% ROM Royal Ontario Museum CIENCES ON = The Status of Smilodontopsis (Brown, 1908) and Ischyrosmilus (Merriam, 1918) A Taxonomic Review of Two Genera of Sabretooth Cats (Felidae, Machairodontinae) C. S. Churcher ROYAL ONTARIO MUSEUM LIFE SCIENCES PUBLICATIONS INSTRUCTIONS TO AUTHORS Authors should prepare their manuscripts carefully accord- ing to the following instructions; failure to do so will result in the manuscript’s being returned to the author for revision. All manuscripts are considered on the under- standing that they are not currently offered for publication elsewhere. 1. General Papers for publication are accepted from ROM staff members and research associates, and from researchers reporting on work done with ROM collec- tions. Monographs on the flora and/or fauna of Ontario may be considered for publication by authors not affiliated with the ROM. Financial contributions to- wards publication will be welcome. Authors are expected to write clearly and concisely and to omit any material not essential for an understanding of the main theme of the paper. 2. Format Manuscripts (including captions, syn- onymies, literature cited, and tables) should be typed with double space on 11’’ x 8%’’ paper with a 1%”’ margin on all sides. Three xerox copies should be submitted to the Senior Editor of the Editorial Board; the original should be retained by the author(s). The submission should include a separate sheet giving the author(s) names and affiliations, the title of the publication, the series for which it is submitted, the number of typed pages, the number of tables, and the number of plates or figures. Manuscripts should normally be organized in the following order: Con- tents, Abstract, Introduction, Materials and Methods, Results, Discussion, Conclusions, Summary (if manu- script is long), Acknowledgements, Appendices, and Literature Cited. Authors are encouraged to include foreign-language translations of the Summary, if appropriate. Main headings should be centred; sub- headings should be left-justified to the text margin. The first line of the first paragraph in each new section should not be indented. Literature citations in the text should be in the form ‘‘Jones (1972)’’ or ‘‘(Jones, 1972)’ or ‘‘(Smith, 1960:71-79, fig. 17)’’. 3. Standard Sources The primary authority on ques- tions of format and style is Guide to Authors, available from ROM Publication Services. For matters not covered in the Guide, consult CBE Style Manual (Fifth Edition). Other standard sources are as follows: for English spelling, The Concise Oxford Dictionary; for Canadian place names and coordinates, Canada Gazetteer Atlas; for the spelling of geographic names, The Times Atlas. . Abstract All papers must be preceded by a short, factual abstract, about one per cent of the text in length. The abstract may be followed by four to six key words in parentheses. . Taxonomy The name of a taxon should be given in full in headings, at the beginnings of paragraphs, and at its first occurrence in the text. Give the authority and date, if appropriate, with the first mention of each taxon, but not thereafter. Taxonomic papers, particu- larly synonymies, should follow the layout in Life Sciences Contributions beginning with No. 136. Inter- national Codes of Biological Nomenclature must be followed. . Literature Cited A complete list of references, in alphabetical order of authors, must be given at the end of the paper. When two or more works of one author are cited, they should be listed chronologically. The names of journals should not be abbreviated. For correct bibliographic form, see Life Sciences Contrib- utions beginning with No. 136. . Tables All tables should be typed on separate sheets and numbered consecutively in arabic numerals in the order of their first mention in the text. Mark the location of each table in the margin of the text. . Plates, Figures, and Text-figures Illustrations may be designated according to the conventions of the author’s discipline; in some disciplines grouped photo- graphs of scientific subject matter are commonly termed Plates, while line drawings and locality and other illustrations that occupy a full page or less are Text-figures. Usage must be consistent throughout the paper. A full-page illustration for a Contribution, with its caption, should be sized to fit an area of 17.3 X 22.75 cm; for Occasional Papers, the area is 14.1 x 21.2 cm. If captions are lengthy, they may be placed on the facing page. A scale or magnification factor should be included. Authors are reminded that when illustrations are reduced magnification factors will change, and that they are responsible for the conver- sion. For details, see Guide to Authors. LIFE SCIENCES CONTRIBUTIONS 140 The Status of Smilodontopsis (Brown, 1908) and [schyrosmilus (Merriam, 1918) A Taxonomic Review of Two Genera of Sabretooth Cats (Felidae, Machairodontinae) C. S. Churcher ROM ROYAL ONTARIO MUSEUM ROYAL ONTARIO MUSEUM PUBLICATIONS IN LIFE SCIENCES The Royal Ontario Museum publishes three series in the Life Sciences: Contributions: a numbered series of original scientific publications. Occasional Papers: a numbered series of original scientific publications, primarily short and of taxonomic significance. Miscellaneous Publications: an unnumbered series on a variety of subjects. All manuscripts considered for publication are subject to the scrutiny and editorial policies of the Life Sciences Editorial Board, and to independent refereeing by two or more persons, other than Museum staff, who are authorities in the particular field involved. LIFE SCIENCES EDITORIAL BOARD Senior editor: J. R. Tamsitt Editor: D. R. Calder Editor: J. C. Barlow External editor: C. S. Churcher Manuscript editor: J. R. Tamsitt C. S. Churcher is Professor in the Department of Zoology, University of Toronto, Toronto, Ontario, and a Research Association in the Department of Vertebrate Palaeontology, Royal Ontario Museum. Canadian Cataloguing in Publication Data Churcher, C. S., 1928- The status of Smilodontopsis (Brown, 1908) and Ischyrosmilus (Merriam, 1918) (Life sciences contributions, ISSN 0384-8159 ; 140) Bibliography: p. ISBN 0-88854-305-0 1. Smilodon — Classification. 2. Felidae, Fossil — Classification. 3. Paleontology — Quaternary. 4. Paleontology — North America. I. Royal Ontario Museum. II. Title. III. Series. QE882.C15C48 1984 569’ .74 C84-099034-0 Publication date: 30 June 1984 ISBN 0-88854-305-0 ISSN 0384-8159 © The Royal Ontario Museum 1984 100 Queen’s Park, Toronto, Canada M5S 2C6 PRINTED AND BOUND IN CANADA AT THE ALGER PRESS Contents Abstract l Introduction 1 Methods Z Tooth Nomenclature 2 Dental Terminology 2 Registration of Specimens 3 Materials 3 Conard Fissure Specimens 3 Cavetown Specimen 4 Port Kennedy Specimens 4 Asphalto Specimen 5 Froman Ferry Specimen 5 Red Rock Canyon Specimen 6 Broadwater Formation Specimens 6 Cita Canyon Specimens 6 Bass Point Waterway I Specimens 8 El Jobean Pit Specimen 12 Santa Fe River I Specimen 12 Inglis IA Specimens 14 Haile XVA Specimens 14 Other Specimens 14 Systematic Discussion 14 Synonymy 14 Comparisons of Brown’s (1908) Smilodontopsis troglodytes and S. conardi with Smilodon 18 Comparisons of Cope’s (1880, 1899) Machaerodus (Smilodon) gracilis and M. (S.) mercerii with Smilodon 20 Specific Distinctions between Machaerodus gracilis and M. mercerii Affinity of Machaerodus gracilis with Ischyrosmilus Merriam, 1918 24 26 Resemblances between the Port Kennedy Species and Species Described within the Genera Machaerodus and Ischyrosmilus Machaerodus ischyrus 26 Ischyrosmilus 26 Ischyrosmilus idahoensis 21) Ischyrosmilus osborni 27 Ischyrosmilus johnstoni 72 | Ischyrosmilus crusafonti 33 The Cita Canyon Homothere 34 The Status of Machaerodus gracilis and the Species of Ischyrosmilus Relationships of the Genus /schyrosmilus Conclusions 41 Acknowledgements 42 Tables 43 Literature Cited Di 35 ill Digitized by the Internet Archive in 2011 with funding from University of Toronto http://www. archive.org/details/statusofsmilodonOOchur The Status of Smilodontopsis (Brown, 1908) and Ischyrosmilus (Merriam, 1918) Abstract The sabretooth genus Smilodontopsis Brown, 1908 is a junior synonym of Smilodon Lund, 1842. Distinctive fossils from Port Kennedy Caverns, Pennsylvania, described as Machaerodus (Smilodon) gracilis and M. (S.) mercerii by Cope (1880, 1899) are referred to Ischyrosmilus Merriam, 1918. Ischyrosmilus gracilis (including I. mercerii) is distinct from Smilodon and Homotherium Fabrini, 1890 (including Dinobastis Cope, 1893) and includes specimens from Florida, Texas, Nebraska, Idaho, and California. /schyrosmilus is intermediate structurally between Megantereon Croizet and Jobert, 1828 and Smilodon. Ischyrosmilus johnstoni Mawby, 1965 includes materials from Homotherium crenatidens Fabrini, 1890. ; Newly recovered materials from Bass Point Waterway I and El Jobean Pit, Florida, the type specimen of Machaerodus (Smilodon) gracilis from Port Kennedy Caverns, Pennsylvania; and reassessments of described specimens assigned to Smilodon gracilis or to species of Ischyrosmilus provide bases for the taxonomic revision. Key words—sabretooth cats, taxonomy, Quaternary, North America, Felidae, Machairodontinae, Smilodontini, Homotheriini, Machaerodus, Smilodon, Megantereon, Smilodontopsis, Ischyrosmilus, Homotherium, Dinobastis, I. gracilis, H. serum. Introduction This taxonomic review is concerned with three problems: (1) the status of the generic name Smilodontopsis Brown, 1908; (2) the identities of specimens usually referred to as Smilodontopsis gracilis or Smilodon gracilis; (3) the relationships of S. gracilis with the North American genus Ischyrosmilus Merriam, 1918, the New World genus Smilodon Lund, 1842, and the Holarctic genera Machairodus Kaup, 1833, Megantereon Croizet and Jobert, 1828, and Homotherium Fabrini, 1890. The genus Smilodontopsis dates from Brown’s original description of specimens from the Conard Fissure, Arkansas, in 1908. His two species Smilodontopsis troglodytes and S. conardi were characterized by the possession of an entepicondylar foramen on the humerus, in which they were to differ from the South American genus Smilodon Lund, 1842, probably on the uncited authority of Gervais (1878) and Cope (1899). S. troglo- dytes was distinguished from Felis (Trucifelis) fatalis Leidy, 1868 (= Smilodon fatalis) by its relatively larger and well-developed metacone on the upper carnassial. Brown (1908:189) noted that the upper carnassial lacks a protocone, the parastyle is large and high, and the prostyle or “‘preanterior cone is a well-developed tubercle, but not as large as in Trucifelis fatalis, although the tooth is much longer’ mesiodistally. Brown (1908) distinguished between S. troglodytes and S. conardi on two features: a shallower median depression on the buccal face of the metacone and a better-developed paracone on the upper carnassial. Merriam and Stock’s (1932) study of a large series of teeth of Smilodon californicus Bovard, 1907 from Rancho La Brea, Califor- nia, demonstrated that these two features are highly variable. Consequently they concluded that Brown’s recognition of two sabretooth species from the Conard Fissure was unfounded and recognized only S. troglo- dytes. Cope described Machaerodus (Smilodon) gracilis and M. (S.) mercerii in 1899 from Port Kennedy Caverns, Pennsylvania, as lightly built cats within the general North American lineage of Pleistocene sabretooths. Merriam and Stock (1932) suggested that these species might be more closely related to Smilodontopsis than to Smilodon and included them within the former genus as a subgenus of Smilodon. They also recognized only one species, S. gracilis, from the Port Kennedy Caverns. In their (1932) classification of the North American Pleistocene sabre- tooths, Merriam and Stock placed all species within the genus Smilodon but assigned the subgenus Trucifelis to include S. fatalis Leidy, 1868, S. floridanus Leidy, 1889, S. californicus Bovard, 1907, and S. nebrascensis Matthew, 1918. Smilodontopsis was also reduced to subgeneric rank to include S. troglodytes Brown, 1908 (including S$. conardi Brown, 1908), S. gracilis (Cope, 1880), and S. mercerii (Cope, 1899), both of which had been originally described as Machaerodus (Smilodon) gracilis and M. (S.) mercerii and whose specific distinction they questioned. They also included Dinobastis as a third subgenus of Smilodon to bring Dinobastis serus Cope, 1893 within the genus. Simpson (1945) followed Merriam and Stock’s (1932) scheme without change, placing the three subgenera within Smilodon Lund, 1842. Slaughter (1963:80) reviewed the known species of Smilodon and concluded that Dinobastis constituted a genus separate from Smilodon: ‘‘There appears to be but three well defined species of Smilodon in North America, S. gracilis, S. floridanus, and S. fatalis.’’ He also synonymized S. californicus with S. floridanus. Churcher (1966) defined the generic distinctness of Dinobastis from Smilodon and concluded that Dinobastis is a junior synonym of Homotherium Fabrini, 1890. Kurtén (1965) accepted Slaughter’s taxonomy in gen- eral and considered S. gracilis to be a smaller Yarmouth- ian sabretooth from which evolved the subsequent Illinoian, Sangamon, and early Wisconsin S. fatalis (including S. trinitiensis Slaughter, 1960). He considered S. floridanus to be a synonym of S. fatalis, and S. californicus the most progressive form, even when compared with Floridan specimens of the same age, and thus apparently a valid species. Webb (1974) accepted Slaughter’s taxonomy but differed from Kurtén in considering S. californicus to be a synonym of S. floridanus. Subsequent workers (e.g., Robertson, 1976) generally followed similar taxonomic schemes and referred to the smaller sabretooth as Smilodon gracilis and to the larger as S. fatalis or S. floridanus . Kurtén and Anderson (1980) recognized Smilodon gracilis as taxonomically intermediate between Megante- reon hesperus and Smilodon fatalis and included all other smilodont sabretooths within S. fatalis, with at best, subspecific status. They separated Ischyrosmilus as an incompletely known genus with an unknown number of species and stated (1980:188-189) that the known specimens ‘‘may represent a single lineage of a sexually dimorphic form with a tendency to size increase and gradual reduction of anterior premolars. . .”’ and that it may derive from Machairodus and also be related to Homotherium. They accepted Homotherium serum as distinct from the smilodonts and possibly derived from the Old World H. latidens or North American H. crenatidens and also possibly distantly descended from Machairodus . Martin (1980) placed ‘‘M. gracilus’’ (sic) with Megantereon rather than Smilodon on the presence of a functional Ps in the former genus. He also considered Ischyrosmilus to be a synonym of Homotherium, and Homotherium to be distinct from Dinobastis . This paper attempts to demonstrate that the holotype of Machaerodus gracilis and other specimens assigned to Smilodon gracilis represent members of the tribe Smilodontini and that M. gracilis is congeneric and probably conspecific with the described species of Ischyrosmilus. Its status is considered intermediate be- tween ancestral Megantereon hesperus and descendant Smilodon fatalis on temporal and evolutionary criteria. Methods TOOTH NOMENCLATURE Individual teeth are identified by a nomenclatural and positional shorthand that gives the type of tooth, its position in the sequence, and whether upper or lower and milk or permanent. Each full notation consists of four elements: 1. 1 or r indicate left or right sides of the arcades. 2. i, c, p or I, C, P, and M indicate whether incisors, canines, premolars, or molars, with lower case letters indicating milk teeth, and upper case permanent teeth. 3. 1, 2, 3, or 4 indicate whether first, second, third, or fourth members of a morphological group within incisors, canines, premolars, and molars. 4. Superscript or subscript positions of the indices 1—4 indicate position in upper or lower jaws respectively. Thus ‘‘Ip*’’ indicates the left milk upper third premolar and ‘‘rM:’’ the right lower first permanent molar. The term ‘‘milk molar’’ is not used since such teeth are molarized milk premolars and referred to as p3 or p4, when either upper or lower and left or right quadrants are equally probable. Within the text other nomenclatural systems are cited, and these are variants of the one used here. Milk teeth may be indicated by ‘‘d’’ placed before an upper case I, C, or P, as in ‘‘dC”’ for the milk upper canine or sabre, and permanent teeth by the absence of the “‘d’’. Brown (1908:191) used ‘‘pa’’ and ‘‘m1’’ to indicate permanent lower teeth, whereas I use ‘‘Pa’’ and ‘‘M:’’. Merriam (1918:524) used ‘‘P3’’ and ‘‘Pz’’ with the bar reinforcing the lower jaw position of these premolars. DENTAL TERMINOLOGY All teeth have four orientations of the crown’s sides: lingual or internally towards the tongue, buccal or labial or externally towards the cheek or lips, mesial or along the dental arcade towards the midline (equivalent to anteriorly on some cheek teeth), and distal or along the dental arcade towards the condyle (equivalent to posteriorly on some cheek teeth). On first or central incisors, mesial is the same as medial and distal as lateral, but on first molars, lingual is the same as medial and buccal as lateral. The standard system of cusp terminology in which terms for the upper teeth end in -cone, -style, -loph, etc., and for the lower teeth in -conid, -conulid, -lophid, etc., is used throughout. Upper cheek teeth are assumed to have three basic cusps arranged in a trigon (triangle)—mesiobuccal paracone, distobuccal metacone, and lingual protocone. Subsidiary conules, e.g., paraconule, may also be present. Lower cheek teeth are assumed to have a similar but reversed trigonid (triangle), with buccal protoconid, mesiolingual paraconid, and distolingual metaconid. Distal to the trigonid may occur an accessory expansion of the crown known as the talonid (heel). The noncarnassialized premolars have three main cusps in line: mesially a paracone (-id), centrally a protocone (-id), and distally a metacone (-id). Additional enamel thickenings mesially or distally are called shelves or cingula. Incisors have a single main cusp, the protocone (-id), with subsidiary small paracone (-id) and metacone (-id) cuspules mesially and distally. Canines have only single cusps, the protocones (-ids). Ridges running along the thin crests of cusps are referred to as the mesial or distal ridges or crests of the cusp. On upper canines that are expanded to form sabres, the surfaces are referred to as the buccal and lingual faces or surfaces, and the crests as the mesial and distal margins or crests. Cusps in which expansion to form a shearing blade has occurred (carnassialization) are referred to as the protocone blade, paraconid blade, and metaconid blade. Roots that directly support cusps may be referred to as protoradix, pararadix, etc., for roots directly beneath the protocone (-id) or paracone (-id). REGISTRATION OF SPECIMENS Specimens mentioned in this report form parts of the collections of a number of institutions and are identified by catalogue numbers composed of a letter prefix and a numeral. The letter prefix identifies the following institu- tions: AMNH American Museum of Natural History, New York NY: ANSP Academy of Natural Sciences, Philadelphia, Pennsylvania. B Brayfields’ Collection, El Jobean, Florida. FDT Florida Diving Tours, Ocala, Florida. LACM Los Angeles County Museum of Natural His- tory, Los Angeles, California. ROM Royal Ontario Museum, Toronto, Ontario. ROM:B Royal Ontario Museum (Bass Point Waterway I), Toronto, Ontario. ROM:P Royal Ontario Museum, Vertebrate Palaeontol- ogy Collection, Toronto, Ontario. TMM _ Texas Memorial Museum, University of Texas, Austin, Texas. UC University of California, Department of Geol- ogy, Berkeley, California, now catalogued as UCMP (see below). UCMP University of California, Museum of Paleontol- ogy, Berkeley, California. UF University of Florida, Florida State Museum, Gainesville, Florida. UNSM