional comments. As defined by Davis (1964:49) immediately behind die lacrimal fossa is a shallow pit, the fossa muscularis, in which the inferior oblique muscle of the eye arises; the thin floor of diis pit is usually broken through on dry skulls, and dien resembles a foramen. This fossa in Ursus is relatively enormous, as large as the lacrimal fossa. Examination of die skull of Cephalogale minor (Vienna Natural History Museum, Vienna A4445) reveals the presence of a slight depression behind the lacrimal fossa delimited by a venfraUy floored ridge. A possible interpretation is diat the condition in Cephalogale might be an intermediate stage in development of the deep, posteriorly positioned fossa seen in Enaliarctos and Pter? onarctos. Following diis fransformation sequence, odier pin? nipeds are hypothesized to have lost this fossa. As noted by Wyss (1987:16) another pinniped synapomor? phy is "die presence of of an unossified space (often termed orbital vacuity) in die venfral wall near the juncture of die frontal, maxilla, and palatine bones...associated with die pinniped configuration of the maxilla" (SYNAPOMORPHY 33). The derived pinniped palate lacks an embrasure pit on die palate between P^ and M^ to receive the crown of the lower camassial, a conspicuous feature in Pteronarctos, Enaliarctos, and terresfrial camivorans (SYNAPOMORPHY 34). This character together with a number of dental modifications including reduction or loss of F* protocone (SYNAPOMORPHY 37), loss of MJ metaconid, entoconid, and hypoconid (SYNAPOMORPHIES 20, 21, 30), and loss of M2 (SYNAPOMORPHY 31) indicate suppression of the camassial function of these teedi, which is viewed as a frend toward homodonty exemplified by modem pinnipeds. Synapomorphies of the pinniped postcranial skeleton (see also Berta et al, 1989; Berta and Ray, 1990) include, for die forelimb, flattening and posterior expansion of the olecranon process of the ulna and sfrong reduction of fifdi intermediate phalanx of the manus. Pelvic girdle and hindlimb spe? cializations include unfused pubic symphysis, loss of pit for ligament on femoral head, reduction or loss of frochanteric fossa on femur, and reduction of processus lateralis on die fibula. Both the manus and pes display hinge-like phalangeal articulations, flattened phalanges, and non-trochleated interphalangeal articulations. All pinnipeds (except living walruses) are characterized by having five lumbar vertebrae. Monophyly of the Otariidae (= Otariinae Mitchell, 1968, 1975; Barnes, 1989) has fraditionally been accepted (Repen? ning and Tedford, 1977; Berta and Demere, 1986; Barnes, 1989). This analysis identified only a single character reversal, the presence of large shelf-like supraorbital processes as uniting otariids (SYNAPOMORPHY 7). Relationships among die Otariidae are discussed further elsewhere (Berta and Demere, 1986; Berta, in prep.). As proposed by Wyss (1987) odobenids and phocids and their fossil allies are sister taxa (Figure 22, node 4). Cranial features that support this clade include reduced contact between the premaxilla and nasals (SYNAPOMORPHY 38), and pefrosal visible through posterior lacerate foramen (SYNAPOMORPHY 42). In addition, there are numerous derived features of the ear region diat are more fully described by Wyss (1987). Among postcranial synapomorphies defined by Wyss in the previously mentioned paper are development of an at least incipientiy caudally directed process on the asfragalas, short calcaneal tuber, and enlarged baculum. A more inclusive clade of pinnipeds including Pin? narctidion, Desmatophoca, Allodesmus, and Phocidae is united by 12 derived cranial features (Figure 22, node 5). These pinnipeds display a mortised contact between the squamosal and jugal (SYNAPOMORPHY 50), posteriorly wide palate (SYNAPOMORPHY 49), and nasals that terminate posterior to die frontal-maxillary contact (SYNAPOMORPHY 48). The lower jaw has a reduced coronoid process (SYNAPOMORPHY 45) and a mandibular flange developed below the ascending ramus (SYNAPOMORPHY 51). Relationships among these taxa are represented in Figure 22, in which Desmatophoca + Pinnarc- NUMBER 69 31 tidion + Allodesmus form an unnamed clade widi Desmato? phoca and Pinnarctidion linked as sister taxa, Pinnarctidion and Desmatophoca possess a squamosal-jugal articulation that is less modified than diat displayed by Allodesmus and phocids in which diis contact is