University of Nebraska State Museum, Lincoln, Nebraska. USNM National Museum of Natural Sciences (United States National Museum), Washington, D.C. WT West Texas State College, Panhandle Plains Historical Museum, Canyon, Texas. All units given are metric unless otherwise specified. Dimensions are given in millimetres (mm) unless other- wise specified. Materials This section comprises annotated inventories of the specimens that have been included within Smilodontopsis, Smilodon gracilis, and Ischyrosmilus and which are considered in the ‘‘Discussion’’. CONARD FISSURE SPECIMENS Brown’s (1908) description of Smilodontopsis troglodytes and of Smilodontopsis conardi from the Pleistocene deposits of the Conard Fissure, 15 miles (24 km) south of Harrison, Arkansas, constituted the first use of this genus and descriptions of the species. Brown (1908:188) founded the genus Smilodontopsis on the type species S. troglodytes, which comprised specimens catalogued as AMNH 11786 and included the ‘‘point of an upper canine, an upper sectorial and an upper incisor; three vertebrae, distal end of humerus, distal ends of radius and ulna and olecranon process of ulna, several carpals, tarsals, metacarpals and phalanges’’, which were described. The second species, S. conardi, was founded (Brown, 1908:190) on a ‘‘complete carnassial, greater part of crown of the upper canine and lower canine’’, catalogued as AMNH 11790. Brown referred several bones and teeth, catalogued as AMNH 11792 and 11785, to the latter species. He described a ‘“‘ps, two other separate ps, a fourth (?lower) premolar, and a lower carnassial m1’’ and ‘‘several dissociated fragmentary limb bones, carpals, metacarpals and a part of associated pes are referred to (S. conardi) although they pertain to Smilodontopsis troglo- dytes equally well’’. These elements comprise three unciforms, three metacarpals (Mc)—one MclII and two Mc V—phalanges, a partial tibia, calcaneum, astragalus, navicular, ectocuneiform, and four metatarsals. Two partial dentaries are catalogued as AMNH 11785 but are not described, although the teeth are probably described. Examination of the specimens in the American Museum of Natural History, New York, revealed more specimens than Brown listed and also a different arrangement by species and catalogue numbers. As it is impossible to recognize with confidence the specimens that Brown mentioned, I shall list them as catalogued when examined by me. Inconsistencies occur, e.g., the absence of a lower canine, which may have been misidentified and be represented by an I*, the absences of Ps, Pa, and M1, and the partial mandible AMNH 11785, part of which may be present within AMNH 11786. The materials of S. troglodytes, catalogued as AMNH 11786, comprise: right upper permanent first incisor (rI’); left upper permanent canine or sabre (IC’); right upper permanent fourth premolar or carnassial (rP*); fragment of a left dentary with condyle and part of a left dentary with part of the mandibular canal, probably constituting one dentary; three partial lumbar vertebrae, possibly III, IV, and V; distal ends of left ulna, left radius, and left humerus; fragment of ?left navicular; parts of proximal end, shaft, and distal articulation of left metacarpal II; three distal articulations of metapodials, one with some shaft; two unidentifiable proximal ends of metapodials; right astragalus; two digital sesamoids; four entire and two fragmentary proximal phalanges; three entire and three fragmentary middle phalanges; and four damaged ungual phalanges. The species S. conardi was described at the same time and founded on three isolated teeth, catalogued as AMNH 11790: right upper permanent second incisor (rl*), possibly identified as a lower canine by Brown; crown of left upper permanent canine or sabre (IC’); and left upper permanent fourth premolar or carnassial (1P*). Additional elements are catalogued as AMNH 11782 (and 11785 in Brown, 1908:190), noted as ‘‘Postcranial of S. conardi?’’, and comprise: distal ends of right humerus and left ulna; left scapholunar lacking sesamoid process; left and right unciforms; right pisiform; distal end of tibia; right calcaneum; left and right proximal phalanges of digit I. Associated with these identified and catalogued speci- mens are two catalogued and six uncatalogued specimens from the same deposit. Some may be listed by Brown within either S. troglodytes or S. conardi but are identified in the collections only as ‘‘Smilodon’’. They comprise: crown of right upper permanent first incisor (rl); partial crown of right upper permanent sabre or canine (rC’); mesial part of right upper permanent fourth premolar or carnassial (rP*); two single-rooted right permanent third lower premolars (rP3’s); left unciform and scapholunar lacking pisiform facet (AMNH 11787); and proximal articulation of proximal phalanx of digit I. When these catalogued and uncatalogued specimens are assessed to determine the minimum number of individuals from which the entire Conard Fissure sample of sabre- tooths could have derived, only two individuals are required. Possibly one has been described as S. troglo- dytes and one as S. conardi, although the skeletal elements are probably mixed in all three samples. Certain characteristics of these two individuals are evident: both Ps’s are single-rooted, and the mesial and distal serrations on the margins of the sabres are of the same type and have the same degree of development as seen in sabres of Smilodon. The upper carnassial fragment assigned to S. troglodytes (rP*, AMNH 11786) lacks any remnant of a protocone, and the entire tooth assigned to S. conardi (1P*, AMNH 11790) possesses a very small tubercle above the protoradix identical to that present on the partial upper carnassial (rP*, uncatalogued) tentatively assigned to ‘‘Smilodon”’ The sabretooth materials collected by Brown from the Conard Fissure derive from at least two individuals of a large cat similar to Smilodon in all respects. The single-rooted condition of Ps’s and the presence of rudimentary protocones on two upper carnassials are both conditions that are not unknown in other isolated specimens of Smilodon and occur in some individuals of S. californicus (Merriam and Stock, 1932:48-51). CAVETOWN SPECIMEN Hay (1920) described the new species Smilodontopsis mooreheadi on an isolated right upper carnassial (rP*, USNM 9212) from Cavetown, Washington County, Mary- land. Dimensions of this tooth were given as: mesiodistal length, 26.5; paracone height, 14; metacone height, 9.5; paracone width, 11.5; and metacone width, 8 mm (Hay, 1920:107). This specimen was said to be smaller than the same tooth in S. conardi or Machaerodus gracilis and was considered to resemble P* of Felis inexpectata. Simpson (1941) re-examined this specimen and concluded that it originated from a puma, probably F. inexpectata, with which I agree. PORT KENNEDY SPECIMENS Cope (1880) described Machaerodus (Smilodon) gracilis and later (1899) M. (S.) mercerii on the remains of gracile sabretooths recovered from Port Kennedy Caverns, Pennsylvania. These specimens were considered by Merriam and Stock (1932) to constitute their revised subgenus Smilodontopsis and to be distinct from the Rancho La Brea Smilodon californicus, but they ques- tioned the validity of the second species. The deposit is considered to be late Pliocene or early Pleistocene and represents the early Irvingtonian faunal age (Kurtén and Anderson, 1980). The Port Kennedy specimens are catalogued in the Academy of Natural Sciences of Philadelphia as ANSP 44-51. No types were designated by Cope, although a broken sabre was the specimen on which M. (S.) gracilis was first described (Cope, 1880), and isolated cheek teeth rP4 and IP? (ANSP 50) those on which M. (S.) mercerii was described (Cope, 1899). The specimens of Machaerodus (Smilodon) gracilis Cope, 1880 (= Smilodontopsis gracilis [Cope] of Merriam and Stock, 1932) comprise: ANSP 44. Crushed and distorted facial region of a skull and mandible, with all four tooth rows but lacking upper molars (M?’s), right lower canine (rC1) and third premolar (rPs). Cope (1899, pl. 20) described this specimen. It has since been freed completely from its matrix and constitutes the basis from which the reconstruction in Fig. 1 was made. In the absence of ANSP 46 (below), ANSP 44 is selected as the replacement neotype. ANSP 45. Unworn right lower molar (rM1). ANSP 46. Large canine. Unlocated, but described in the Academy’s catalogue as ‘‘?right sabre’’. This specimen may be the first recognized sabre (Cope, 1880; 1895:240) stated to be in “‘the Wheatley Collection’’, and may be the holotype. ANSP 47. Six isolated specimens comprising: crown of left upper permanent sabre or canine (IC’); entire left and fragmentary right upper permanent carnassials (P*’s); left upper permanent molar (1M?); part of a left dentary; and a fragment of bone. ANSP 48. Eight broken and fragmentary pieces of postcranial skeleton comprising: distal end of left humerus; proximal end of left radius; proximal ends of right metacarpals II and III; small ungual phalanx; entire and damaged right calcanea; and right astragalus. The specimens on which Machaerodus (Smilodon) mercertt Cope, 1899 (= Smilodontopsis mercerii [Cope] of Merriam and Stock, 1932) is based comprise: ANSP 49. Left and right dentaries that articulate to form a single mandible. ANSP 50. Isolated right lower fourth premolar (rP4; B.1.4) and left upper third premolar (IP?; B.3.6), designated as the lectotypes in the catalogue, possibly in Cope’s handwriting. These specimens constitute the holotypes of M. (S.) mercerii. Three premolars were originally collectively described as Uncia mercerii by Cope (1895) but only the two catalogued as ANSP 50 formed the basis for the description of M. mercerii in 1899. The third, a right P*, ANSP 52, probably formed the basis for the identification of Uncia inexpectata (= Crocuta inexpectata, Cope, 1895; Cope, 1899). ANSP 51. Thirty-seven bones and fragments of bones, including: distal articulation and part of shaft of ?right humerus; shaft of left metacarpal II; left metacarpals III and V; damaged left metacarpal IV; damaged right metacarpals IV and V; left metatarsal II; damaged left metatarsal V; damaged right metatarsals II, HI?, and V; six proximal phalanges; six middle phalanges; and one ungual phalanx. There appears to be no reliable basis on which elements from one individual or species may be certainly identified. When available elements are assessed to determine the minimum number of individuals from which they could derive, only three are required. The more extensive individual was described as M. gracilis, and the less complete individual as M. mercerii, although postcranial elements from the two individuals appear to have been mixed. The third individual is represented by the partial left dentary (ANSP 47 partim). ASPHALTO SPECIMEN Merriam (1905) described Machaerodus? ischyrus on a dentary from Asphalto, Kern County, California, from the western margin of San Joaquin Valley at the foot of the Tremblor Range. This specimen probably derived from the Tulare Beds (Merriam, 1918) and is probably early Blancan (Kurtén and Anderson, 1980). In 1918 Merriam founded the genus /schyrosmilus on this specimen when describing a second species, /schyrosmilus idahoensis, from Froman Ferry, Idaho. The age of the Asphalto specimen was given as Quaternary or late Pliocene (Merriam, 1905) and Blancan (Mawby, 1965). The type specimen (UCMP 8140) is a left dentary lacking the coronoid, condyle, and angle, but with well-preserved Ii—I2, Ci—M1, and alveolus for Is. FROMAN FERRY SPECIMEN Merriam (1918) described /schyrosmilus? idahoensis on a partial left dentary from Froman Ferry on the Snake River, southwestern Idaho. It was recovered from beds referred to the Idaho Formation at locality 3036C (Museum of Paleontology, University of California) and considered to be of late Pliocene age (Merriam, 1918) or late Blancan faunal age (Kurtén and Anderson, 1980). The type specimen (UCMP 22343) comprises the ramus of a left dentary, broken obliquely from beneath Ps to behind M1, and lacking the genial flange. Roots of all teeth I11—M1: are preserved, with parts of the crowns of Ci and Ps. RED ROCK CANYON SPECIMEN Merriam (1919) described Ischyrosmilus osborni on a partial dentary from the lower part of the Ricardo Beds in Red Rock Canyon, north of Barstow and west of the El Paso Range, California. He had earlier named it (Mer- riam, 1918:525) without description or illustration. The age of this specimen was considered early Pliocene by Merriam (1919), but it is Clarendonian, now considered late Miocene at about 10 My ago (Evernden et al., 1964: 167). The type specimen (UCMP 19476) comprises the anterior two-thirds of a left dentary broken posterior to M1, with li-Is, Pa—M1, and root of Ps. The alveolus for Ci is empty and the genial flange is slightly abraded. Gregory (1942) considered /. osborni very similar to Eusmilus whitfordi (Barbour and Cook, 1915) and that both were incorrectly assigned generically. He concluded that they were congeneric and probably conspecific, but that paucity of material did not allow the latter conclusion with confidence. He assigned them to ‘‘Megantereon’’ sensu lato, but recognized that they were more primitive and could well belong within Sansanosmilus or Albano- smilus. Kitts (1957) placed Merriam’s (1919) /. osborni in ‘‘Albanosmilus?’’ and Mawby (1965) placed it in ‘‘San- sanosmilus?’’ which has the more complete holotype. Schultz, Schultz, and Martin (1970) placed /. osborni in their new genus Barbourofelis as B. osborni on its deep symphysis which is considered a characteristic of the tribe Barbourofelini, and EF. whitfordi was tentatively placed in the genus as ?Barbourofelis whitfordi because of a characteristically deep fossa for the superficial masseter and the ‘‘long-thin’’ carnassial, but without detailed discussion. They considered B. osborni and ?B. whitfordi to be Clarendonian (lower part of the Ash Hollow Formation, then middle Pliocene). Ischyrosmilus osborni thus represents an earlier sabre- tooth taxon than either the Smilodontini or Homotheriini and belongs within the lineage leading to Barbourofelis morrisi of late Clarendonian or early Hemphillian age (middle of the Ash Hollow Formation) and B. fricki of ‘*Kimballian’’ (channel fill above the Ash Hollow Formation) or early Blancan age (Schultz, Schultz, and Martin, 1970). BROADWATER FORMATION SPECIMENS Schultz and Martin (1970) described /schyrosmilus crusafonti on a partial left dentary from the Dan Bowman Ranch, 5% miles east and % mile north (8 km E and 1.2 km N) of Broadwater, Nebraska. The geological horizon of the specimens is the Lisco Member of the early Pleistocene Broadwater Formation. The type specimen UNSM 25493 comprises a partial left dentary with I1—Ci, Ps—M:, and alveolus for Ps. It is damaged on the genial flanges and broken behind M1. An isolated left C! UNSM 25503 is referred to this species. Schultz and Martin (1970) also identified /schyrosmilus sp. from the same horizon but 3 miles east and 2 miles north (5 km E and 3 km N) of Lisco, Nebraska. This specimen UNSM 1105 comprises damaged premaxillae with left and right I?—C! and alveoli for I'’s. CITA CANYON SPECIMENS Mawby (1965) founded /schyrosmilus johnstoni on 19 cranial and postcranial elements of sabretooths recovered from the Cita Canyon locality, Randall County, Texas (UCMP locality V-3721), of Blancan age. These materials were originally collected for the Panhandle Plains Histori- cal Museum, Canyon, Texas, but five specimens have since been transferred to the Museum of Paleontology, University of California, Berkeley, California. The specimens assigned to Ischyrosmilus johnstoni Mawby, 1965 comprise: WT 1239 (Type). Mandible with r + IPs, r +1Mz, IPs, roots of rCi and rPs. The type is stated to possess ‘‘right and left Ps and M1, left P3, roots of left C and Ps, all teeth heavily worn’’? (Mawby, 1965:576), and his illustration (1965:579, fig. 2) shows the left dentary possessing IPs—M:z in alveoli and empty alveoli for 1li—C1. Mawby designated the mandible as the type and referred the following specimens to the species: UCMP 66485. Left maxilla with P*. UCMP 66486. Partial right dentary with heavily worn P4 and M1. wT 1025. Upper canine or sabre, C! (listed by Mawby as WT 1026). Fig. 1. Ischyrosmilus gracilis (Cope, 1880). Partial reconstructions of the facial and mandibular regions of the sabretooth originally described as Machaerodus (Smilodon) gracilis Cope, 1880 from the Port Kennedy Caverns, Pennsylvania, based upon the neotype ANSP 44, with details of the ear region from ROM:B 4733. Only preserved parts are shaded, but probable outlines are restored to complete impressions. Left dentary dotted in A to allow full view of sabre. Scale in centimetres. A Anterior aspect of facial region. B_ Left buccal or lateral aspect of facial region with mastoid and glenoid processes and external auditory meatus from ROM:B 4733 from Bass Point Waterway I, Florida. C Palatal aspect of facial fragment. D_ Occlusal aspect of mandible. E Genial aspect of mandible. F Buccal or lateral aspect of mandible. wT 1238. Right dentary fragment with newly worn P4 and M1. wT 1860. Left maxilla with I'-I°, P* and stump of C! and associated damaged braincase. wT 2615. Left dentary lacking teeth, ?juvenile. wT 1026, 1528, 1834, and 2429. Fragments of sabres, Ces: The following postcranial specimens were considered to be probably referable to /. johnstoni by Mawby: UCMP 66487. Left humerus lacking proximal end. UCMP 66488. Left femur, lacking distal end and most of greater trochanter. UCMP 66489. Complete tibia. WT 625. Right humerus lacking tuberosities and part of caput. WT 626. Shaft of right femur. WT 628. Proximal end of right ulna. wT 1066. Right femur, slightly damaged and distorted. WT 1691. Distal end of right humerus. No specimen number. Distal end of left humerus. Assessment of these specimens for taxonomic homogeneity and for the minimum number of individuals from which the sample could have been derived shows that two taxa and five individuals may be involved. The two taxa comprise a homothere similar to Homotherium crenatidens and a smilodont referable to /schyrosmilus gracilis, which are represented by three and two possible individuals, respectively (see Discussion, p. 34). BASS POINT WATERWAY I SPECIMENS The area of the site is located in Section 19, Township 39 south, Range 22 east, Town of Northport, Sarasota County, Florida, at latitude 27° north, longitude 82° west. Many vertebrate fossils were recovered, both in situ and in the spoil thrown up in the dredging of the waterway during July and August 1977, and in subsequent years, from the site which is located on the southeast bank of a northeast-southwest trending canal. The site comprises a single in situ excavation in the side of the canal bank, about 2 metres below the land surface, and usually below water level. It was first revealed at a time of especially low water. Bones have also been collected from a considerable distance both to the northeast and southwest of the excavation on the bank as they weathered out from the spoil thrown up by the drag-line. At one place at least the sample is heterochronic and comprises a mixture with probable Rancholabrean materials. The name Bass Point Waterway I (BP I) is reserved for materials excavated in situ on three occasions from the site in the southeast bank. Surface-collected specimens are identified as Bass Point Waterway (BP) as they lack geological provenance. The fossils are hard and black (Figs. 2-7), with most larger elements broken but not widely dispersed. In a few instances rows of associated horse teeth were found in close proximity but lacking any alveolar bone. The specimens derive from a thin layer suggestive of a bone-lag and appear to have been water-sorted, as very little microvertebrate material was obtained by screening the surrounding matrix. The brownish grey matrix immediately enclosing the specimens has not been certainly identified but is likely a fine silt which overlies the surface disconformity of highly aquiferous ‘‘barnacle-hash’’ coquina. This coquina may be the local facies of the Tamiami Formation of probable early Pliocene age. At BP I the fossiliferous horizon is overlaid by a mainly sandy stratum, which appears to extend to the surface and which contains lenses of shells of relatively modern aspect. The mammalian fauna obtained from BP I is recognized as Blancan (A. G. Edmund, pers. comm.) and includes a chelonian, scimitar cat (Homotherium), cat or bobcat (Felis or Lynx), small equid (Nannippus phlegon), zebra (Equus [Dolichohippus]), small camelid (Hemiauchenia or Palaeolama), small cervid, and giant armadillo (Kraglievichia). All, or almost all, of the Blancan fossils found in the Northport—Port Charlotte area are associated with erosional surfaces of the Tamiami Formation, as far as can be determined, but these surfaces usually lie below the present water-table. The presence of later Ranchola- brean fossils, including Smilodon floridanus, in the surface sample derived from the drag-line spoil bears evidence to a possible marine transgressional-regressional cycle following the deposition of the Blancan BP I remains. The collections are chiefly in the possession of Mr. William and Mrs. Leila Brayfield, of El Jobean, Florida. Fig. 2. Ischyrosmilus gracilis. Cranial fragments and cheek teeth from Bass Point Waterway I, Florida. A,B Right maxillary fragments and cheek teeth, rP?