dorsoventrally expanded (SYNAPOMOR? PHY 50). These taxa also lack development of large orbital vacuities (SYNAPOMORPHY 33) that are present in both Allodesmus and phocids. The Phocidae are monophyletic (for a recent systematic freatment see Wyss, 1988b). The above analysis clearly shows that "enaliarctids" are paraphyletic. Enaliarctos* is the sister group of the remaining pinnipeds, and Pteronarctos and Pinnarctidion are placed in less inclusive clades. This is in contrast to the traditionally accepted phylogeny of "enaliarctids" as basal otariids involved only in the ancesfry of some pinnipeds (sea lions, walruses, and their fossil allies (desmatophocids and allodesmids), but not true seals) (Bames, 1979, 1989). According to Bames (1989:19) the "Enaliarctinae are a horizontal group,...a grade rather than a clade." Recognition of the paraphyletic grade "Enaliarctinae" is an inappropriate systematic procedure. Such an approach is undesirable for many reasons, die most important of which is that it misrepresents the evolutionary history of pinnipeds. If one accepts the evidence reviewed here, then the phylogenetic arrangement of diese taxa should reflect that knowledge. By grouping Pteronarctos and Pinnarctidion in the "Enaliarctinae" and arguing for "die taxonomic utility [of doing so], considering the still comparatively meager state of our knowledge of early otariid evolution," Bames (1989:19) obscures what we do know about the interrelationships of these taxa. Conclusions Three new species of the archaic pinniped Enaliarctos* from the marine late Oligocene and early Miocene (Arikareean and Hemingfordian or early Barstovian correlatives) of coastal Oregon are described based on well-preserved crania and dentitions: E. barnesi (near Yaquina Formation-Nye Mudst? one contact), E. emlongi (Nye Mudstone or Astoria Formation), and E. tedfordi (Yaquina Formation). An additional skull is referred to a previously described species, E. mitchelli (Nye Mudstone). These taxa differ principally in facial morphology and details of the dentition. Phylogenetic analysis of cranial and dental characters among these species and die genotype E. mealsi results in recognition of Enaliarctos as a metataxon, since monophyly of this taxon cannot be unambiguously determined. Relationships among Enaliarctos species are largely unresolved aldiough there is some character evidence to support die common ancestry ofE. tedfordi, E. emlongi, and E. mitchelli. Phylogenetic analysis of 52 cranial and dental features among fossil taxa {Enaliarctos, Pteronarctos, Desmatophoca, Allodesmus, and Pinnarctidion) and extant pinniped families (Otariidae, Odobenidae, and Phocidae), supports die view that the "Enaliarctinae" (= "Enaliarctidae") is paraphyletic as well as the recognition that Enaliarctos* is the sister taxon of all other pinnipeds. Other taxa previously included in the "Enaliarctidae" {Pteronarctos and Pinnarctidion) show differ? ent relationships: Pteronarctos + all other pinnipeds form one monophyletic clade and intemested within diis clade are several other clades, one of which includes Pinnarctidion + Desmatophoca, Allodesmus, and the Phocidae. Literature Cited Addicott, W.A. 1976. Neogene MoUuscan Stages of Oregon and Washington. In A.E. Fritsche, H. Ter Best, Jr., and W.W. Womardt, editors, The Neogene Symposium, pages 95-116. San Francisco: Society of Economic Paleontologists and Mineralogists [Armual Meeting, Pacific Section]. Armentrout, J.M. 1981. Correlation and Ages of Cenozoic Stratigraphic Units in Oregon and Washington. Geological Society of America Special Paper, 184: 137-148. Armentrout, J.M., D.A. HuU, J.D. BeauUeu, and W.W. Rau, editors 1983. Correlation of Cenozoic Stratigraphic Units of Western Oregon and Washington. Oregon Department of Geology and Mineral Industries Oil and Gas Investigations, 7: 90 pages, 1 chart. Barnes, L.G. 1979. Fossil Enaliarctine Pinnipeds (MammaUa: Otariidae) from Pyramid HUl, Kern County, California. Contributions in Science. Natural History Museum of Los Angeles County, 318:1-41, figures 1-22, tables 1-4. 1987a. An Early Miocene Piimiped of the Genus Desmatophoca (Mam? maUa: Otariidae) from Washington. Contributions in Science. Natural History Museum of Los Angeles Couniy, 382:1-20, figures 1-9, 1 table. 