—P*, (ROM:B 4230, 4234, and 4262). A—buccal or lateral; B—occlusal aspects. Note canine alveolar surface mesially on ROM:B 4234, broken protoradix on ROM:B 4230, and alveolus for M!? on ROM:B 4262 (behind ROM:B 4230). C_ Right glenoid and mastoid processes and external auditory meatus, dorsolateral aspect (ROM:B 4733). Note space between processes leading to the meatus. D-F Left maxillary fragment with P%, (ROM:B 4225). D—buccal or lateral; E—occlusal; and F—lingual or medial aspects. Note canine alveolar surface mesially (D, E), ventral margin of infraorbital canal dorsally (D, F), and broken protoradix and pararadix of P* distally (D-F). CENTIMETRES The elements recovered from BP I duplicate some of the more important elements recovered from the Port Kennedy Caverns, Pennsylvania, and dentaries from Florida and elsewhere, and thus allow significant direct comparisons. The Bass Point Waterway I sabretooth materials are deposited in the Royal Ontario Museum, Collection of Vertebrate Fossils, and catalogued with the prefix ‘**ROM:B’’. Selected specimens are illustrated in Figs. 2-7. The materials comprise: ROM:B 4100. Proximal phalanx, left pes, digit IV; articulates with 4232 (Fig. 6F). ROM:B 4205. Partially fused premaxillae, with worn stump of el?, and alveoli for el’, el?, and rI*?. ROM:B 4225. Left maxillary fragment with P?, partial alveolar surfaces for C! and mesial roots of P*, and ventral margin of infraorbital canal (Fig. 2D-F). ROM:B 4228. Right permanent upper canine or sabre, rC', central portions of crown and lingual enamel margin only (Fig. 4D). ROM:B 4229. Right dentary fragment with distal half of fourth premolar, rP4, and alveoli for first molar or carnassial, rM1 (Fig. 4N-P). ROM:B 4230. Right upper permanent fourth premolar or carnassial, rP*, lacking protocone, protoradix, and mesial half of metacone-hypocone root; paracone root fits well into distal alveolus of 4234 and distal root into alveolus on 4262 (Fig. 2A,B, Fig. 3L—-N). ROM:B 4231. Left permanent upper canine or sabre, IC’, crown fragment only, lacking distal portion near enamel margin (Fig. 4A,B). ROM:B 4232. Left metatarsal IV (Fig. 7A-C). ROM:B 4233. Left metatarsal III, lacking phalangeal articulation (Fig. 6A—D). ROM:B 4234. Right maxillary fragment with permanent third premolar, rP?, and alveolar surfaces for root of C! and mesial root of P*; paracone root of 4230 fits well into distal alveolus (Fig. 2A,B). ROM:B 4235. Right permanent upper third incisor, rl? (Fig. 3D-F). ROM:B 4236. Left permanent lower canine, IC: (Fig. 4H-J). Fig. 3. Ischyrosmilus gracilis. Isolated upper incisors and cheek A-C_ Left permanent upper third incisor, II*, (ROM:B 4237). A—lateral or distal; B—lingual; and C—medial or mesial aspects. D-F Right permanent upper third incisor, rl?, (ROM:B 4235). D—medial or mesial; E—lingual; and F—lateral or distal aspects. G,H_ Right permanent upper fourth premolar or carnassial, rP*, distal root and crown fragment, (ROM:B 4289). G—lateral or buccal; and H—medial or lingual aspects. Note worn shear surface and exposed pulp cavities in H. 10 ROM:B 4237. Left permanent upper third incisor, II? (Fig. 3A-C). ROM:B 4238. Left astragalus, slightly larger in articula- tion than calcaneum 6007 (Fig. SA—F). ROM:B 4249. Left permanent lower second incisor, Ilz (Fig. 4E-G). ROM:B 4250. Left permanent lower canine, 1Ci, with damaged crown (Fig. 4K-—M). ROM:B 4253. Right maxillary fragment with alveoli for P? and alveolar contacts for root of C! and pararadix of P*, and ventral margin of infraorbital canal. ROM:B 4260. Left maxillary fragment with alveolus for M? and alveolar contacts for distal surface of distal root of Pe ROM:B 4261. Tip of ?left permanent upper canine or sabre, ?1C! (Fig. 4C). ROM:B 4262. Right maxillary fragment with alveolus for M! and alveolar contacts for distal surface of distal root of P* 4230 (Fig. 2A, B). ROM:B 4288. Right permanent upper canine or sabre, rC!; fragment of crown above enamel margin (Fig. 4D). ROM:B 4289. Right permanent upper fourth premolar or carnassial, rP*; distal root and metacone hypocone blade only, worn through to pulp cavity (Fig. 3G, H). ROM:B 4733. Right glenoid and mastoid processes with external auditory meatus (Figs. 1B, 2C). ROM:B 4734. Right premaxillary fragment with alveolar surfaces for roots of I? and C’; root of 4235 fits well into incisive alveolus. ROM:B 4735. Right maxillary fragment with alveoli for distal root of P* and root of M}; distal root of 4230 fits well into alveolus for distal root of P* (Fig. 2A,B). ROM:B 6007. Left calcaneum; see 4238 (Fig. 5G—J). ROM:B 6037. First or proximal phalanx, ph. I, of digit I, right manus (Fig. 7E). ROM:B 6084. Left permanent upper fourth premolar or carnassial, 1P*, mesial half only; split through metacone near protocone (Fig. 30-R). ROM:B 6088. Left permanent upper third premolar, 1P* (Fig. 3I-K). ROM:B 6574. Right metatarsal III, lacking phalangeal articulation (Fig. 6E—-H). teeth from Bass Point Waterway I, Florida. I-K_ Left permanent upper third premolar, 1P?, with broken mesial root, (ROM:B 6088). I—lingual or medial; J—buccal or lateral; and K—occlusal aspects. L-N_ Right permanent upper fourth premolar or carnassial, rP*, with broken protoradix, (ROM:B 4230). L—buccal or lateral; M—lingual or medial; and N—occlusal aspects. O-R_ Left permanent upper fourth premolar or carnassial, 1P*, paracone and root, (ROM:B 6084). O—lingual or medial; P—buccal or lateral; Q—occlusal; and R-——distal or posterior aspects. Note evidence of protoradix in O, Q, and R. op) lJ x = uJ = aes = uJ O 11 The minimum number of individuals from which this sample could be derived is two, one of which was aged and the other a young adult. The aged individual possessed well-worn carnassials and is catalogued as ROM:B 4289, with possibly 4228, 4234, 4250, 4253, and 4261. The young adult is represented by the other cranial specimens and is identified by the lack of wear on the shearing surfaces of the carnassials 4230 and 6084 and on the canine tips. The glenoid and auditory meatus specimen 4733 may derive from either individual, as may the pedal elements. The astragalus 4238 and calcaneum 6007 fit together imperfectly and may represent either the older and younger animals, respectively, or an individual with more tolerance between the facets of the two elements. The Bass Point Waterway collection is composed of black, well-mineralized specimens which preserve small features and show no effects of weathering or rounding by water although many have been broken before deposition or during recovery. The Bass Point specimens include parts of upper and lower dentitions, partial dentaries, astragalus, and metatarsal III, which are comparable to those of the Port Kennedy M. gracilis discovery and from which they do not differ markedly either in detail or size. The upper canine (4228) is more gracile (perhaps indicating indi- vidual, sexual, or regional variation), and P? (4225, 4234) is longer, probably reflecting variation intrinsic to this tooth. EL JOBEAN PIT SPECIMEN An isolated left maxilla (B 8400) with third and fourth premolars, IP?—P*, was recovered from El Jobean Pit, Charlotte County, Florida, by Mr. and Mrs. William Brayfield in August 1982. The El Jobean Pit is located in Section 21, Township 40, Range 21 East, Charlotte County at approximately latitude 27° north, longitude 82° west, and is an operating gravel pit. The specimen was found in a black sand and green silt member similar to that from which the Bass Point Waterway I specimens derived and which, in the El Fig. 4. Ischyrosmilus gracilis. Isolated upper canine or sabre Point Waterway I, Florida. A,B Left permanent upper canine or sabre, IC’, crown fragment, (ROM:B 4231). A—lingual; and B—buccal aspects. C ?Left permanent upper canine or sabre, IC’, tip, (ROM:B 4261). D_ Right permanent upper canine or sabre, rC’, basal portion of crown, (ROM:B 4228). Buccal aspect. E-G_ Left permanent lower second incisor, Ilz, (ROM:B 4249). E—lateral or distal; F—lingual; and G—medial or mesial aspects. 12 a Jobean Pit, overlies the Caloosahatchee Shell Beds. Associated faunal elements comprise a gomphothere proboscidean (Cuvieronius?), small equid (Nannippus phlegon), and giant armadillo (Holmesina floridanus), indicating a Blancan age (W. Brayfield and A. G. Edmund, pers. comm., 1982). The specimen (Brayfields’ Collection B 8400: Fig. 8) comprises most of a left maxilla in which the palatal shelf at the level of P® and alveolar margins of the canine alveolus are damaged. The premolars, P*® and P*, are nearly undamaged, although both are worn, especially P*. Alveoli for C? and M! are preserved. Sutural contacts with the premaxilla, frontal and palatine, are preserved, and the anterior buttress of the jugal is present to preserve the inferior margin of the orbit and the base of the zygomatic arch. The individual from which B 8400 derived was old or very mature from the worn state of P*, which correlates with the fused condition of the anterior root of the jugal. The carnassial P* lacks a protocone but has a protoradix (Fig. 8D). P* and P® have a small distolingual wear facet. Both premolars show a vertically striated pattern of wear different from that in the Bass Point Waterway I specimens (Fig. 3G—-R). M! was triangular in section and presumably styliform. The facial aperture of the infraorbi- tal canal is deep and wide, as in the type M. gracilis (Fig. 1A,B) and in smilodonts, but in lateral aspects the lateral margin is less curved and nearly vertical. SANTE FE RIVER I SPECIMEN An isolated lower fourth premolar (rP4, FDT 488) was recovered from Santa Fe River locality I, in Gilchrist and Columbia counties, Florida (Webb, 1974). Kurtén (1965:220) suggested that this site is heterochronic, with many fossils ‘‘undoubtedly Late Pleistocene’’ while ‘‘others may be as old as the Blancan’’ and some of the felid materials ‘‘might be middle Pleistocene in age’’. Kurtén (1965:241, fig. 14 Above) illustrated the tooth and dated it as ‘‘probably Middle Pleistocene’’. Webb (1974:150) agreed with Kurtén, dated the river site as fragments, lower incisors, and fragment of dentary from Bass H-J Left permanent lower canine, ICi, (ROM:B 4236). H—lateral or buccal; I—lingual or posterior; and J—medial or mesial aspects. K-M_ Left permanent lower canine, IC1, with damaged crown, (ROM:B 4250). K—lateral or buccal; L—lingual or posterior; and M—nmedial or mesial aspects. N-P_ Right dentary fragment with distal half of Pa and alveoli for Mi, (ROM:B 4229). N—buccal or lateral; O—lingual or medial; and P—occlusal aspects. Note root of P4 in N protruding below the ventral boundary. op) Lu oO be WW = ro Ra ve) O 3 ‘‘late Blancan’’, and assigned the specimen to Smilodon gracilis Cope although Kurtén had assigned the tooth to ““Smilodon sp., possibly Smilodon fatalis or Smilodon gracilis”’. INGLIS IA SPECIMENS Inglis Locality IA is located in Citrus County, Florida, and yields an early Irvingtonian fauna (Klein, 1971; Webb, 1974; Kurtén and Anderson, 1980). The sabretooth remains include cranial, dental, and postcranial elements of gracile form. The materials are deposited in the Florida State Museum, University of Florida, Gainesville, Florida, and have been described by Annalisa Berta (pers. comm., 1982). I examined casts of a right dentary (UF 20065), and these show a permanent fourth premolar and first molar (P4s—M1), the alveolus for Ps, partial alveoli for Is—C1, and the symphysis and genial flange, but lack the coronoid, condyle, and angle. This small machairodont was referred to Smilodon gracilis by Klein (1971), Webb (1974), Kurtén and Anderson (1980), and A. Berta (pers. comm., 1982). The cast almost matches in its preserved areas the left dentary from Port Kennedy Caverns and is comparable with the dentaries assigned to /schyrosmilus by Merriam (1905, 1918, 1919) and Mawby (1965). HAILE XVA SPECIMENS The Haile XVA sabretooth specimens form part of a Blancan fauna (Webb, 1974) of latest Pliocene age (Robertson, 1976) recovered from a fissure in Ocala (Eocene) limestone near Haile, Alachua County, Florida. The specimens comprise the distal end of a right tibia and a right astragalus (UF 17496, jointly catalogued), which articulate well together, and a right metacarpal II (UF 17498), which are presumed to derive from a single individual (Robertson, 1976). Robertson (1976:157) re- ferred the specimens to Smilodon because of ‘‘(1) the presence of the astragalar foramen; and (2) the medial facet for the calcaneum and facet for the navicular are merged’’. The fusion of the navicular and medial calcanear facets was also observed by Merriam and Stock (1932). These specimens are not directly comparable to the other described Florida materials, although the astragalus may be compared to that from Bass Point Waterway I and to astragalus ANSP 48 from Port Kennedy Caverns which was described by Cope (1880) as Machaerodus (Smilo- don) mercerii. Webb (1974) and Robertson (1976) assigned these specimens to Smilodon gracilis. OTHER SPECIMENS A. Berta (pers. comm., 1982) reported and described additional specimens from El Jobean, Inglis IA, and McLeod A. Systematic Discussion SYNONYMY Familiarity with names of the North American Pliocene and Pleistocene sabretoothed cats is insufficient for the nomenclature and taxonomy to be clear, as names have been misspelled and distinctions between members of the tribes Smilodontini and Homotheriini ignored or confused. The following synonymy presumes the conclusions that will become evident at the end of this discussion and is inserted here for the benefit of readers. The synonymy is limited to selected North American taxa or taxa that were present in North America, although other Eurasian and South American taxa may have had representatives in North America. Each taxon is cited with its synonyms and with geographical localities or distribution. Family Felidae Gray, 1821:302 (includes Machaerodontidae Woodward, 1898:399) Eurasia, Africa, North and South America Subfamily Machairodontinae Gill, 1872:59 Eurasia, Africa, North and South America Fig. 5. Ischyrosmilus gracilis. Astragalus and calcaneum from Bass Point Waterway I, Florida. A-F Left astragalus, (ROM:B 4238). A—distal or navicular; B—medial; C—ventral or calcanear; D—lateral; E—proximal or tibial; and F—posteroproximal aspects. 14 G-J Left caleaneum, (ROM:B 6007). G—distal or cuboid; H—medial or sustentacular; I—dorsal or astragalar; and J—lateral aspects. CENTIMETRES Tribe Homotheriini Kurtén, 1963:97 Eurasia, Africa, and North America Genus Homotherium Fabrini, 1890:176 Eurasia, Africa, and North America Machairodus—Fabrini, 1890:175, Val d’ Arno, Italy. Dinobastis Cope, 1893:896, Western Oklahoma. Epimachairodus—Simpson, 1945:120, Eurasia. Smilodon—Hibbard, 1952:12, McPherson, Kansas. Ischyrosmilus?—Kurtén, 1963:98, America. Species Homotherium serum (Cope, 1893) North America Dinobastis serus Cope, 1893:896, Western Oklahoma. Homotherium serum—Churcher, 1966:272: Friesenhahn Cave, Texas; Hamman Pit, Kansas. Rawn-Schatzinger and Collins, 1981:18: Old Crow Basin, and Dawson, Yukon; American Falls Lake Beds, Idaho; San Francisco Bay, California; Delmont, South Dakota; Sand Draw, Nebraska; Sandahl, and McPherson, Kansas; Hennesey, Oklahoma; Slaton, Bagett Cave, Gilliland, Laubach Cave, and Friesenhahn Cave, Texas; Gassaway Fissure, Tennes- see; Reddick IA and Inglis IA, Florida. ?Ischyrosmilus johnstoni Mawby, 1965:573 partim, Cita Canyon, Texas. Tribe Smilodontini Kurtén, 1963:97 Eurasia, Africa, North and South America Genus Ischyrosmilus Merriam, 1918:524 North America Ischyrosmilus Merriam, 1915:262 nomen vanum, Ricardo, California. Ischyrosmilus Merriam, 1917:425, 430 nomen vanum, Tulare, California. Species Ischyrosmilus gracilis (Cope, 1880) North America Machaerodus (Smilodon) gracilis Cope, 1880:857, Port Kennedy Caverns, Pennsylvania. Uncia mercerii Cope, 1895:392 partim, Port Kennedy Caverns, Pennsylvania. Smilodon gracilis—Cope, 1895:448, Port Kennedy Caverns, Pennsylvania. Machaerodus (Smilodon) mercerii—Cope, 1899:240, Port Kennedy Caverns, Pennsylvania. Machaerodus? ischyrus Merriam, 1905:173, Asphalto, California. Smilodontopsis (Machaerodus) gracilis—Brown, 1908:191, Port Kennedy Caverns, Pennsylvania. Smilodontopsis (Machaerodus) mercerti—Brown, 1908:191, Port Kennedy Caverns, Pennsylvania. Dinobastis (Machaerodus) ischyrus—Brown, 1908:191, Asphalto, California. ?Ischyrosmilus ischyrus Merriam, vanum, McKittrick, California. Ischyrosmilus? idahoensis Merriam, 1918:524, Froman Ferry, Idaho. Ischyrosmilus ischyrus Merriam, 1918:524, McKittrick, California. Machaerodus? gracilis—Matthew, 1918:229, Port Ken- nedy Caverns, Pennsylvania. Machaerodus? mercerii—Matthew, 1918:229, Port Ken- nedy Caverns, Pennsylvania. Smilodon (Smilodontopsis) gracilis—Merriam and Stock, 1932:15, Port Kennedy Caverns, Pennsylvania. Smilodon (Smilodontopsis) mercerti—Merriam and Stock, 1932:15, Port Kennedy Caverns, Pennsylvania. Machaerodus gracilis—Slaughter, 1960:490, Port Ken- nedy Caverns, Pennsylvania. Ischyrosmilus johnstoni Mawby, 1965:583 partim, Cita Canyon, Texas. Ischyrosmilus crusafonti Schultz and Martin, 1970:34, Broadwater, Nebraska. M(eganterion) gracilus Martin, 1980:149, no referred material. 1917:425 nomen Fig. 6. Ischyrosmilus gracilis. Metatarsals II from Bass Point Waterway I, Florida. A-D Left third metatarsal, mt. II, lacking phalangeal or distal articulation, (ROM:B 4233). A—medial or internal; B—dorsal or anterior; C—lateral or external; and D—proximal or cuneiform aspects. E-H_ Right third metatarsal, mt. III, lacking phalangeal or 16 distal articulation, (ROM:B 6574). E—medial or internal; F—dorsal or anterior; G—lateral or external; and H—proximal or cuneiform aspects. Note puncture, marks of gnawing, and abrasions on dorsal surface seen in F and G. CENTIMETRES 17 Genus Smilodon Lund, 1842:193 North and South America Felis—Leidy, 1868:175, Hardin County, Texas. Trucifelis Leidy, 1868:175, Hardin County, Texas. Machaerodus—Lydekker, 1884:33, Hardin County, Texas. Drepanodon—Leidy, 1889:14, Ocala, Marion County, Florida. Machairodus—Leidy, 1889:14, Ocala, Marion County, Florida. Smilodontopsis Brown, 1908:188, Conard Fissure, Arkan- sas. Species Smilodon fatalis (Leidy, 1868) North America Felis (Trucifelis) fatalis Leidy, 1868:175, Hardin County, Texas. Trucifelis fatalis Leidy, 1868:175, Hardin County, Texas. Machaerodus fatalis—Lydekker, 1884:333, Hardin County, Texas. Drepanodon or Machairodus floridanus Leidy, 1889:14, Ocala, Marion County, Florida. Smilodon floridanus—Adams, 1896:433, Ocala, Marion County, Florida. Smilodon californicus Bovard, 1907:155, Rancho La Brea, California. Smilodontopsis troglodytes Brown, 1908:188, Conard Fissure, Arkansas. Smilodontopsis conardi Brown, 1908:190, Conard Fis- sure, Arkansas. Smilodon nebrascensis Matthew, 1918:228, Hay Springs Quarry, Sheridan County, Nebraska. Smilodon (Trucifelis) californicus—Merriam and Stock, 1932:16, Rancho La Brea, California. Smilodon (Trucifelis) fatalis—Merriam and_ Stock, 1932:16, Hardin County, Texas. Smilodon (Trucifelis) floridanus—Merriam and_ Stock, 1932:16, Ocala, Marion County, Florida. Smilodon (Trucifelis) californicus—Merriam and Stock, 1932:16, Ocala, Marion County, Florida. Smilodon (Trucifelis) nebrascensis—Merriam and Stock, 1932:16, Hay Spring Quarry, Sheridan County, Nebraska. Smilodon (Trucifelis) californicus brevipes Merriam and Stock, 1932:161, Rancho La Brea, California. Smilodon trinitiensis Slaughter, 1960:487, Trinity River, Second Terrace, Dallas, Texas. COMPARISONS OF BROWN’S (1908) SMILODONTOPSIS TROGLODYTES AND S. CONARDI WITH SMILODON When the various specimens assigned to Smilodontopsis troglodytes, S. conardi, and Smilodon sp. from the Conard 18 Fissure are compared with each other, no obviously distinctive differences are apparent among comparable elements. Measurements of the dental elements fall within the recorded size ranges available for Smilodon from Rancho La Brea, California, and Talara, Peru (Table 1). Merriam and Stock (1932:13) remarked on the presence of single-rooted Ps’s in both individuals of the Conard Smilodontopsis since such teeth occur in only 6 per cent of S. californicus. They suggested that this tooth ‘“‘was more frequently developed in the Arkansas form than in the type from California. Moreover, its presence may have been typical in Smilodontopsis .’’ They also said the teeth in the Conard specimens showed less reduction than in S. californicus, and Brown (1908:190) stated ‘*Ps is single rooted and oval in outline, with the internal face flattened; the apex of the cone above the base of the crown, inside measurement, is equal to the transverse measurement of the tooth at this point; there is a distinct anterior basal cone about half the size of the posterior basal cone. Two other separate teeth, p?, show the same characters.’’ Merriam and Stock (1932) considered that Ps of Smilodon californicus 1s not of this type, all the cusps being reduced, and concluded that Smilodontopsis appeared to be more primitive than the species from Rancho La Brea. However, my examinations of Ps’s of S. californicus indicate that examples matching the size and development of specimens ascribed to Smilodontopsis troglodytes and S. conardi are present, and consequently Merriam and Stock’s conclusion is not generally supportable. Merriam and Stock (1932) considered that the postcra- nial elements assigned to Smilodontopsis conardi resem- bled those from Rancho La Brea. They stated (1932:14) that the scapholunar ‘“‘is of relatively small size but resembles more closely in structure the comparable element in [Smilodon] than in true cats,’’ but that ‘‘the palmar border of the surface of the magnum does not show the characteristic notch seen in specimens from. . .”’ Rancho La Brea. Measurements of the metapodials lie at the lower end of the range of variation for metapodials of S. californicus but are larger than those referred to Smilodon c. brevipes by Merriam and Stock (see Table 2). Merriam and Stock (1932:14) concluded that ‘‘all the characters, with one exception, which Brown (1908:188) regarded as generically distinctive of Smilodontopsis are found to apply equally well to Smilodon californicus. The single exception, namely the presence and character of P3, has already been commented upon.’’ This exception, as discussed above, is not valid, and matching examples are present in the Rancho La Brea population of S. californicus. Kurtén and Anderson (1980) list both Smilodontopsis troglodytes and S. conardi as junior synonyms of Smilodon fatalis . When measurements of the various elements originally attributed to Smilodontopsis troglodytes and S. conardi are CENTIMETRES Fig. 7. Ischyrosmilus gracilis. Metatarsals IV and phalanges from Bass Point Waterway I, Florida. A-D Left fourth metatarsal, mt. IV, (ROM:B 4232). A—medial or internal; B—dorsal or anterior, C—lateral or external; and D—proximal or cuneiform aspects. E Proximal or first phalanx, ph. I, of digit I, right manus, compared (Table 2), no distinctions are evident beyond normal variation within a population. Because no quantita- tive or qualitative basis exists that distinguishes the Conard Fissure sabretooth material at the specific or higher taxonomic level, all specimens, whether described as Smilodontopsis or Smilodon?, can be placed in a single species. When these specimens are compared with similar elements from Rancho La Brea, they are again indistin- guishable on qualitative or quantitative criteria. The Conard sabretooth is thus properly referred to (ROM:B 6037). Dorsal or anterior aspect. F Proximal or first phalanx, ph. I, probably from digit IV, left pes (ROM:B 4100). Dorsal or anterior aspect. Articulates with ROM:B 4232. Smilodon Lund, 1842 and, as the Fissure’s deposits are Pleistocene (and their age cannot be determined more accurately), and the animal indistinguishable from S. californicus, assignment to the species S. fatalis (Leidy, 1868), S. floridanus Leidy, 1889, and S. californicus Bovard, 1907 must be considered. Slaughter (1963) and Kurtén (1965) considered S. fatalis to have existed during Illinoian, Sangamon, and early Wisconsin times and S. floridanus and S. californicus during late Wisconsin and possibly early Holocene times. Kurtén (1965) postulated 19 that §. floridanus might include S. californicus, and Webb (1974) confirmed their single identity. Webb also noted that S. floridanus was recovered from Sangamon deposits at Arredondo IA in Florida. Kurtén and Anderson (1980) placed Smilodontopsis troglodytes and S. conardi within Smilodon fatalis. Thus the Conard Fissure Smilodon may be referred to S. fatalis (Leidy, 1868) which, in its turn, may be the senior synonym of S. floridanus Leidy, 1889 when more extensive samples of the Illinoian populations are available. The genus Smilodontopsis Brown, 1908 is thus a junior synonym of Smilodon Lund, 1842, and the species troglodytes Brown, 1908 and conardi Brown, 1908 are junior synonyms of floridanus Leidy, 1889, in which is also included californicus Bovard, 1907. COMPARISONS OF COPE’S (1880, 1899) MACHAERODUS (SMILODON) GRACILIS AND M. (S.) MERCERIT WITH SMILODON Machaerodus (Smilodon) gracilis and M. (S.) mercerii were originally described by Cope (1880, 1895, 1899) on material derived from Port Kennedy Caverns, Pennsy]- vania. This material probably represents remnants from three individuals, and it is again arguable whether three individuals associated in a single deposit could derive from two nearly identical species. Merriam and Stock (1932) questioned the validity of Cope’s specific distinc- tion but did not propose that M. (S.) mercerii was conspecific and synonymous with M. (S.) gracilis. However, they did consider that these species were definitely distinct from Smilodon (Smilodon) californicus and assigned the two species to the subgenus Smilodontop- sis rather than the subgenus Smilodon to emphasize the divergence from the latter and a probable closer affinity to the Conard Fissure sabretooth. As shown above, the genus Smilodontopsis Brown, 1908 is based on Smilodontopsis troglodytes (and Smilodontopsis conardi), which are conspecific with Smilodon floridanus, and thus Smilodontopsis Brown, 1908 is a junior synonym of Smilodon Lund, 1842. As Machaerodus gracilis and M. mercerii differ markedly from Smilodon both morphologically and temporally, they must be considered separately for assignment to generic or subgeneric taxa. Merriam and Stock (1932:15) gave the main characteris- tics of M. gracilis and M. mercerii as the presence of Ps, the greater “‘inferior development of the forward flange of the mandible,’’ a noticeably smaller Ci which is only slightly more massive than Is, and a slenderer sabre. They suggested that the presence of Ps and the smaller Ci indicate affinity to Smilodontopsis Brown (= Smilodon Lund) and possibly to Dinobastis Cope (= Homotherium Fabrini) as Ci is always heavier than I3 in Smilodon californicus. Churcher (1966) showed that North Ameri- can Dinobastis serus Cope, 1893 was distinct from Smilodon and congeneric with Homotherium Fabrini, 1890 as Homotherium serum (Cope, 1893), and that the closest affinity is to the Chinese H. ultimum from the Villafranchian of Locality 9 at Choukoutien. Examination of the specimens referred to as Machaerodus gracilis and M. mercerii show that they bear little resemblance to those described as Smilodon troglo- dytes and S. conardi (i.e., Smilodon floridanus) and in fact differ markedly from the general suite of characters that distinguish Smilodon and with which the Conard Fissure materials agree. Thus the use of Smilodon as a subgenus for the Port Kennedy sabretooth, as originally suggested by Cope (1880, 1899), may be unwarranted. The facial region and mandibles of M. gracilis and M. mercerli are more slender and more lightly built than are those of S. troglodytes (including S. conardi = S. floridanus: Fig. 1, Table 3). The sabre (Fig. 9C) is more gracile, proportionately shorter, and with the serrations on the margins differing in quality and degree. The serrations in M. gracilis are less beadlike and extend relatively farther across the distal faces of the tooth, whereas those of Smilodon are more rounded and are restricted to the mesial and distal margins of the sabres. Serrations are present on the distal margin but are nearly absent on the mesial in both taxa, and are finer in M. gracilis or M. merceriit, where they are smaller. Serrations are absent from all other teeth except C1. Four well-developed cusps on P? correspond to the parastyle, paracone, metacone, and buccodistal style of P? in Smilodon. However, these cusps show little or no reduction, although the tooth is low crowned, and thus cannot be considered to match the condition of P?’s in Smilodon. The upper carnassial, P*, lacks the lingual protocone but has a well-developed protoradix. The paraconule is large and usually lacks a mesial ectoparastyle. The mandible is smaller in all dimensions in M. gracilis Fig. 8. /schyrosmilus gracilis. Partial left maxilla with third and fourth premolars, P3—P4, (ROM:B 8400), from El Jobean Pit, Florida. Compare B 8400 with neotype of M. gracilis (Fig. 1|A—C) and premolars with those from Bass Point Waterway I (Fig. 2A,B,D-F, Fig. 3G-R). A-D A—anterior; B—medial; C—buccal or lateral; and D—palatal aspects. O == a i —— 4% = = t = | CENTIMETRES than in Smilodon (Table 4), and the mental or genial flanges of the chin are thinner and extend ventrally well beyond the margin of the ramus, a development that is absent in all known specimens of Smilodon (Fig. 10B,C). The mental foramina are double, and two or three accessory foramina may occur medial to the genial flanges. The condition of these foramina varies in Smilodon, but a single mental foramen is usual although double foramina occur in the Talara, Peru, and Californian populations of Smilodon, and the anterior accessory foramina vary in number from two to five. The jaws are less robust and thinner buccolingually than in Smilodon and in morphology greatly resemble some of the earlier sabretoothed carnivores (e.g., Hoplophoneus, Dinictis, Machairodus , or Megantereon). The dentition as a whole is smaller and less massive than in Smilodon and readily distinguishes the two taxa. The lower canine, C1, is serrated on the distal margin of the crown as in the sabres and P* is higher crowned but with cusps that are not as elongate nor as slanted distally. When measurements of the preserved cranial elements of M. gracilis and M. mercerii are compared with the observed ranges of variation for measurements of similar elements of Smilodon from Rancho La Brea and Talara, overlaps occur only in the lower range of 10 of 36 dimensions and are indicated by asterisked superscript indices (Table 3). Characters 1-3 involve the length of upper diastema, a dimension that is highly variable in many mammals, as it is dependent on size and age and degree of maturation of the individual. Characters 4, 5, 7, and 10 involve the mesiodistal diameters of I’, P?, P*, and M!. The three teeth I', P?, and M! are relatively small and variable in size as they are undergoing reduction in Smilodon. Character 7 (mesiodistal diameter of P*) is dependent on the development of the mesial parastyle (Character 8) which may or may not be present, and is functionally interrelated to Character 5 (mesiodistal diameter of P*) in the provision of a shearing blade. The presence of a parastyle on P* usually correlates with a smaller P? in Smilodon. Character 6 (crown height at protocone of P?) reflects the development of the shearing function on P*, with smaller protocones on P® correlating with greater carnassialization of P*. In Character 9 (metacone blade length of P*) all of the measurements for M. gracilis lie within the lower part of the size range for P*’s of Smilodon and reflect the gracile form of the former sabretooth. Characters 8, 9, and 10 (dimensions of the crown of M?) reflect the small size and variability inherent in a tooth that has undergone reduction and is variable in all Felidae, and overlapping measurements are expected in taxa of similar size. Most of the overlaps at the lower extremes of the size ranges indicate that the Port Kennedy sabretooths had a dentition that was about the same size as that of the smaller individuals of Smilodon from the late Pleistocene of Rancho La Brea and Talara. Table 4 gives measurements of the mandible and lower dentition of the Port Kennedy sabretooths compared with specimens referred to as Smilodon gracilis from Florida and ranges of variation in Smilodon from Rancho La Brea and Talara. Overlaps in the ranges of variation for the Port Kennedy sabretooths’ dimensions with those of Smilodon from Rancho La Brea and Talara occur in 13 of 47 dimensions measured and are indicated by asterisked superscript indices (Table 4). Characters 1-5 involve lengths over a series of teeth and mainly estimated measurements are available. Characters 6-9 involve Fig. 9. Outlines of sabres or upper canines (C!’s) of Smilodontini and Homotheriini. Sabres are drawn to the same crown-root length for easy comparisons. Pillars to the left of the tips of the crowns indicate relative scales; associated figures give crown- root lengths in mm. Smilodontini (upper row) A Megantereon megantereon from Val d’ Arno, Italy. Redrawn in reverse from Kurtén (1962, fig. 1). B Megantereon megantereon from Puebla de Valverde, Spain. Redrawn in reverse from Kurtén and Crusafont (1977, fig. 15). C Machaerodus (Smilodon) gracilis from Port Kennedy Caverns, Pennsylvania. Drawn from crushed facial specimen ANSP 44 in the Academy of Natural Sciences, Philadelphia, Pennsylvania, U.S.A. D Smilodon neogaeus from Talara, Peru. Drawn from gracile specimen ROM:P 2120 in the Royal Ontario Museum, Toronto, Canada. E Smilodon neogaeus from Talara, Peru. Drawn from robust specimen ROM:P 2228 in the Royal Ontario Museum, Toronto, Canada. Homotheriini (lower row) F Machairodus crenatidens from Val d’ Arno, Italy. Redrawn in reverse from Kurtén (1962, fig. 1). G Homotherium crenatidens (= Ischyrosmilus johnstoni par- tim) from Cita Canyon, Texas. Drawn from specimens wT 1026 (root) and wT 2429 (crown) in the Panhandle Plains Historical Museum, West Texas State College, Canyon, U.S.A. H Homotherium ultimum from Choukoutien (Locality 13), China. Redrawn from Teilhard de Chardin (1939, fig. 3). I Homotherium serum from the Friesenhahn Cave, Texas. Drawn from specimen TMM 933-1923, in the Texas Memorial Museum, University of Texas, Austin, U.S.A. mesiodistal diameters of teeth in the symphyseal or diastematic regions. For Characters 6, 7, and 9 (mesiodis- tal diameters of Iz, Ci, and Mz) the overlap in measurements is at the lower end of the range and is probably not significant. For Character 8 (mesiodistal diameter of Ps) the larger overlapping measurement is estimated and the range of the Ps’s for the Rancho La Brea and Talara Smilodon is founded on teeth which may or may not have been naturally present in life, as this tooth was undergoing reduction and is usually absent. Character 10 is the greater dimension of the symphysis and may be unreliable, as the symphyseal region has been distorted in both Port Kennedy mandibles. Characters 12 and 13 (depth of the dentary beneath Ps and mesial to Ps) record small overlaps at the lower extremes of ranges that include estimated measurements and are also probably not significant. The Port Kennedy sabretooth is smaller than Smilodon from Rancho La Brea and Talara in 38 of 44 dimensions (Table 5) but is similar to or overlaps the lower ends of the ranges for Smilodon in 6 dimensions. These dimensions are the trochlear diameter of the humerus, the proximal transverse diameters of metacarpal and metatarsal II, and three dimensions of the astragalus. As the sample available for the Port Kennedy sabretooth is inadequate for statistical treatment, it may be that the smaller sabretooth possessed elements that were either smaller or approached in size those of small individuals of the Pleistocene American Smilodon. The comparatively large astragalus and metapodials II suggest large extremities, and the dimensions of the calcar and astragalar facet on the calcaneum support this conclusion. Tables 3-5 compare dimensions of the Port Kennedy sabretooth materials with similar data for Smilodon derived from the Rancho La Brea, California, and Talara, Peru, samples. In total, overlap of ranges occurs in only 30 of 134 dimensions available, and most occurs marginally at the lower limits of the size ranges for Smilodon. The Port Kennedy sabretooth is therefore considered to be dissimilar from the Pleistocene Smilodon, as was also concluded by Merriam and Stock (1932), in that it is generally smaller and more gracile, and differs in qualitative characters of the dentitions, skull, and mandible. Table 6 gives measurements of Bass Point Waterway I and El Jobean Pit sabretooth materials. Comparisons of these with those of Machaerodus gracilis (including M. mercerii) from Port Kennedy show agreement in most dimensions for all elements except the sabre, C’. Measurements of this tooth indicate that the Bass Point Waterway individual was considerably smaller than those from Port Kennedy and may reflect individual, geograph- ic, or sexual differences. In all other dimensions the Bass Point Waterway dnd El Jobean Pit specimens are not significantly divergent in dimensions from the Port Kennedy Cavern M. gracilis, and I consider them to represent the same species. The Haile XVA and Bass Point Waterway I remains are similar in details and size (Table 6). ROM:B 4238 has a well-developed astragalar foramen, about 4 x 2 mm in transverse and anteroposterior diameters, respectively, and a flattened outline to the navicular facet in broad aspect, as described for Smilodon californicus by Merriam and Stock (1932:142) and by Robertson (1976) for the Haile XVA specimen UF 17498. However, the medial calcanear and lateral navicular facets are not broadly confluent on the distal face of the astragalus, but only connected by an unfaceted ridge about 4 mm long. SPECIFIC DISTINCTIONS BETWEEN MACHAERODUS GRACILIS AND M. MERCERII Cope (1899) considered the Port Kennedy sabretooth to represent two species, although only one was described in 1880 and mentioned by him in 1895, and the second described in 1899. I found no significant divergences in conformation or size between Machaerodus gracilis and M. mercerii. Qualitatively the specimens resemble one another in all respects and quantitatively their dimensions Fig. 10. Left lateral aspects of dentaries of Smilodontini and Homotheriini. Specimens are drawn to similar incisive-condylar lengths. Bars beneath ventral margins represent 50 mm. Smilodontini (left column) A Ischyrosmilus johnstoni from Cita Canyon, Texas, U.S.A. Redrawn from Mawby (1965, fig. 2). B_ Ischyrosmilus ischyrus from Asphalto, California, U.S.A. Drawn and restored from cast of UCMP 8140, Museum of Palaeontology, University of California, Berkeley, U.S.A. C Machaerodus (Smilodon) gracilis from Port Kennedy Caverns, Pennsylvania. Drawn from dentary specimen ANSP 44, Academy of Natural Sciences, Philadelphia, Pennsylvania, U.S.A. D Smilodon neogaeus from Talara, Peru. Drawn from robust 24 specimen ROM:P2228, Royal Ontario Museum, Toronto, Canada. Homotheriini (right column) E Homotherium nestianum (= Machairodus nestianus) from Perrier, France. Redrawn from de Bonis (1976, fig. 5). F Homotherium palanderi (= Machairodus palanderi) from locality 113, Pao-Te-Hsien, Shansi, China. Redrawn in reverse from Zdansky (1924, Tafel XXVI, fig. 2). G Homotherium serum (= Dinobastis serus) from Friesenhahn Cave, Texas, U.S.A. Redrawn from Churcher (1966, fig. 1B). vary less than do similar dimensions of Smilodon (Tables 3-5). Therefore these three individuals are considered to represent the species Machaerodus gracilis Cope, 1880. Other specimens from Florida (Bass Point Waterway I, Sante Fe River Locality I, Inglis Locality IA, and Haile Locality XWA) resemble the Port Kennedy species wherever comparisons are possible. These specimens, except for the hitherto unpublished Bass Point Waterway and El Jobean Pit materials, from time to time have been referred to as Smilodon gracilis (Slaughter, 1963; Kurtén, 1965; Webb, 1974; Robertson, 1976; Kurtén and Ander- son, 1980). The specific identity and distinctiveness is correct, but the generic assignation is questionable on the evidence discussed previously. It is thus necessary to reassess the generic status of the taxon originally named Machaerodus gracilis by Cope (1880). AFFINITY OF MACHAERODUS GRACILIS WITH ISCHYROSMILUS MERRIAM, 1918 Resemblances between the Port Kennedy Species and Species Described within the Genera Machaerodus and Ischyrosmilus Comparisons were made among the M. gracilis ‘‘type’’ materials from Port Kennedy, the Florida specimens, and other described North American sabretooth cats reported from the Pleistocene or late Pliocene. Machaerodus ischyrus Merriam (1905:173) considered the Port Kennedy sabre- tooth to resemble Machaerodus when compared with his new species Machaerodus(?) ischyrus from the late Pliocene of Asphalto, Kern County, California. He said: ‘“‘Of the North American species Machaerodus gracilis Cope, from the Port Kennedy Fissure, resembles the California species in the form of Ps, while the heel of Mi has almost disappeared. M. ischyrus differs from this species particularly in the shortness of the diastema and probably of the whole jaw, in the absence of a posterior basal cusp on Pa, and in the complete reduction of the heel of M1. Judging from Cope’s figure* (E. D. Cope, Jour. Philad. Acad. Nat. Sci., 2nd ser., vol. 11, pl. 20, fig. 1—footnote) of the type of M. gracilis the mandibular flange is not as prominent and the jaw as a whole somewhat weaker than in M. ischyrus.’’ The measure- ments in Table 4 substantiate this estimation. Merriam (1905) also compared M. ischyrus with Smilodon fatalis, Dinobastis serus (= Homotherium serum), Felis imperialis, and M. catecopsis but concluded that all four are distinct from M. ischyrus. He noted (1905:173) that H. serum did not ‘‘differ from M. ischyrus 26 greatly in size and had large external incisors with a moderately elongated superior canine. The superior sectorial has a large protostyle, but the basal cusp anterior to this is rudimentary. This means that, as in Hop- lophoneus, the posterior part of Ps opposing it was probably relatively shorter than in Smilodon. The charac- ters mentioned all suggest correlation with M. ischyrus, although there would at present be no justification for considering them identical.’’ His conclusion is reasonable as M. ischyrus (known only from the anterior part of a dentary and later described as /schyrosmilus ischyrus) was deduced to possess large sabre-like upper canines in contrast to the known prominent but relatively lower crowned, deltoid, and thinner canines of H. serum. Merriam also compared M. ischyrus with M. palaeindicus from the Upper Siwalik beds of India, which he considered resembled M. ischyrus greatly. Merriam’s type of M. ischyrus was referred to by Bovard (1907:163) as ‘‘characterized by the great reduc- tion of Ps, the presence of a single posterior cusp on Pa, the absence of both metaconid and heel from M1, the shortness of the diastema, the possession of a prominent flange below the symphyseal region and the abbreviation of the jaw.’’ He concluded that ‘‘the presence of Ps and the absence of metaconid and heel from M1, the shortness of the jaw and the heavy flange indicate that S. californicus and M. ischyrus belong to widely separated species if not genera.”’ Brown (1908:191) wrote that M. ischyrus probably belonged to the genus Dinobastis ‘‘as indicated by the nearly equal size of paraconid and metaconid in mi, and the erect position of ps without posterior basal heel. This correlation was suggested by Dr. Merriam.’ This suggestion was unfortunate as Merriam (1918) proposed the new genus /schyrosmilus, for which he made M. ischyrus the type species. Ischyrosmilus The description of Jschyrosmilus is confusing. Merriam (1915:262) mentioned ‘‘/schyrosmilus (Sabre-tooth cat)’’ from the Ricardo Pliocene but gave no description or discussion of the new generic name. In 1917 (p. 425) he noted that /schyrosmilus ischyrus had been obtained from the Upper Pliocene (Tulare) of California, /schyrosmilus osborni n. gen. n. sp. (= Barbourofelis osborni, vide infra) from the Lower Pliocene (= late Miocene) Ricardo of the Mohave Desert (p. 430), and ‘“‘/schyrosmilus n. sp.’’ from the Upper Pliocene Snake Creek of Idaho, which he presumed represented ‘‘the Tulare stage in California’ (p. 432). However, Merriam described neither the new genus nor the new species in this publication. Ischyrosmilus idahoensis Merriam (1918) gave the first description of the genus when he described /? idahoensis, and Simpson (1945) dated the genus from this description. However, /? idahoensis is not the type species, as Merriam (1918:524) described ‘‘/schyrosmilus, new genus’’, founded upon ‘‘Machaerodus? ischyrus’’, in a footnote appended to the description of the new species ‘‘/schyrosmilus? idahoen- sis’’. Merriam’s (1918) remarks effectively parallelled those he gave later, in 1919, and constitute the first description of the genus. Merriam (1918) referred to the later publication as ‘‘in press’’. He (1918:524) described the new species /? idahoensis as differing from ‘‘the typical S. californicus in dentition by the presence of a small Pz. . . . The Idaho mandible differs from that of S. californicus very decidedly in the size and proportions of the flange below the diastema. In S. californicus the flange is comparatively short, ending posteriorly rather abruptly some distance in front of Ps. In no. 22343 the posterior end of the flange fades out below the anterior end of Pz. A flange of this type seen in the Idaho specimen is found also in Ischyrosmilus ischyrus of the Tulare Pliocene of California, in J. osborni of the Ricardo Pliocene, and in Machaerodus palaeindicus of the upper Siwalik beds of India.’ Merriam added that the Idaho mandible is as large as that of Smilodon californicus, possesses Ps and three mental foramina, and is tentatively dated as earlier than Rancho La Brea. Osborn (1918:31) dated /? idahoensis as early Pleis- tocene or late Pliocene and referred to a sabretooth ‘resembling ?Machaerodus ischyrus’’, presumably re- ferring to the taxon rather than to an isolated tooth as no such specimen can be located. Ischyrosmilus osborni (= Barbourofelis osborni) Merriam (1919:543) reiterated the generic characteristics of Ischyrosmilus at length. This description is effectively the same as that in his 1918 footnote but with minor differences. The 1919 description, with textual di- vergences from the 1918 version in square brackets, reads: ‘“Mandible massive; flange clearly marked, relatively wide anteroposteriorly, slightly deeper [wider] than in Smilodon, not as strongly developed as in_ typical Machaerodus; length of [‘‘length of’ omitted] diastema much as in Machaerodus, but shorter than in Smilodon. Ps very small, with one root. Pa with single posterior cusp or with [*‘with’’ omitted] incipient division of this cusp. [End of 1918 description.] Mi without metaconid and heel. This group known only from beds referred to the Pliocene.’ Merriam considered the genus to be less advanced than Smilodon. I. osborni is said to be smaller than J. ischyrus, flatter in the anterior face of the symphyseal region, with much thinner or transversely narrower cheek teeth, and in the division of the posterior lobe of Ps into two cusps. Hay (1927) mentioned only /. idahoensis in his review of the fossil Mammalia from the western states and dated it as Upper Pliocene or Lower Pleistocene, presumably following Osborn (1918). Merriam (1919:545) described Ischyrosmilus osborni from the Ricardo Beds (Pliocene) near Red Rock Canyon, California, represented by a partial left dentary (UCMP 19476, holotype) with Ps and damaged M1, roots of li—Is and Ps, and alveolus for C1. Merriam distinguished the species from others of the genus by its smaller size, transversely thinner cheek teeth, division of the posterior lobe of Ps (?metaconid) into two cusps, and a flatter anterior face to the symphysis. He also remarked on the dorsoventrally narrow masseteric fossa beneath Mi, and speculated that this might indicate a different conforma- tion to the jaw joint. The genial flange on the specimen is broken anteriorly and ventrally, so that the symphyseal face appears flatter in lateral aspect. Ps is single rooted, and shows no sign of a double origin. P4 and M: are as in other /schyrosmilus, except for the double-cusped metaconid on Ps, which is possibly formed from the talonid. The characters described by Merriam (1919) are those selected by Gregory (1942), Kitts (1957), and Mawby (1965) to separate the Ricardo ‘“‘/schyrosmilus’’ osborni from other /schyrosmilus. Schultz, Schultz, and Martin (1970) placed ‘‘/.’’ osborni in their new genus Bar- bourofelis but mentioned only three of the characters enumerated by the earlier authors. Ischyrosmilus johnstoni Johnston and Savage (1955:39) listed ‘‘Jschyrosmilus or Homotherium’’ from the Blancan (Plio-Pleistocene) of Cita Canyon, Texas. This tentative identification was based on a restored skull composed of separated facial and cranial sections (WT 1860) and a mandible (WT 1239) which derive from individuals the size of Homotherium or Smilodon. Other cranial, dental, and postcranial materials are associated with these two specimens and have been compared with materials of other large sabretooths. Also, a cast of the restored skull and mandible (ROM 30136 = WT 1860 + 1239) was available for study in the Royal Ontario Museum. Mawby (1965) described the Cita Canyon machairodont materials originally reported by Johnston and Savage (1955). He described a new species, /schyrosmilus johnstoni, based on the mandible (WT 1239, holotype) and referred to it other mandibular specimens (UCMP 66486 [ = WT 1834], wT 1238, 2615), cranial specimens (WT 1860, Pa UCMP 66485), and isolated canine specimens (WT 1026 [= 1025], 1026, 1528, 1834, 2429). Some fragmentary postcranial specimens (humeri, WT 625, 1691, UCMP 66487, unnumbered; ulna, WT 628; femora, WT 626, 1066, UCMP 66488; tibia, UCMP 66489) are tentatively referred to I. johnstoni. Mawby (1965:583) compared /. johnstoni with /. ischyrus and Smilodon californicus, but ‘‘no close relationship between the two genera is implied. On the contrary, Smilodon and Ischyrosmilus, respectively, rep- resent two quite distinct machairodont lineages, for which Kurtén (1963) has proposed the tribal names Smilodontini and Homotheriini.’’ (Churcher [1966] incorrectly cited Kurtén as proposing two subfamilies, Smilodontinae and Homotheriinae. ) Mawby (1965:584) compared /. johnstoni with Eurasian Homotherium and concluded that ‘‘there are many similarities between Homotherium crenatidens [from the Villafranchian of Europe and Asia] and /schyrosmilus jJohnstoni. Both were long-limbed, relatively slender animals, with deep massive mandibles. The dentitions are similar, with relatively short, flattened sabres, and high, narrow carnassials. There are numerous points of re- semblance in the skull, including the form of the mastoid process, the shape of the bulla, the keel and depressions of the basioccipital and division of the sagittal crest. It is likely that direct comparison of J. johnstoni with specimens of H. crenatidens would show the two to be congeneric.’’ This discussion will show that Mawby’s (1965) conclusions are partly erroneous, as two sabretooth cats, one of which is a homothere and the other a smilodont, are represented in the material assigned to /. johnstoni. In Homotherium (Fig. 11) the incisors are set in a rounded arcade and are equally spaced, the sabres are proportionately more laterally compressed and only mediumly hypsodont compared to other machairodonts, P? is small but always present, P* is a two-rooted vertical blade lacking all remnants of protocone or protoradix in the late Pleistocene or Holocene species, and M? is a more or less styliform tooth with a triangular crown. In the lower jaw, Ps is two-cusped and more hypsodont, and P4 and Mi: are hypsodont and show a vertical pattern of wear. All teeth in Homotherium, whether milk or permanent, are serrated, and all incisors possess mesial and distal cuspules that are also serrated (Churcher, 1966). Another character that appears to be distinctive in the facial region is a small elongate oval infraorbital foramen, which is divergent from the larger and broader infraorbital foramina typical of smilodonts. Fabrini (1890) established Homotherium on Machairodus crenatidens and M. nestianus. Cope’s (1893) Dinobastis serus from North America morphologi- cally resembles the Chinese H. ultimum (see Churcher, 1966) and, as has been reviewed above, /schyrosmilus (based on/. ischyrus) is not congeneric with Homotherium or even closely allied to it. Mawby (1965:580, fig. 3) illustrated a specimen with Ps and Mi (wT 1238) with enlarged detail drawings of crenelations on the mesial margin of the protoconid of P4, a mesial flange on M1, with crenelations on its margin, and a slightly damaged protoconid. As the type mandible (WT 1239) possesses only well-worn teeth, Mawby (1965:577-578) referred the teeth on wT 1238 to J. johnstoni and ascribed their characteristics to his new species. Churcher (1966:267) described Ps and Mi: of the Friesenhahn Cave, Texas, Homotherium serum thus: ‘‘The fourth lower premolar is absolutely smaller than that in Smilodon and does not show an incipient mesial or distal style. In the unworn condition all three major cusps show fine serrations. The lower molar is set closely distal to or slightly overlapping Ps, and both Ps and Mi have distally tilted crowns. In absolutely unworn erupting Mi’s, the whole mesial, distal, and occlusal edges of the crowns are serrated, especially the mesial and distal surfaces; Mi is proportionately more buccolingually flattened. . .”’ than Ps. The characteristics of Ps and Mi inH. serum are equally descriptive of teeth in WT 1238 from Cita Canyon, with the reservation that Mawby’s illustration tilts the occlusal plane some 15° from the horizontal and thus provides a distad tilt to the teeth. I observed serrations on P4 and Mi teeth in H. serum, and Figure 4 (reproduced from Churcher, 1966:266) illustrates their extent. The referred specimen WT 1238, from which the lower cheek teeth of I. johnstoni were described, therefore derived from Homotherium. Consequently, some of the other referred Fig. 11. Homotherium serum (Cope, 1893). Reconstruction of skull and mandible of the Friesenhahn Cave scimitar-tooth from Churcher (1966). Reconstruction based mainly on specimens in the Texas Memorial Museum, Austin, Texas (skulls TMM 933-3231 and 3582 and mandible TMM 933-1) with additional details from isolated teeth and elements from the Friesenhahn Cave and specimens from Kansas and Oklahoma. Scale in centimetres. A Lateral aspect of skull. B_ Lateral aspect of left dentary. C Occlusal aspect of mandible. D_ Nuchal aspect of skull. E Ventral aspect of left side of skull. F Dorsal aspect of left side of skull. 29 specimens or probably referred postcranial elements could also derive from an individual of Homotherium. Mawby (1965:576) listed two fragments of upper canines (WT 1026, 2429) among the referred materials. In 1966 the late Dr. C. W. Hibbard and I examined these fragmentary specimens (and most of the other material assigned to /. johnstoni), then at the Museum of Paleontology, University of California, Berkeley, and we concluded that they probably derived from a homothere because of the mesiodistally wide alveolar region, the relatively short crown height or medium development of hypsodonty, the more acute angle between the axes of the crown and the root, and the configuration of the enamel margin and the serrated mesial ridge (Fig. 9G). Kurtén (1962:103, fig. 1) gave outline comparisons of sabres of Megantereon and Homotherium crenatidens from Val d’Arno (Fig. 9A,F). Mawby’s (1965:586, fig. 5) illustra- tion and cast of the skull (WT 1860) show a restored sabre that is more hypsodont and slender than are those in the Friesenhahn Cave H. serum, very worn P*’s that curve inwards as only in very old individuals, no P?’s or M?’s, but alveoli for both M?’s, and possible traces of resorbed alveoli for the P*’s (Fig. 12C). He described P*® in UCMP 66485 as possessing two well-developed separate roots as determined by the alveoli. The P*’s are too worn to reveal any of the crown’s details. Their roots are shorter than in S. californicus and there are separate roots that either supported a protocone or just buttressed the crown. The infraorbital foramen is oval and about half the size of that in Smilodon. The incisors appear to have been set in a more rounded arc than in Smilodon, since all three are visible in lateral aspect, as in Homotherium, but are too worn to reveal their crown morphology. The characters on the facial fragments indicate that UCMP 66485 probably represents a homothere because of the presence of P*, as does wT 1860 (facial region) because of the small infraorbital foramina, the strongly concave or deep palate and the incisors set in an arc. The distance from the sabre to the mesial margin of P* is short and could not support a fully developed three- or four-cusped P? (as in M. gracilis or Smilodon) but could accommodate a small single- or double-cusped tooth with fused or closely set roots as in Homotherium (Table 3). Measurements available from the dentitions and skull of the Cita Canyon specimen’s facial region (WT 1860, Table 3) show that all dimensions are absolutely greater than any from the type specimens of M. gracilis or M. mercerii, except for the crown height of the sabre, C!, which is less. These measurements generally lie within the size ranges for Smilodon and Homotherium serum and indicate that the Cita Canyon facial specimen derived from an individual larger than the Port Kennedy Caverns sabre- tooth and as large as either Smilodon or H. serum. Among the other Cita Canyon material are canines that are similar 30 in height and proportions to those of H. serum (Fig. 9G). The entire C’ (wT 1025; listed as 1026 by Mawby, 1965) is estimated by Mawby as 150 mm in overall length. C. W. Hibbard and I estimated the same dimensions for the combined WT 1026 and 2429 as 160 to 165 mm. Sabres in the Port Kennedy M. gracilis are about 160 mm long, although the total length had to be estimated as the roots are still implanted and the six sabres of the Friesenhahn Cave H. serum range between 118 and 126 mm in total length. Although the size of wT 1025 and the composite figure based on WT 1026 and 2429 more closely approach that for M. gracilis, their conformation indicates affilia- tion with homotheres and not with smilodonts. Mawby (1965:78) commented: *‘Near the root, the anterior