1987b. Aetiocetus and Chonecetus, Primitive OUgocene Toothed Mysticetes and the Origin of Baleen Whales. [Abstract.] Journal of Vertebrate Paleontology, 7 (Supplement to no. 3), lOA. 1989. A New EnaUarctine Pimuped from the Astoria Formation, Oregon, and a Classification of the Otariidae (MammaUa: Carnivora). Contributions in Science. Natural History Museum of Los Angeles County, 403:1-26, figures 1-9, tables 1-2. Berggren, W , D.V. Kent, J.J. Flynn, and J.A. Van Couvering 1985. Cenozoic Geochronology. Geological Society of America Bulletin, 96:1407-1418. Berta, A. In prep. [New specimens of Pteronarctos (Pinnipediformes) from the Miocene of Oregon and Its relationship to the Otariidae.] Berta, A., and T.A. Demere 1986. Callorhinus gilmorei n. sp., (Carnivora: Otariidae) from the San Diego Formation (Blancan) and its ImpUcations for Ouriid Phylogeny. Transactions of the San Diego Society of Natural History, 2l{l):lll-l26. Berta, A., and C.E. Ray 1990. Skeletal Morphology and Locomotor CapabUities of the Archaic Piimiped, Enaliarctos mealsi. Journal of Vertebrate Paleontology, 10(2):141-157. Berta, A.. C.E. Ray, and A.R. Wyss 1989. Skeleton of the Oldest Known Pinniped, Enaliarctos mealsi. Science, 244:60-62. Davis, D.D. 1964. The Giant Panda: A Morphological Study of Evolutionary Mecha? nism. Fieldiana: Zoology Memoirs, 3:1-334. De Beaumont, G. 1965. Contribution a I'etude du genre Cephalogale Jourdan (Carnivora), Schweizerische Paldontologische Abhandlungen, 82:1-34, figures 1-28. Domning, D.R, C.E. Ray, and M.C. McKenna 1986. Two New OUgocene DesmostyUans and a Discussion of Tethyth? erian Systematics. Smithsonian Contributions to Paleobiology. 59: 56 pages. Donoghue, M.J. 1985. A Critique of the Biological Species Concept and Recommendation for a Phylogenetic Alternative.Bryolog'ist, 88:172-181. Dubrovo, LA. 1981. A New SubfamUy of FossU Seals (Pinnipedia Kamtschatarctinae subfam nov.) Proceedings ofthe Academy of Sciences ofthe USSR, 256:970-974 [in Russian]. 1984. Miocene Seals from the Pacific Region. Paleontological Journal, 18(l):50-60 [in Russian]. Emlong, D. 1966. A New Archaic Cetacean from the OUgocene of Northwest Oregon. Bulletin of the Museum of Natural History. University of Oregon, 3:1-51. Hynn, J.J., N.A. Neff, and R.H. Tedford 1988. Phylogeny of the Carnivora. In M.J. Benton, editor, The Phylogen? etic Classification of the Tetrapods. Volume 2 (Mammals), pages 73-116. Oxford: Qarendon Press. Gauthier, J.A. 1986. Saurischian Monophyly and the Origin of Birds. In K. Padian, editor, The Origin of Birds and the Evolution of FUght. Memoirs of the California Academy of Sciences, 8:1-55. Gauthier, J.A., R. Estes, and K. de Quieroz 1988. A Phylogenetic Analysis of Lepidosauromorpha. In R. Estes and G. PregUl, editors, Phylogenetic Relationships of Lizard Families: Essays Commemorating Charles L. Camp, pages 15-98. Stanford Univ. Press. Hunt, R.M., Jr. 1974a. The Auditory BuUa in Carnivora: An Anatomical Basis for Reappraisal of Carnivore Evolution. Journal of Morphology, 143(l):21-76. 1974b. Daphoeniclis, a Cat-Uke Carnivore (MammaUa, Amphicyonidae) from the Oligcxene of North America. Journal of Paleontology, 48(5): 1030-1047. 1977. Basicranial Anatomy of Cynelos Jourdan (MammaUa: Carnivora), an Aquitaiuan Amphicyonid from the AlUer Basin, France. Journal of Paleontology, 51(4):826-843. King, J.E. 1983. Seals of the World. Second edition, 240 pages. Ithaca, New York: ComeU University Press. Maddison, W.P, M.J. Donoghue, and D.R. Maddiscwi 1984. Outgroup Analysis and Parsimony. Systematic Zoology, 33:83-103. McLaren, LA. 1960. Are the Pinnipedia Biphyletic? Systematic Zoology. 9:18-28. Merriam, C.W. 1930. Allocyon, a New Canid Genus from the John Day Beds of Oregon. University of California Publications in Geological Sciences, 19:229-243, figures 1-5, plates 35, 36. MitcheU, E.D. 1968. The Mio-Pliocene Pinniped Imagotaria. Journal of the Fisheries Research Board of Canada, 25:1843-1900. 1975. ParalleUsm and Convergence in the Evolution of Otariidae and Phocidae. Rapports et Proces-verbaux des Reunions, Conseil International pour I'Exploration de la Mer, 169:12-26. MitcheU, E., and R.H. Tedford 1973. The EnaUarctinae: A New Group of Extinct Aquatic Carnivora and a Consideration of the Origin of the Otariidae. Bulletin of the American Museum of Natural History, 151(3):201-284, figures 1-27 tables 1-4. 32 NUMBER 69 33 Mivart, G. 1885. Notes on the Pinnipedia. Proceedings of the Zoological Society of London, 1885:484-500. Muizon, Ch. de 1982. Les relations phylogenetiques des Lutrinae (MusteUdae, Mam? maUa). Geobios. Memoire Special. 