edge of wT 1025 turns somewhat medially, less sharply than in Dinobastis but much as in Smilodon ,’’ which suggests that WT 1025 represents a smilodont and disagrees with C. W. Hibbard’s and my observations based on wT 1026 and 2429. Specimens WT 1528 and 1834, both fragmentary sabres, are not discussed by Mawby, although measurements of WT 1834 are given with those of wr 1025 (Mawby, 1965:578, table 2—as 1026). As wT 1025 is considered to derive from Homotherium, perhaps WT 1834 does also, for its measurements fall within the range of variation of C!’s for that genus. The cranial portion of WT 1860 is not directly comparable with the type of M. gracilis, which lacks that region, but is comparable with those of H. serum and S. californicus (Fig. 12). Mawby (1965:581, fig. 4) illus- trated and described the basicranial region of J. johnstoni but compared it only with S. californicus. A major diagnostic difference between a smilodont and a homothere is the degree of closure of the ventral re-entrant into which the external auditory meatus opens; that is, whether the mastoid process lies close to the posterior surface of the glenoid process or not. In S. californicus the separation is from | to 2 mm, and inH. serum, it is about 20 mm. Mawby’s figure 4 shows a wide separation that is about 15 mm on the cast, and he (1965:580) stated that the mastoid does not approach the glenoid process. Other characteristics of homotheres are also present in the basicranial region. The foramen ovale is separated from the eustachian canal by a ridge of bone in Smilodon and by a flat area in Homotherium and in WT 1860. The stylomastoid foramen is separated from the tympanohyal pit in Smilodon by an area of the wall of the bulla, but in Homotherium they are apposed because of the posterior inclination of the mastoid process, as in WT 1860. The basioccipital-basisphenoid has a central ridge that ends anteriorly in two round pits at the level of the mastoid and slightly posterior to the midpoint of the bullae in wT 1860, as in Homotherium but not in Smilodon, where the ridge continues to the level of the eustachian canal, bifurcating Fig. 12. Left lateral aspects of skulls of Smilodontini and Homotheriini, with restorations of the fragmentary skull of Ischyrosmilus johnstoni Mawby, 1965. Specimens are drawn to similar basicranial lengths. Bars beneaih the ventral margins represent 50 mm. Smilodontini (left column) A Megantereon megantereon from Seneze, France. Redrawn from Piveteau (1961, fig. 172). B_ Smilodon floridanus (= S. californicus) from Rancho La Brea, California, U.S.A. Redrawn in reverse with sabres restored to in vivo position from Merriam and Stock (1932, pl.1, fig. 1). C_ Ischyrosmilus johnstoni from Cita Canyon, Texas, U.S.A. Redrawn from Mawby (1965, fig. 5). Homotheriini (right column) D Homotherium nestianum (= Machairodus nestianus) trom Perrier, France. Redrawn from de Bonis (1976, fig. 8a). E Homotherium serum (= Dinobastis serus) from Friesenhahn Cave, Texas, U.S.A.. Redrawn from Churcher (1966, fig. LA). F Homotherium sp. (= Ischyrosmilus johnstoni partim) from Cita Canyon, Texas, U.S.A. Redrawn with facial and basicranial fragments WT 1860 reoriented within the outline of a Homotherium skull and with a homothere sabre to illustrate a more likely conformation. and ending near the roots of the alae. Mawby (1965) also reported that the sagittal crest bifurcated posterior to the parietals, as in Homotherium but not in Smilodon. The cranial and facial parts of WT 1860 are thus assigned to Homotherium as both differ in all observed characters from those found in smilodonts, including Ischyrosmilus. The single fragment of the cranium of M. gracilis recovered from Bass Point Waterway I (ROM:B 4733, Figs. 1B, 2C) has a ventrally closed external auditory opening, with the mastoid process apposed to within 5 mm of the posterior glenoid surface. This small fragment indicates that the condition in /schyrosmilus was probably typical of smilodonts but not as derived as in Smilodon. The type mandible of /. johnstoni (WT 1239), of which the left dentary is illustrated by Mawby (1965:579, fig. 2), has been broken through the diastema at the level of the posterior mental foramen; the condylar and angular area is separated from the coronoid and body; the mental flange, parts of the ramus beneath Pa and M1, and a small area posterior to Mi are missing; and the buccal walls of the incisive and canine alveoli are absent. Comparison of the type mandible of /. johnstoni WT 1239 with the type of M. gracilis ANSP 44 is informative (Figs. 1OA and C). The dentaries of M. gracilis are well preserved but compressed anteriorly and may be used as a standard for comparison. The incisors and lower canine were set with their crowns in a slightly bowed transverse row, as 1s suggested by both the type and the Cita Canyon dentary WT 1239. The mental foramina are double, although not as well separated, and without a subdivision of the anterior foramen. This could be explained in WT 1239 by the absence of the posterior foramen and by the separation of the twinned openings of the anterior foramen. Ps is small, with two fused roots, and occurs separated mesially from P4 in all specimens examined. Pa and M1, where preserved, are similar in morphology and in distal inclination to those in the type of M. gracilis. The ventral margin of the ramus, the slopes of the anterior margins of the coronoids, and the conformation of the masseteric fossa are all similar to those of the type of M. gracilis but only fragments of the latter two are available. In wT 1239 the margin of the mental flange is absent. This area 1s preserved in the type of M. gracilis and shows a posterior step below the mental foramen, which may be unusual for the taxon. The type mandible of /. johnstoni (WT 1239), that of M. gracilis (ANSP 44), and casts of a right dentary from Inglis IA (UF 20065) ascribed to Smilodon gracilis by Webb (1974) are similar. Both wT 1239 and UF 20065 are broken in the incisive and canine alveolar regions in an almost identical manner and have lost comparable areas of the condylar, coronoidal, and angular regions. Morphologi- cally these specimens are conformationally identical. However, the Inglis IA specimen is linearly about 70 per cent the size of the Cita Canyon specimen and presumably is from an animal as small as those from Haile KXVA (Robertson, 1976) or Bass Point Waterway I and slightly smaller than the type dentaries of M. gracilis from Port Kennedy Caverns. Mawby’s (1965:586, fig. 5) reconstruction is thus a composite. The jaw is the type of /. johnstoni and is correctly considered similar to other /schyrosmilus, but it also resembles M. gracilis and other specimens that are usually referred to in the literature as Smilodon or Smilodontopsis gracilis. The skull derives from a homothere, somewhat more primitive than H. serum from Texas. No contacts are present between the facial and cranial portions of WT 1860, and error has been introduced in the reconstruction because the cranial portion has been placed lower with respect to the palatal plane than it should be and also because the glenoid-mastoid-occipital- condylar alignment has been tilted at about 30° to the occlusal plane rather than about 20° as in Homotherium (Fig. 12C). In Smilodon the inclination is about 40°. This placement accommodated the /schyrosmilus jaw, allowing it to have the space for opening, but in Smilodon the glenoid is placed well above the palatal and occlusal planes, and in Homotherium it lies only slightly above these planes. This ventral displacement further distorts the cranial outline of the skull, which becomes nearly parallel to the basicranial and palatine planes instead of rising distally to a high lambda, as in Smilodon and Homotherium (Fig. 12B,E). When relative positions of the fragments are altered, the skull outline and basicranial and palatal relationships conform to the normal homothere configuration (Fig. 12F). The specimens referred by Mawby (1965) to his new species Ischyrosmilus johnstoni thus derive from two different sabretooth species, one a homothere and the other a smilodont. The type mandible represents the smilodont /. johnstoni, and the skull is that of a species of Homotherium which is more primitive than H. serum. I. johnstoni is represented by the type specimen: a mandible with right and left P4’s and M1’s, left Ps, roots of left C1 and right Ps, all heavily worn (WT 1239), a partial right dentary (UCMP 66486, ex WT 1834), an edentulous, possibly juvenile, left dentary (WT 2615), a left maxilla with P* (UCMP 66485), and an isolated sabre (WT 1025 = wT 1026 in Mawby’s list of materials). The genus Homotherium is represented by a fragment of a right mandible with nearly unworn P4 and Mi (WT 1238), a left maxilla with ’-I°, P*, stump of sabre and two separated cranial fragments (WT 1860, restored as a skull), and two fragments of sabres (WT 2429 crown, WT 1026 root). Another fragmentary sabre (WT 1528) is unassigned. Mawby (1965:581-582) gave a number of measure- ments for postcranial elements that he considered probably derived from /. johnstoni and commented that ‘‘they indicate that Ischyrosmilus was a long-legged animal, probably with a slender build like that of Homotherium (Schaub, 1925) rather than the squat proportions of Smilodon or the sloping stance of Dinobastis (Meade, 1961).’’ As Dinobastis and Homotherium are congeneric (Churcher, 1966), this statement should be modified to emphasize the long-leggedness of the homotheres con- trasted to the heavy build of the smilodonts. When measurements of the postcranial elements of the sabretooths from Cita Canyon are compared with those given by Meade (1961) for Homotherium serum (Table 7), the humeri (WT 625, 1691, UCMP 66487) are found to be smaller than for Smilodon and slightly smaller than for H. serum. Mawby (1965) noted that these humeri generally are similar to the humerus of Machaerodus catacopsis from the Hemphillian of Texas. The humeri may therefore be derived from /schyrosmilus, as they appear to be of about the appropriate size and morphology for a smilodont smaller than Smilodon and about the size of M. gracilis. The ulna (WT 628) appears to be smaller than the ulnae of both Smilodon and Homotherium from the few data available, and may also be derived from /schyrosmilus. The femur (WT 1066) is as long as that of Smilodon and some 12 per cent longer than the femora of Homotherium serum from Friesenhahn Cave, Texas. The other femoral measurements approximate those of H. serum and S. californicus, except those of the caput and condyles, which are less than in S. californicus. In these latter dimensions WT 1066 and UCMP 66488 resemble Homotherium. The tibia (UCMP 66489) resembles the tibiae of both Smilodon and Homotherium in its measurements. How- ever, as Mawby stated, it is long and slender. Both the femora and tibia are long, longer than might be expected to correspond to the humeri and ulna, and slenderly built. Mawby (1965) noted that the morphology of the suture joining the lesser trochanter on the femur to the greater trochanter resembled that in Homotherium. Thus the femora and tibia probably derive from a homothere and may be associated with the skull and other elements identified above as Homotherium. The postcranial elements thus represent both a smilo- dont and a homothere, with humeri (WT 625, probably WT 1691, UCMP 66487) and the ulna (WT 628) from Ischyrosmilus johnstoni and the femora (WT 1066, UCMP 66488) and tibia (UCMP 66489) from Homotherium. Ischyrosmilus crusafonti Schultz and Martin (1970) described /schyrosmilus crusafonti from a partial left dentary (UNSM 25493) with well-worn C1, Pa, and M1, crownless Is, and alveolus for Ps, and referred a partial upper sabre (C' UNSM 25503) to that species. Both specimens derived from the early Pleistocene Lisco Member of the Broadwater Formation on the Dan Bowman Ranch, near Broadwater, Merrill County, Nebraska. The species is diagnosed as the “smallest known species of [the] genus; Ps single-rooted, but proportion- ately large’? by Schultz and Martin (1970:34), who commented that the dentary 1s ‘“‘considerably smaller and more lightly built than the holotype (ramus) of /schyro- smilus ischyrus Merriam, but it resembles the latter in most other respects.’’ Measurements of the described dentaries of /schyrosmilus (Table 4) show J. crusafonti to have measurements that differ from those of /. ischyrus, whether taken by Merriam (1905, 1918), Schultz and Martin (1970), or Kurtén (1982, pers. comm.), by being 2 to 4 mm smaller, and thus cannot be said to be ‘considerably smaller’’ in such a large animal. When the observed ranges of variation of similar dimensions for the better-known sabretooth Smilodon are used as a standard, all measurements for any one dimension on available specimens of any species of /schyrosmilus fall within a smaller range than exists in Smilodon. The dentary USNM 25493 possesses a damaged genial flange, a twinned foramen on the lateral face of the flange, and a second mental foramen beneath Ps, as in /. johnstoni. Such a morphology is present also in the Port Kennedy M. gracilis jaws and the Inglis IA (UF 20065) dentary. Schultz and Martin (1970:34) stated ‘*The incisors of J. crusafonti are caniniform, each consisting of a single cusp which is recurved posteriorly. The incisors and canines are serrated.... The I: is small and compressed while I2 and Is become progressively larger. The [lower] canine is relatively large for a machairodont cat.”’ Schultz and Martin (1970:35, fig. 1B) illustrated buccal and occlusal aspects of the dentary and showed no serrations or curvature on the lower incisors or canines but did show that the crown of Is is missing, that of Ci worn to a stub, and those of I: and Iz worn or faceted. The P3 was small, separated from P4 by about 8 mm, and possessed a fused double root, of which the mesial portion was half the diameter of the distal. The size and position agree with those of the Ps’s observed in M. gracilis and other Ischyrosmilus, although the proportions of the fused roots vary. Schultz and Martin (1970:34) stated *‘The Ps appears to have been proportionately larger than that in/. ischyrus.”’ The measurements given in Table 4 for the alveolus are not comparable to those of crowns of Ps3’s as Ps is usually tilted distally, and thus the alveolar margin traces an oblique line across the face of the root, and because cingula on the crown make measurements of crowns greater than those of roots that support them. They further stated ‘‘The Pa of J. crusafonti is slightly recurved posteriorly and has two roots. It consists of a paraconid, a 33 larger protoconid and a metaconid about equal in size to the paraconid. An accessory cusp (talonid) also may have been present posterior to the metaconid, but the tooth is too worn to be certain of this.”’ Schultz and Martin (1970:35, fig. 1B) illustrated the Pa in both buccal and occlusal aspects. The paraconid is well preserved but heavier than in other /schyrosmilus or M. gracilis, the protoconid is higher and similar in section to the metaconid, but both the latter are too worn to be certain of details. In /schyrosmilus the protoconid is large, almost three times the mesiodistal length and twice the height of either the paraconid or metaconid, and the talonid is a distinct distal or distolingual shelf. These characters are not discernible in Schultz and Martin’s diagrams. Both Ps and Mi: are well worn distal to the protoconid of Ps, and some characters described by Schultz and Martin may no longer be present on the specimen. Schultz and Martin (1970:34,35, fig. 1A) also reported Ischyrosmilus sp. from the early Pleistocene Lisco Member of the Broadwater Formation but from Garden County, Nebraska. This identification is based on UNSM 1105, which comprises left and right damaged premaxillae and left and right I?, I?, and C!. The sabres are curved as in Smilodon and appear to have slipped from the sockets 10 or 15 mm. Schultz and Martin (1970:34) note that this specimen is “‘larger than /schyrosmilus crusafonti from the Broadwater Local Fauna, and is near the size of /. Johnstoni Mawby. The canines are proportionately shorter than in Smilodon although they are almost as broad. They are also more coarsely serrated.’’ Schultz and Martin (1970, fig. 1A,C) figured the premaxillae and sabres (UNSM 1105) of Ischyrosmilus sp. and a partial left upper canine (UNSM 25503) assigned to /. crusafonti to the same scale (3/5). These diagrams show both sabres to have equivalent mesiodistal diameters at the bases of the crowns (18 mm on figures = +30 mm actual) and to be comparable in outlines. Dr. Bjorn Kurtén measured the gross dimensions of the teeth in both specimens, and these measurements agree with those for other /schyrosmilus specimens available and for M. gracilis (Tables 3, 4). No basis of comparison exists for stating that UNSM 1105 is from a larger individual than the type of J. crusafonti, as all the Nebraska specimens agree in dimensions with those recorded for M. gracilis. The sabres of UNSM 1105 (Schultz and Martin, 1970, fig. 1A) are similar to those of Smilodon in curvature and robustness, rather than to those of Homotherium (Fig. 9H,I). ‘‘The incisors in UNSM 1105 are arranged in a broad curve anterior to the canines, and are large and widely spaced. They show distinct wear facets in the regions where they interlocked with the lower incisors, as in Smilodon’’ (Schultz and Martin, 1970:34). In fact the incisors lie in a shallow curve, with I'’s and I?’s located on 34 a transverse axis and I°’s slightly posterior to it. The wear facets show truncations of the crown from edge-to-edge bites, as in Smilodon, and show the effects of both crown-to-crown and interlocking occlusions. The size of I’’s is unknown, as they are absent and the alveoli damaged, but they were subequal to or smaller than [?’s from the space available. The specimens described as Ischyrosmilus crusafonti (UNSM 25493, partial dentary, and UNSM 25503, partial sabre) and Ischyrosmilus sp. (UNSM 1105, partial pre- maxillae with sabres), all from the Lisco Member of the Broadwater Formation of early Pleistocene age from Nebraska, do not differ significantly either qualitatively or quantitatively from the range of specimens previously described within /schyrosmilus, and the dentary closely resembles the type of J. ischyrus. THE CITA CANYON HOMOTHERE This Homotherium may not be considered as H. serum with confidence because of the Blancan age of the Cita Canyon fauna and the late Pleistocene age of the Friesenhahn Cave remains (a dischroneity of some 2 X 10® years), and because of the longer limb bones and morphological differences in the basicranial and dental characteristics, although the cranial and dental dimensions of the Cita Canyon specimens generally fall within extremes of size observed in the Friesenhahn Cave H. serum. Cranial material of the Cita Canyon Homotherium is distinguishable from that of the Friesenhahn Cave H. serum on the following characteristics: 1. The incisors are not set in as rounded an arcade and are not as separated from each other. 2. The sabre (C') is more hypsodont and may be considered intermediate between H. serum (or H. ultimum of China) and a less specialized serrated condition. 3. P® is small and two-rooted (not single-rooted as in H. serum). 4. The upper carnassial (P*) has a protoradix. 5. Serrations on P4 may be restricted to the paraconid and metaconid. 6. The lower carnassial (M1) is serrated and is broadly concave on the medial surface rather than flat or shallowly concave. 7. M1 is proportionately broader buccolingually than Ps. The ratio is 3:2 (14.0:9.3 mm) in the Cita Canyon (WT 1238) and 4:3 (12.7:9.1 mm) in the Friesenhahn Cave H. serum (TMM 933-3231). 8. The tympanohyal pit and stylomastoid foramen are subequal. These structures are separated by a distance equal to the diameter of the foramen and set in a transverse line, rather than separated by a greater distance and with the tympanohyal pit lying anterior to the stylomastoid foramen at an angle of 45° to the midline. 