6:259-277. Munthe, J., and M.C. Coombs 1979. Miocene Dome-skuUed ChaUcotheres (MammaUa, Perissodactyla) from the Western Uruted Slates: A PreUminary Discussion of a Bizarre Structure. Journal of Paleontology, 53(1):77-91. Repenning, C A . 1976. Adaptive Evolution of Sea Lions and Walruses. Systematic Zoology. 25(4):375-390. Repenning, C.A., and R.H. Tedford 1977. Otarioid Seals of the Neogene. Professional Papers of the United Stales Geological Survey, 992: 93 p)ages. Sivertson, E. 1954. A Survey of the Eared Seals (FamUy Otariidae) with Remarks on the Antarctic Seals CoUected by M/K "Norvegia" in 1928-1929. Det Norske Vedenskaps-Akademii Oslo, 36:1-76. Snavely, P.D., Jr., N.S. MacLeod, H.C. Wagner, and D.L. Lander 1980. Geology of the West-central Part of the Oregon Coast Range. In Geologic Field Trips in Western Oregon and Southwestem Washington, Bulletin of the Oregon Department of Geology and Mineral Industries, 101:39-58. Snavely, RD., Jr., N.S. MacLeod, H.C. Wagner, and W.W. Rau 1976. Geology of the Yaquina and Toledo Quadrangles, Oregon. United States Geological Survey MisceUaneous Investigations Series Map 1-867, scale 1 : 62,500. Snavely, RD., Jr.. W.W. Rau. and H.C. Wagner 1964. Miocene Stratigraphy of the Yaquina Bay Area. Newport, Oregon. OreBtn, 26(8):133-151. Swofford, D.L. 1989. Phylogenetic Analysis Using Parsimony (PAUP), Version 3D. Champaign: DUnois Natural History Survey. Tedford. R.H. 1976. Relationship of Pinnipeds to Other Carnivores (MammaUa). System - alic Zoology. 25(4):363-374. Wozencraft, C. 1989. The Phylogeny of the Recent Carnivora. In John L. Gittleman, editor. Carnivore Behavior, Ecology, and Evolution, pages 495- 535. New York: ComeU University Press. Wyss. A.R. 1987. The Walrus Auditory Region and Monophyly of Pinnipeds. American Museum Novitates, 2S71:1-31, figures 1-7, tables 1-2. 1988a. Evidence from FUpper Structure for a Single Origin of Pinnipeds. Nature, 334:427-428. 1988b. On "Retrogression" in the Evolution of the Phocinae and Phylogen? etic Affinities of the Monk Seals. American Museum Novitates, 2924: 38 pages. REQUIREMENTS FOR SMITHSONIAN SERIES PUBLICATION Manuscripts intended for senes publication receive sut>stantive review (conducted by their originating Smithsonian museums or offices) and are sut}mitted to the Smithsonian Institution Press with Fonm SI-36, which must show the approval of the appropriate authority designated by the sponsoring organizational unit. Re? quests for special treatment?use of color, foldouts, case-bound covers, etc.?require, on the same form, the added approval of the sponsoring authority. Review of manuscripts and art by the Press for requirements of sehes format and style, completeness and danty of c(^y, and arrangement of all matenal, as outlined below, will govern, within the judgment of the Press, acceptance or rejection of manuscripts and art. Copy must be prepared on typewriter or word processor, douttle-spaced, on one side of standard white bond paper (not erasable), with 1%" margins, submitted as rit>bon copy (not carbon or xerox), in loose sheets (not stapled or bound), and accompanied by onginal 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 ttie established format; table of contents with indents reflecting the hierarchy of heads in the paper: also, foreword and/or preface, if appropriate. First page of text should carry ttie title and author at the top of the page: second page should have only the author's name and professional malKng address, to be used as an unnumbered footnote on the first page of printed text. Center heads of whatever level should be typed with initial caps of major words, with extra space above and below the head, but no other preparation (such as aH caps or underline, except for the underlir^ necessary for generic ar\d specific epithets). Rurt-in paragraph heads should use period/dashes or colons as neces? sary. Tabulations within text (lists of data, often in parallel columns) can be typed on the text page where ttwy occur, Ijut they should not contain rules or numbered table captions. Formal tat)les (numbered, with captions, boxheads, stubs, rules) should be submitted as carefuUy typed, double-spaced copy separate from the text; they will be typeset unless ottierwise requested, if camera-copy use is anticipated, do not draw rules on manuscript copy. Taxonomic keys in natural history papers should use the aMgned-couptet form for zoology and may use the multi-level indent form for t>otany. If cross referencing is required t>etween key and text, do not include page references within the key, but number the keyed-out taxa, using the same numbers with their comrapondir^ heads in the text. Synonymy in zook>gy must use the ^lort form (taxon, author, year:page), with full refo'ence at the end of the paper under "Uterature Cited." For botany, the kxig form (taxon, auttior, at>breviated joum^ri or book title, vokime, page, year, with no reference in "Literature Cited") is optional. Texl-retorence system (autttor, yesrrpage used within the text, with full citatk>n in "Literature Cited" at the end of the text) must be used in place of bitHkigre^k: footnotes in aH Contributions Series and is strongly recommended in the Studies Series: '(Jones, 1910:122)" or "...Jones (1910:122)." If bibliographfc footnotes are required, use the short form (author, brief title, page) with the full citation in the biblbgraphy. Footnotes, when few in number, whether annotative or bitrtk)- graphic, should be typed on separate sheets and inserted imme? diately after the text pages on which ttie references occur. Ext^- sive notes must be gathered together and placed at the end of ttie text in a notes section. Bibliography, depending upon use, is twined "Literature Cited," "References," or "Bibliography." SpeH out titles of txioks, artkdes. joumals, and monographk; senes. For book and article titles use sentence-style capitalization according to the rules of the lan? guage employed (exception: capitalize all major words in English). For journal and series titles, capitalize the initial word and ail subsequent words except articles, conjunctk>ns, and prepositions. Transliterate languages that use a non-Roman alphabet acoorc^ to the Library of Congress system. Undertine (for italics) titles of joumals and series and titles of books that are not pv t of a series. Use the parentheses/cokxi system for volume (numt)er): paginatton: "10(2):5-9." For alignment and arrangement of ele? ments, follow the format of recent publications in the series for which the manuscript is intended. GukJelines for jyeparing tjitrfi- ography may be secured fr?n Series Section. SI Press. Legends for illustrations must be sutHnitted at the end of the manusc:ript. with as many legends typed, dout>le-spaced, to a page as convenient. IHustrations must be submitted as original art (not copies) accompanying, but separate from, the manuscript. GukJelines for preparing art may be secured from Series Section, SI Press. All types of illustrations (photographs, line drawings, maps, etc.) may be intermixed throughout the printed text. They shoukj be tenmed Figures and should be numbered consecutively as they will appear in the monograph, if several iUustratkins are treated as components of a single composite figure, they shoukj be desig? nated by towercase italic letters on the illustratkx); also, in the tegend and In text references the itafic letters (underlined in coj^) should be used: "Figure 9b.' Illustrations that are intended to foHow the printed text may be termed nates, and any components should be similariy lettered and referenced: "Plate 9b." Keys to any symbols within an illustratkin shoukl appear on the art rather than in the legend. Some points of style: Do not use periods after such atHsrevi- ations as "mm, ft, USNM, NNE." SpeB out numbers "one" throu^ "nine" in expository text, txjt use digits in aH otiier cases if possit^. Use of ttie metrk: system of measurement is (deferable; vt/tiere use of the English system is unavoklat^. supply metric equiva? lents in parentheses. Use the decimal system far precise meas? urements and relationships, common fractions for af^oximatkins. Use day/month/year sequence for dates: "9 April 1976." For months in tabular Hstings or data sections, use three-letter at:4)re- viattons with no periods: "Jan. Mar. Jun," etc. C^it space between initials of a personal name: "J.B. Jones." Arrange and paginate sequentially every sheet of manu? script in the following order: (1) titie page, (2) atjstract, (3) con? tents. (4) foreword and/or preface, (5) text, (6) appendixes, (7) notes section, (8) glossary, (9) bibliography, (10) legends, (11) tatrtes. Index cqsy may t>e sut>mitted at page proof stage, but plans for an index should t>e indk:ated when manuscript is sub? mitted.