9. The infraorbital canal is dorsal to the mesial root (pararadix) of P* rather than between the mesial and distal roots. These differences may not remain diagnostic as more specimens become available, but they give bases for comparisons with the Eurasian Pontian-to- Villafranchian sabretooths in which homothere characters are present. Considerations of descriptions and illustrations of homotheres in the literature (e.g., Zdansky, 1924; Schaub, 1934; Teilhard and Piveteau, 1930; Teilhard and Leroy, 1945; Teilhard, 1939; de Bonis, 1976) show that there was a gradual evolution in the specialization of the carnassial functions of P* and Mi: from late Miocene to late Pleistocene times. Variation from the typical condition of the dentition for any one time may be expected, so that the degree of modification of the dentition of a North American homothere provides only a rough check on its synchroneity with the Eurasian sequence, where the number of extensively preserved specimens is also not great. The conditions of the protocone and the mesial margin of the sabre are unknown in the Cita Canyon homothere, but the presence of a protoradix suggests either a vestigial cone or only a supporting root. The sabres were probably serrated on both edges as this is the normal condition. P? is absent, P? is double-rooted or may be present and variably developed or absent, and M! as represented by an alveolus is small and transversely elongated. This places the Cita Canyon homothere intermediate in its specialization between the Chinese Pontian (late Miocene) Machairodus palanderi (including M. tingii) and the late Pleistocene Homotherium ultimum, and at a stage of evolution comparable to that of H. cf. crenatidens, M. nihowanen- sis, M. inexpectatus, or M. maximiliani. In Europe a similar evolutionary stage is described for H. nestianum (sic, H. nestianus) from Perrier (Rocca Negra), France, by de Bonis (1976) from the late Villafranchian (Plio- Pleistocene Boundary). The variability of the presence or absence and development of P*, the protocone on P*, the transversely elongate or styliform crown of M/’, and serrations on the cheek teeth are not known for homotheres. In most deposits carnivores are relatively rare, and in all occur- rences of homotheres only two or three specimens are usually recovered, often incomplete. It is also likely that P® might reappear as an atavism even in otherwise highly evolved populations if the postcanine diastema could accommodate it. Moreover, the protocone on P* might be less likely to reappear because of the requirement that the carnassial occlusion be unhindered, and the protoradix might be variable in expression long after the protocone had been lost or reduced to a shallow style on the lingual face of the carnassial blade. The protoradix might provide some transverse stability to the carnassial under shearing stress but might be partly dependent on the thickness of the dentary at that point, as homothere dentaries are thinner than those of smilodonts. The conformation of the crown of M! also might be variable, in that its presence as an oblique blade distal to the upper carnassial does not always occur in Smilodon. In the mandible, Ps is separated from Ps in M. nihowanensis and H. nestianum but is close to Pa in H. ultimum or H. serum. The right dentary from Cita Canyon (WT 1238) preserves no Ps, although it was probably present. The Cita Canyon homothere is most similar to the Villafranchian specimens from China known as H. cf. crenatidens, M. nihowanensis, M. inexpectatus, and M. maximiliani, and especially M. nihowanensis. But the variation within this group of species suggests that the Villafranchian taxa constitute a record of evolving populations, with H. cf. crenatidens and M. nihowanensis representing conservative, and M. inexpectatus and M. maximiliani progressive, specimens. Of these names, H. crenatidens Fabrini, 1890 has priority. Mawby (1965:584) remarked on the ‘“‘many similarities between Homo- therium crenatidens and Ischyrosmilus johnstoni. .. . The dentitions are similar, with relatively short, flattened sabres, and high, narrow carnassials. There are numerous points of resemblance in the skull, including the form of the mastoid process, the shape of the bulla, the keel and depressions of the basioccipital and the division of the sagittal crest.”’ The homothere specimens from Cita Canyon, Texas, described by Mawby (1965) under the name of /schyro- smilus johnstoni, are here referred to Homotherium crenatidens, and the Chinese species known as Machairodus (Megantereon) nihowanensis, M. inexpec- tatus ,and M. maximiliani are also referred to H. crenati- dens Fabrini, 1890, with the Chinese species recognized as junior synonyms of H. crenatidens. THE STATUS OF MACHAERODUS GRACILIS AND THE SPECIES OF ISCHYROSMILUS All characteristics given by Merriam for species of the genus /schyrosmilus, except variation in size, are descrip- tive and accurately diagnostic of the Port Kennedy Smilodon gracilis specimens. When measurements of Merriam’s type specimens of J. ischyrus (UCMP 8140) and I. idahoensis (UCMP 22343) and J. crusafonti (Schultz and Martin, 1970) are compared with those for the Port Kennedy sabretooth (Table 4), in no case is there sufficient disparity for a single identity to be unlikely. While in general the measurements for/. idahoensis and J. Johnstoni are greater than those of the Port Kennedy and Florida materials, those of /. ischyrus and/. crusafonti are a5 within the ranges of variation that might be expected in an extensive series. The Port Kennedy sabretooth, originally described as Machaerodus gracilis and M. mercerii, or referred to as Smilodontopsis gracilis and Smilodon gracilis in sub- sequent literature, is therefore congeneric with taxa described within the genus /schyrosmilus Merriam, 1918. The acceptance of five species in this genus, founded on six occurrences for /. gracilis (Port Kennedy Caverns, Pennsylvania, Sante Fe I, Inglis IA, Haile XVA, Bass Point Waterway I, and El Jobean Pit, Florida), and single occurrences for J. ischyrus (Asphalto, California), /. idahoensis (Froman Ferry, Idaho), and /. crusafonti (Dan Bowman Ranch, Nebraska), with two possible individuals of J. johnstoni (Cita Canyon, Texas), appears unlikely. The four smaller species (Table 4) constitute a single taxon on the evidence of morphology and size, and the two larger specimens may represent either a second species or larger variants attributable to sexual dimorphism or simple size variation. Moreover, the extremes of measurements for S. californicus are similar to those observable in Ischyrosmilus (Tables 3-5). Size variation in [schyro- smilus may thus reflect either simple variation or sexual dimorphism, if all the specimens were contemporaneous. However, Evernden et al. (1964) and Mawby (1965) considered that /schyrosmilus was restricted to Blancan faunas. All known specimens derive from Blancan or Irvingtonian deposits, with those of J. idahoensis, I. Johnstoni, and I. crusafonti being later (early Pleistocene) and others earlier (late Pliocene or earliest Pleistocene). It is thus possible that /. idahoensis, I. johnstoni, and 1. crusafonti represent somewhat later populations in which the average size and massiveness were greater than in the better-known earlier populations. The genus /schyrosmilus Merriam, 1918 must take precedence over North American names, in that Smilodon Lund, 1842 is not available, Smilodontopsis Brown, 1908 is a synonym of Smilodon, and Homotherium Fabrini, 1890 does not apply to any of the type specimens. Cope (1880) originally described the Port Kennedy materials as Machaerodus gracilis and Uncia mercerti (1895) or M. mercerii (1899). Machaerodus is a variant of Machair- odus Kaup, 1833, which under Article 33(b) of the International Code of Zoological Nomenclature (Stoll et al., 1961:37) is an ‘‘incorrect subsequent spelling’’ and “‘has no status in nomenclature and therefore does not enter into homonymy and cannot be used as a replacement name.” The correct name for the Port Kennedy sabretooth is Ischyrosmilus gracilis (Cope, 1880), and Machaerodus is therefore neither a synonym nor available as a replacement name. The smaller species of Ischyrosmilus described by Merriam, /schyrosmilus ischyrus (Merriam, 1905) (= Machaerodus ischyrus Merriam, 1905), and 1. crusafonti 36 Schultz and Martin, 1970 (but not 7. osborni Merriam, 1919) are synonyms of /schyrosmilus gracilis (Cope, 1880), including Machaerodus (Smilodon) mercerii Cope, 1899, but the genotype is /schyrosmilus ischyrus, as Merriam (1918) founded the genus on the specimen previously described (1905) as Machaerodus? ischyrus. The status of Ischyrosmilus idahoensis Merriam, 1918 (including /. johnstoni Mawby, 1965) is dependent on the significance of greater size. At present, because of the paucity of specimens, and because no evidence exists to distinguish /. idahoensis from the /. gracilis specimens on criteria that are independent of possible sexual or normal size variation, it is wise to consider /. idahoensis synonymous with /. gracilis. The geological dates given for the /schyrosmilus specimens are late Pliocene for /. ischyrus and I. Johnstoni, late Pliocene or early Pleistocene (Blancan or Irvingtonian) for the Florida specimens of /. gracilis and for I. idahoensis, early Pleistocene for /. crusafonti, and Pleistocene for /. gracilis (including /. mercerii). Actual age of the Port Kennedy Cavern deposits is probably less certain than that of any of the other /schyrosmilus-yielding deposits (Kurtén and Anderson, 1980), as stratigraphic evidence within caves can be misleading, and the associated faunal remains either inadequate for dating or dischronous. Hibbard (1955:94) considered the Port Kennedy mic- rotine rodents ‘‘more advanced than those occurring in the earlier Pleistocene faunas such as the Hagerman of Idaho, the Sand Draw of Nebraska, the Deer Park and Sanders of Kansas.’ Latest ages for these faunas are: Hagerman 3.7 My, Sand Draw 3.1-2.8 My, Deer Park 3.1—3.0 My, and Sanders 2.8 My, approximately, and all represent Blancan faunal assemblages (Lundelius et al., in press). It is likely that at least part of the Port Kennedy fauna dates from the later part of the Blancan, or from the early Irvingtonian land mammal ages, and that an early Pleistocene date is probable. Evolutionary trends in /schyrosmilus appear to be from a smaller and more gracile form toward a larger and more robust one. The temporal range of the genus may be restated as early Pliocene to early Pleistocene in North America, and its distribution to include the areas of the present Great Plains, Idaho, California, Florida, Pennsyl- vania, and northwards, possibly as far as the Canadian prairies, in a pattern reminiscent of that of its later descendant Smilodon. At present the known sample is too small to permit suggestions of regional variants, although the specimens from Florida are the most numerous and indicate considerable variation in individual size. RELATIONSHIPS OF THE GENUS ISCHY ROSMILUS The first description of specimens of /schyrosmilus from the Port Kennedy Fissure, Pennsylvania, and Asphalto, California, were within the genus Machaerodus (Cope, 1880; Merriam, 1905), and both Cope and Merriam commented on resemblances to Eurasian species of Machairodus. Mawby (1965:584) presented an opposing view: “‘It is likely that direct comparison of J. johnstoni with specimens of H. crenatidens would show the two to be congeneric.’’ He added that ‘‘it seems best to retain the name /schyrosmilus for the early Pleistocene homotheres of North America.’’ However, this opinion was probably influenced by the homothere characters of the skull and hind limb of the Cita Canyon sabretooth specimens. Taxonomic confusion has obscured the systematic relationships of Pliocene and Pleistocene machairodonts of the Holarctic and Ethiopian regions in the past. They differ from Felis in the presence of well-developed mastoid processes, reduced coronoid processes on the mandible, elongated and massive mandibular symphyses, elongated and hypsodont (sabre-like or dagger-like) upper canines, premolars reduced both in number and in effective shearing length, and strong and trenchant carnassials. Their bodies were strong to massive, the tail apparently always short, as in Lynx, and the limbs strong with powerful claws. All were large, similar in size to the lion (Felis leo), leopard (F. pardus), or jaguar (F. onca), and readily distinct from these extant large cats on morpholog- ical differences. Piveteau (1961:786) avoided some of the taxonomic problems within the Machairodontinae, referred to them as the ‘‘Machairodontoid Lineage’, and commented “‘Il y a toujours eu une grande confusion dans la nomenclature de ces formes; d’abord dans les désignations spécifiques quand elles étaient toutes rangées sous la dénomination de Machairodus; ensuite dans les désignations génériques quand le genre initial fut a son tour découpé, parfois d’une fagon abusive et sans justification suffisante, en de nombreux autres genres.”’ Croizet and Jobert (1828) described Megantereon from the Villafranchian of Seneze, France, and Kaup (1833) described Machairodus from the Pontian of Eppelsheim, Germany. These two genera represent two already separate lineages of sabretooths, recognized by Kurtén (1963) as the tribes Smilodontini (dirk sabretooths) and Homotheriini (scimitar sabretooths), respectively, but named from the middle and late Pleistocene genera Smilodon and Homotherium. Megantereon is known from Eurasia and Africa and has been reported from the Blancan of North America (Schultz and Martin, 1970; Kurtén and Anderson, 1980) and also from the late Hemphillian (Berta and Galiano, 1983). It was as big as a cougar or panther and increased in size through time. Its dentition is characterized by strongly hypsodont upper canines that are slim and dirk-like and lack serrations on their margins (Figs. 9A,B, 12A). The upper third premolars, P*’s, are somewhat reduced but functional. The upper carnassials, P*’s, possess single parastyles, paracones, metacone blades, and _ strong protocones, and the upper molars are transversely elongate and reduced. The lower carnassials, M1’s, lack evidence of metaconids or talonids, except as cingula. The sabres, C!’s, occlude posteriorly free of the lower canines and within the lower diastemata, and functional P3’s are present. The skull is massive, with a short muzzle, convex dorsal outline, low nasals, and flaring nuchal crest. The glenoid process extends to the level of the external auditory meatus and down to, or below, the occlusal plane. The mastoid process is large and lies posterior to the meatus and bulla. The mandibular symphysis is strong, vertical, with well-developed genial flanges almost double the depth of the dentary in the diastema. Machairodus is known from the Pontian of Eurasia, and from Africa, and is reported from North America by Martin and Schultz (1975, Hemphillian and Kimballian), Martin (1980, Hemphillian or late Clarendonian), and Harrison (1983, late Hemphillian). Kurtén and Anderson (1980) did not recognize this genus in North America but suggested that Homotherium (Fig. 11) might derive from that genus. Machairodus was the size of a lion. Its dentition is characterized by a less hypsodont scimitar-like canine that is broader mesiodistally than that of Megante- reon, proportionately more compressed buccolingually, and strongly serrated on both edges. The sabre occludes past the lower canine’s distal edge, with the distobuccal ridge on the lower canine serrated, and the incisors set ina rounded arc. The second upper premolars are single- rooted and reduced, if present. The second lower premolars are larger and two-rooted with differentiated crowns, and the third and fourth premolars may have serrated margins. The upper carnassials have variably developed protocones, integral with the side of the paracone, strong shearing blades, and double parastyles, and the upper molars are oval or round, but not usually transversely elongate. The lower carnassial is trenchant and may have vestigial metaconids or talonids. The skull of Machairodus is low-crowned, with a less convex dorsal outline than Megantereon, the nasals are higher and the muzzle longer, the infraorbital canal is oval in section, and its diameter is less than 40 per cent of the maximum orbital diameter, the tympanic bullae are not highly inflated, the mastoid process is massive and rugose, and the glenoid process extends to the occlusal plane. The mandibular symphysis is inclined at an obtuse angle to the axis of the horizontal ramus, and the genial flanges are smaller than in Megantereon. The limbs are longer and the claws less massive. The salient points in which Megantereon differs from Machairodus are the more dirk-like sabres that lack serrations and which occlude distal to the unserrated lower 37 canines, unserrated incisors, distinct protocones and double parastyles on the upper carnassials, the trans- versely elongated upper molar, the absence of a rudimen- tary upper second premolar, the vertical symphysis with the large genial flanges, the developed lower third premolar, the lower nasals and the convex outline of the skull, the inflated bullae, and the shorter and more massive limbs and claws. Ischyrosmilus is similar to Megantereon in the lower nasals, probably a convex dorsal outline to the skull, the proportions and section of the sabres and their occlusion entirely posterior to the lower canines, the simple parastyle and the occasional presence of a protocone on the upper carnassial, the vertical orientation of the mandibular symphysis with sharp genial flanges, and the transverse alignment of the incisors and lower canine. The mastoid and glenoid processes are ventrally elongated and approach each other but do not almost touch, as in Smilodon. Smilodon (Fig. 12B) is derived from the Megantereon condition by further emphasis on the hypsodonty of the sabre, reduction of the third upper premolars, the almost completed loss of the third lower premolars, and greater development of the sagittal and nuchal crests. Smilodon also has serrations on the margins of the upper canines and on the distobuccal ridge of the lower canine, but these are less wel! developed than in Machairodus or Homotherium and are different in form. It had newly evolved an elongated mastoid process that extends anteriorly and ventrally to below the level of the external auditory meatus and to within | or 2 mm of the postglenoid process. Ischyrosmilus gracilis (Fig. 1) represents an inter- mediate transition towards that of Smilodon in_ the reduction of the third lower premolars, development of fine serrations on the canines and other teeth, and an elongate mastoid process that approaches the postglenoid surface to within 5 mm. The Neogene machairodonts therefore represent two distinct lineages which had a common Eurasian origin during the early or middle Miocene (Fig. 13). The separate lineages, begun by Machairodus and Megantereon, were well established by the early Pliocene and widespread across Eurasia and Africa. By Blancan or Villafranchian times, the Machairodus lineage had evolved from the level exemplified by M. aphanistus to that of Homotherium (= Machairodus) crenatidens and was trending towards the advanced level shown in Homotherium (= Epi- machairodus) ultimum or H. serum. The changes involved in the evolution of Homotherium from Machairodus appear to have been so gradual, and the variations in the morphological characters within coeval populations so great, that the examples of the populations available do not provide characters to distinguish the grades, nor does the fossil record provide convenient accidental gaps by which to separate genera or species. Megantereon is known from the earliest Villafranchian in Europe, possibly from the Pliocene of India, and became extinct in Europe at the end of the Villafranchian but survived in Africa and Asia into the middle Pleistocene. By early Pliocene times the lineage may be represented in North America by Megantereon hesperus, from which a new species, /schyrosmilus gracilis, may have arisen by late early or middle Pliocene times. This species later differentiated into the large and successful New World sabretooth Smilodon, known from middle Pleistocene times to after the retreat of the Wisconsin ice and from both North and South America. The smilodont lineage paralleled the homothere in the development of serrations on the teeth, albeit different in detail, and in the reduction of Ps’s. This lineage introduced a new conformation within the Machairodontinae in the close apposition of the mastoid and postglenoid processes in response to a variant mechanism for closing the jaw. Similar modifications for the areas supporting the glenoid had occurred earlier in the Oligocene Eusmilus and middle Miocene (Clarendonian) Barbourofelis . Kurtén (1963), Slaughter (1963), and Webb (1974) referred the Port Kennedy specimens to Smilodon, and Martin (1980) referred them to Megantereon. Ischyro- smilus gracilis can be distinguished from both Megante- reon and Smilodon by size and morphologies which show interesting intermediate evolutionary stages. The evolutionary changes that took place in the smilodonts were greater than those in the homotheres, which effectively experienced only refining of already-evolved structures. The smilodonts continued to reduce the premaxillary dentition, separately acquired serrations on the canines, and altered the conformation of the auditory region. [schyrosmilus is as distinct from Megantereon as it is from Smilodon, and thus it is a morphospecific taxon which, because it may be recognized qualitatively and without recourse to fine measurements, probably deserves generic rank. This may only be confirmed by the discovery of additional specimens. Fig. 13. Selected occurrences of Homotheriini and Smilodontini in Eurasia and the Americas, arranged as lineages. Times of the original migrations from Eurasia to North America or from North America to South America are approximate, as are temporal positionings of the occurrences. Later exchange migrations across the Bering Strait in either direction or 38 southwards across the Panamanian Isthmus after the original immigrations are not indicated. Time-Scale in 10® years (My) and Magnetic Reversal Scale in black (normal polarity) and white (reversed polarity) column. HOMOTHERIINI My EURASIA “Machairodus’ 1 inexpectatus Homotherium (= Epimachairodus ) ultimum , T} Choukoutien, China Homotherium crenatidens partim: Rocca Negra, France Machairodus nestianus M. nihowanensis partim 3 Nihowan, China Machairodus maximiliani Machairodus aphanistus Machairodus cultridens N. AMERICA Friesenhahn Cave, Texas Homotherium (= Dinobastis ) sarum Homotherium crenatidens Cita Canyon, Texas “Machairodus. coloradensis Cheyenne Co., Neb. EURASIA Megantereon spp. Eurasia M. crenatidens partim: Rocca Negra, France M. nihowanensi partim Nihowan, China Megantereon sp. India Megantereon megantereon Europe SMILODONTINI N. AMERICA Smilodon fatalis Ischyrosmilus gracilis Port Kennedy, | Pa. Inglis 1A, Fla. 0 Haile XVA, Fla. Ischyrosmilus gracilis Brayfields’ Pit, Fla. $ Megantereon hesperus S. AMERICA Smilodon neogaeus Smilodon ensenadensis Pal © CoEINGE PLEISTOCENE MIOCENE Confusion has long existed over the identification of sabretooths within the Machairodus-Epimachairodus- Homotherium (including Dinobastis) and _ the Megantereon-Ischyrosmilus-Smilodon lineages. Schaub (1934) discussed this confusion and the taxonomic problems of separating Machairodus or Epimachairodus (= Homotherium) from Megantereon in the early Pliocene. The inclusion by Teilhard and Piveteau (1930) within Machairodus nihowanensis of specimens from the late Pliocene of Nihowan, China, that are assignable to both Epimachairodus crenatidens and Megantereon, and the inclusion by Bielawski (1905) within Machairodus nestianus of specimens from the late Pliocene of Rocca Negra, Perrier, France, assignable to both Epimachairodus crenatidens and Megantereon, only added to this confusion. Mawby’s (1965) inclusion of both a homothere (Homotherium) and a smilodont (/schyrosmilus) within Ischyrosmilus johnstoni has already been discussed. De Bonis (1976) reviewed the machairodonts from Rocca Negra, France, reconsidered the skull described by Schaub (1934), and identified it to Machairodus nestianus as Homotherium nestianus (sic) from the Pliocene- Pleistocene boundary (middle Villafranchian). He (1976:17) suggested that the differences between H. nestianum and H. crenatidens are insufficient for a specific separation. Schaub (1934:405) had already suggested, after studying the same skull from Rocca Negra, that the distinctive specific characters invoked by Fabrini (1890) had value only at the level of individual or sexual differences. De Bonis (1976) considered that Meade (1961) included within Dinobastis serus the two crania TMM 933-3582 and -3231 of which TMM 933-3231 represents /schyrosmilus. He stated (1976:20) that in TMM 933-3231, C!, ‘‘la canine supérieure est proportionnellement plus longue et semble plus finement crénelée, la troisieme prémolaire supérieure est plus longue, la carnassiere est plus courte relativement a la taille de l’animal, le profil supérieur du crane est plus bombé et surtout l’apophyse glénoide est plus longue et la cavité glénoide se trouve nettement au-dessous du méat auditif externe.’’ Cranium TMM 933-3231 is an adult with fully erupted sabres, whereas TMM 933-3582 has sabres deeply set within the alveoli, probably as a result of postmortem compaction into the alveoli. The dorsal outline of TMM 933-3231 appears more rounded because its nasals are not preserved. P®’s cannot be compared as none is extant in TMM 933-3582. The right P* of TMM 933-3582 differs by less than | mm from the left P* of TMM 933-3231 (37.6 X 12.7 versus 38.6 X 12.0 mm for mesiodistal and buccolingual diameters) and thus is longer, although the total lengths of the two skulls show TMM 933-3231 to be longer (362 versus 324 mm for the premaxillary-condylar length). 40 Moreover, the level of the glenoid surface in TMM 933-3582 relative to the external auditory meatus is the result of a combination of minor distortions of the fossil and the artistry in Meade’s (1961, fig. 1) illustration. De Bonis’s (1976) contention that /schyrosmilus is also present in the Friesenhahn Cave materials assigned to Dinobastis serus by Meade is erroneous and appears to be based on his misinterpretation of Meade’s illustrations and the possible influence of Mawby’s (1965) identification of a homothere skull as the smilodont /schyrosmilus Johnstoni. Martin and Schultz (1975) described and figured Machairodus cf. coloradensis from the Kimball Forma- tion (Upper Pliocene) of Cheyenne County, Nebraska, and from the upper part of the Ash Hollow Formation (Pliocene, Hemphillian) of Sherman County, Nebraska, the latter as a new subspecies, M. coloradensis tanneri. They concluded (1975:55) that their ‘“‘new subspecies demonstrates evolutionary trends which seem to be leading to Ischyrosmilus from the lower Quaternary (Blancan).’’ It appears that they also relied on Mawby’s (1965) description of /schyrosmilus johnstoni in their evaluation of the relationships of their specimens and in their concluding remarks. Their two Machairodus are typical of the early stages of the genus sensu lato, probably at the level of Homotherium crenatidens or H. palanderi, and show none of the derived characters that typify the Ischyrosmilus stage in the Megantereon to Smilodon lineage. Martin (1980:149-150) compounded the Martin and Schultz (1975) error when he stated: ‘“‘Along with Megantereon, we find a _ long-legged, short-tailed scimitar-toothed cat (Homotherium) occurring in Eurasia, Africa and North America. The North American form has been described as Ischyrosmilus, as 1 have generally supported the separation of the two genera (Schultz and Martin, 1970; Martin and Schultz, 1975), but I now regard them as congeneric. A related genus, Dinobastis, has a cruciate sulcus on the brain (based on the McPherson Equus Beds specimens of Dinobastis), and | regard Homotherium and Dinobastis as felids.’’ Martin’s first two sentences state that Homotherium and Ischyrosmilus are congeneric. His third sentence places Dinobastis (= Homotherium) in the Felidae. All the genera named are Felidae, but /schyrosmilus belongs to the tribe Smilodon- tint and Homotherium and Dinobastis to the tribe Homotheriini, and although the latter two may be congeneric, Homotherium is not a possible congener of Ischyrosmilus. Last, to my knowledge, the scimitar cats (genus Homotherium) have never been intentionally placed in Jschyrosmilus, except by Martin (1980), probably because their dental and basicranial characters are essentially dissimilar. However, Mawby (1965) referred Ischyrosmilus to the Homotheriini fide Kurtén (1963:98) where ‘“‘/schyrosmilus?’’ is referred to as a homothere. The confusion that has existed concerning identification of the homothere and smilodont lineages still exists, despite the diagnoses and efforts of previous workers (e.g., Pilgrim, 1931; Schaub, 1925, 1934; Kurtén, 1963; Churcher, 1966; Kurtén and Anderson, 1980). It is necessary to have a clear concept of the two functional conformations and to decide to which lineage a specimen belongs on the basis of a knowledge of their morphologi- cal differences before attempting to place a name on the specimen. Some of the confusion appears to have occurred because of reversals of this procedure, and the inclusion of specimens from both coeval and sympatric but different lineages within a single taxon. Conclusions The names that have been applied to the gracile sabretooth Ischyrosmilus. gracilis are: Machaerodus (Smilodon) gracilis Cope, 1880; Smilodon gracilis Cope, 1895; Uncia mercerti Cope, 1895; Machaerodus (Smilodon) mercerii Cope, 1899; Smilodon (Smilodontopsis) gracilis Merriam and Stock, 1932; Machaerodus? ischyrus Merriam, 1905; Ischyrosmilus? idahoensis Merriam, 1918; Ischyrosmilus ischyrus Merriam, 1918; J. johnstoni Mawby, 1965; J. crusafonti Schultz and Martin, 1970; and M(eganterion) gracilus Martin, 1980. These names have been applied by various authors (e.g., Slaughter, 1963; Kurtén, 1965; Webb, 1974; Robertson, 1976; and Kurtén and Anderson, 1980) in different ways to refer to specimens included within these taxa. Machaerodus Cope, 1880 is an invalid variant of Machairodus Kaup, 1833. The genus Smilodontopsis Brown, 1908 was first applied to true Smilodon and thus is a junior synonym of Smilodon Lund, 1842 and therefore unavailable. Smilodontopsis troglodytes and S. conardi are therefore true Smilodon. Ischyrosmilus gracilis (Cope, 1880) is the first de- scribed species of the genus /schyrosmilus Merriam, 1918, of which the type species is /schyrosmilus ischyrus Merriam, 1917, originally described as Machaerodus? ischyrus Merriam, 1905. The genus /schyrosmilus com- prises the species /. gracilis (Cope, 1880), with/. mercerii (Cope, 1895), /. ischyrus (Merriam, 1905), and /. crusafonti Schultz and Martin, 1970 as_ conspecific synonyms. Larger specimens described as /. idahoensis Merriam, 1918 and /. johnstoni Mawby, 1965 are probably only sexually dimorphic or morphological variants of J. gracilis but may constitute a later taxon for which the name /. idahoensis Merriam, 1918 is available. Ischyrosmilus is intermediate between Megantereon and Smilodon in characters of its dentition and jaw articula- tion. It evolved from an immigrant Megantereon, possibly M. hesperus from the early Pliocene of North America, is found in deposits of early or middle Pliocene to early Pleistocene ages (Blancan and Irvingtonian faunal ages), and probably evolved into Smilodon by middle Pleistocene times. Smilodon became extinct probably in early Holocene times. /schyrosmilus belongs to the tribe Smilodontini and possesses the dirk-like sabres, the ventrally extended glenoid process, the reduced lower third premolar, and other characters typical of the tribe. 4] Acknowledgements This report stems from my interest in the genus Smilodon sensu lato and from conversations with many vertebrate palaeontologists about it. I have to thank the curators, assistants, and students in many North American museums who since 1965 have assisted me, allowed me to examine and borrow specimens, and discussed the anatomy and phylogenies of sabretooths. I am particularly indebted to Dr. Bjorn Kurtén of the University of Helsinki with whom I have corresponded through the years about Smilodon, Dinobastis, and Ischyrosmilus johnstoni; Dr. David Gillette, then of the Academy of Natural Sciences, Philadelphia, who allowed me to examine the type materials of Smilodontopsis troglodytes, S. conardi, and S. mooreheadi; Dr. Theodore Downs of the Los Angeles County Museum, Los Angeles, who allowed me to examine the collections of Smilodon californicus from Rancho La Brea; Dr. Ernest L. Lundelius, Jr., of the University of Texas and Texas Memorial Museum, Austin, who allowed me to study and measure the collections of Homotherium serum from the Friesenhahn Cave; and Dr. S. David Webb of the University of Florida and Florida State Museum, Gaines- ville, who allowed me to examine and refer to the cast of the Inglis IA dentary assigned to Smilodon gracilis . Dr. Annalisa Berta, who is currently studying the Florida sabretooths, discussed the phylogeny and history of the Pleistocene sabretooths, read and commented on my paper in draft, and allowed me to read a draft of her paper, ““The sabrecat Smilodon gracilis from Florida and a discussion of its relationships’’. Dr. Kurtén, Dr. John E. Mawby of Deep Springs College, California, Dr. Thomas S. Parsons of the University of Toronto, and Dr. Christopher McGowan of the Royal Ontario Museum, Toronto, read drafts of this paper and offered critical advice. Miss Airlie Armstrong and Miss Judy Hann, my technical assistants, typed the manuscript, and, with Dr. James A. Burns, then of the Department of Zoology, University of Toronto, now of the Provincial Museum of Alberta, Edmonton, checked for inconsistencies and errors with great good humour. Mr. John Glover of the Faculty of Arts and Science, University of Toronto, Photographic Facility, produced the photographs used for the plates and those from which many of the line diagrams were made efficiently and quickly. Last, and most important, I thank Mr. and Mrs. William Brayfield of El Jobean, Florida, and Dr. A. Gordon Edmund of the Royal Ontario Museum, who collected at the Bass Point Waterway and El Jobean Pit sites and allowed me to incorporate their new materials into this report. Without their interest, patience, and support, this paper would have lacked the new data by which many of the conclusions are substantiated. ASG yisuay [BIOL (Z) 8°S-€'8 UL VP Ue jys1ay UMOID (OL SSW O'L “89 JaWeIp [ensuljooong (7) $'9-7'01 L6 ‘101 IOJOUWIP [BISIPOISAP fq (LOS WSO) = PSI Gash apr[q auocrjawW Jo yWBueT (9€) €:91-M9'I7Z = csI aT'LI QuOdRIDW IYSIOY UMOID (61) 9°07-6'F7 6!) 79 juasqe au0d0}01d 19A0 jYSIOY UMOID GO) Fl -€ Le = 96 = auoovied 19A0 JYysIay UMOID (81) L'8-6'II 96°91 - 6'LI ajnuooeied 19A0 1YysIay UMOID (pZ) Juasge—9T ‘p] SE €°8 6'8 ajAqsered 19A0 1Yys1ay UMOID (6b) I 11-881 aL Fl 9°SI Lvl Ja}QWeIp jensuljooong (8b) p €€-€ OF : GILE CLE JQ}OWLIP [PISIPOISAW .d IGE £00! ele JUDWSPIy JO yYBUFT (pI) €°87I-L'ISI (9°L6) 1UsI9Y UMOIO [RISIG (ST) 90'€ZI-F Ibl (€001) IYSIoy UMOIO [RISA (8€) €°SI-1'P7Z (8°€1) J9}aWeIP [ensuljooong (Zr) O'9€-20'1S (€'1€) JO}WIP [PISIPOISAW ,D (9) 0'Ob-7'8r 6 9F yyBua] [RIO] (€€) 9 vI-1 7 €°072 1yslay UMOID (8h) 1'OI-9'F1 9TI I9}OWeIp [ensuljooong (SO) SUSIE L‘6 JO]OWIP [BISIPOISAW, z] (p) O'SE-28° Sb : 7'Sb yisug| [P10] z SIE yisua] 100y (S€) 9 TI-9E'81 6'€1 rol 1YysIay UMOID (br) 8°8-L‘TI 68 9°6 Ja}2WeIp [ensuljooong (97) 0'S-9'L ee 0'8 J9}OWLIP [RISIPOISAW ;] eas oidutes. =. (N) L8L11 HNIW O6LI1 HNWY O8L11 HNIWY sesueyly sesueyly sesueyly nog ‘eIe[eL pue ‘eIUsOsITED ‘oInssiq preuoDd ‘QINSsiy pIeuoDd ‘gINSsIy pleuoDd suolsuawiq ‘eolg eT] OYyouRYy Wo poAlop 1paDuod sajxpo]soy pue UOPO]IUG IO} Sd9BuUeI IZI1S {uopopMs sisdojuopopius sisdojuopopius sjuawa|q ‘Q[QeINSPOWIUN SI UOISUSUTIP dy) JU) ,,-,, PUR ‘UOISUSWIP d}9;dWIOSUT UR SosoyUoIed *UIOM SI YOO} ay} JO UMOID ay} WY) ,.M,, ‘UONBUT]SO UP Sd}BOIPUT ,.9,, “SONOUIIT[IU UT ore SJUDOINSeaYy “JOYINe oy) Aq SJUSWOINSROW WO’ puR (ZEH[) YOOIG Pue WLU WO] POALIOp ore ‘nJog ‘PIR[e], pue ‘eIWIOJI[ED ‘vag eT] OYOUPY WO UOpO]IUG JO SUOISUSUIP Ie[IWIs 10} SABRI IZ1g “JOYINE dy) Aq UdyR) SJUSWIINSBa| “SESUBYAIY ‘IINSSLY PABUOD 9Yy} WOIY CUOpo]NUS 0} pausIsse sjUdua[a [eJUapP payelosse Jo pue 1psvU0d “¢ puke sajdpojs0.4) sisdojuopojNus yo sadAyojoy [ejUap ay} JO SyuoWaINseayy T IGQFL (81) €°S7—-'SE LIzé THT JOQOWeIP [esIpoutxoid wiNnwIxe/y (LI) 9: 1€-L'bh 0'87Z L°@€ yidap sewyedosiop wnuwixeyy Yel ne) (Jacky woyisid Buryory) reunjoydessg (9) 687-7 7E Cb yeys JO Ja}0WeIp JoUIa}sodoi9}ue WINUTUTI] (91) I°S7—-O'€E ECS YeyYs JO IOJOWeIP ISIOASURI} WINWIUIY] (07) 9'SE€-S' Ih Eve J9}9WIP [e}SIP JOUa\sodos9}ue WNWTXeYy (07) O'SP-€'€9 Tes JOOWEIP [BISIP ISIOASURI) WINWIxe|| PIQIL (OZ) 9°S€-7 8r PTE J9WeIp [eISIP JOWa}sodos9jue WINUTxe|| (OZ) b 6b-€°L9 €or JOIWIP [BISIP ISIOASUBI] WINWIXe|y Yel snipey (IZ) 1° 61I-8°€€ 7 07 €'6l IQJOWLIP [BIESIP ISIOASUL) WINUITXeJ (1Z) €'1€-9 9b Pee TEE J\IWIP [eISIP JOUIA}sodosa}uK WNUTXe] Yel Vel euln (SZ) Z°SZ-0°9€ Elz 9°€Z Jeg]yooy JeuyN Jo Ja}9WeIp WNUTTUT (ZZ) L’€6-€ 8TI 26° LL €°98 IOJOWLIP [B}SIP ISIOASULI] WINWIXe| Yysu Ye] sniouiny] CDT LS8sGell = 6°78 sisXydida [enuas JoWa}ue Jo winyUad sAoge oulds [eInau Jo yYyZIOH (81) 9'P7-0'SE 9°L7Z €°67Z 6°87 sisXydida yenuss s0La}sod yo dag (81) S'8€-F 8 Lit L€p € lr stsAydida [eUsd JOLIA}sOd sso1Oe YIPIAA (91) p'7E-97S - OIr 20'6P ypim jeaskydodesAz1s0g (91) 99° Lp-9IL - 2p €S - ypIM easkydodeiajy (91) 0°6S-6'18 0'€9 8°6S wBua] jeaskydodeshz (81) p'6€-v 7S 7 6r Z'0S LLY winUsd Jo WBUeT ¢ pure “p ‘¢ “SON Cé (ON bi “ON €i “ON aeIqeHaA Jequiny] (€v) 791-7 SZ 091 g[Apuod Jo JO}9WIeIP [eNUSAOSIOG (pr) €'6€-7'8S 9°Ob a[Apuos jo Ja}9WeIpP [ensuljooong Arequoqd zis ajduies = (N) L8L11 HNWY Z6LIT HNWY 98L11 HNWV Niog ‘eIeyey, pue sesueyly SUOISUDWIG ‘eIWIOJI[eD ‘vag eT] OyoUeY ‘gnssiJ preuod sesueyly ‘oimssiy preuod sesueyly ‘oinssi.j preuog pue WOIJ PSALIOp UopojIMs (uopojIMs ipapuo) sisdojuopopims sajkpojso4] sisdojuopojius s}udwa|q JO} SadueI IZIS ‘Q[QRINSLOLIUN SI UOISUSWIP dy) WEY) ,,-,, PUR UONBUWISO UL SOIVIPUT ,.9,, “SOIJOUN][IW UI JI S]JUDUAINSeIP, “JOYINE ay) Aq sJUSWIAINSeAW WO] pure (ZE61) YOOIG pue WILD] WO POALap sie ‘NO ‘vIL[LL, pue ‘eIUIOJI[eD ‘Bog eT] OYUeY WO VOPO]NUG JO SUOISUSWIP Ie[IWIS JO} S9dULI ZIG “sesUBYIY ‘d1INSSIY PIBUOD IY) WOI ‘ ;UOpOIMS 0} paUsdisse s}uUdUII]a [eIUe19}S0d pajelsosse ay} JO pue ‘1pspu0s ‘¢ puke Sajk{pojdo0.4j sisdojuopopimg yo sadAjopoy 9y} Jo (sedueyeyd jdadxa) syudua]9 [eIUe.ID}SOd JO sjUdWIAINSBIy, Z AGU (SZ) 0°€8-9'601 = 20001 = ysus] wWnwixey YI YI YI mW AI Il Il erpediese oy (ST) pb I1Z-L 62 LEGG . J90R] [BSIeIe}OW SsOIOR YIPIM WINWIXey (II) 9°€I-¥'81 9°SI Ja9WeIp [eIsIpowlxoid wnwixeyy (SI) I'S7-6'8P OLE y3ua| reueldosiop wnwixeyy Ys wuojJ19UuNd0}9q (S$) 96-7 FI € Ol SsoUyorIy) [eIsIpoulrxoid wWnwixeyy (S) 6 01-Z'El 6 TI yigus] UNWIxew 0} [PULIOU JO}OUIeIP aSIOASUPI], (S) Z 17-982 9'PT ysue] Wnwixeyy Yel ULIOJIOUNdOSa|y (O€) L'97—-O'8E 10€ ¢'0€ 1°87 peoy Jo Ja}0WeIp WINWIxe| (1€) O'17-S$ 67 8°77 9°77 S17 * -yoou Jo JO}OWeIP WNUIIxe| (TE) 0: 9b-7'19 bSP 887 (OG JOOWIP ISIOASUBI) WINUITXR|y (€€) 9'pr-7' 19 CLP LLy err yi3us] WNWIxey| su 2] su snjesensy (ZZ) 1° S€-6 8h CLE goRjIns [e19}e] uO ydap wnuTxEyy (27) S'7E-T bP BSE }20R] PlOgNd J9A0 YIPIM WINWIXeY] (ZZ) v' O-PS Lb UIN[NeIWAISNS I9AO YIPIA WNWIXe| (ZZ) p'6L-8 901 v'r8 wN3us] UNWIxey Wsu umnouesed UOMS THEA Giz 1°SZ uidap sejuejdosiop wnuixeyy (iC)eElZ=y LE v'SZ JO}OWIP asIdasueI) WNUTXeYy (IZ) L'p@-¥' IE 9°€7 JaaUIeIP [eIstpoutxoid WINWIxe|] 2] ploqng (91) p'87-F LE 9°87 JO}WIP asJOAsueIL, (91) 0'L7-0'8r Lp yydap sejuejdosiog su Te[NOIAeN (91) S'6I-I°L7Z Ly? FSC Sil YIPIM [esIOp WNWIxe (91) L'87-8 OP 6'0€ €°67 b'0€ a}2uUeIp JeuUR]dosiop WNUIXe] (91) 8° €7-7 PE 9°LZ ONG 9°LZ JO}OWIP [PISIPOWIXOIg isu isu Yl WOsIIUA) (IZ) p 1I-€'91 Gael peoy Jo Ja}0WeIp [eIsIpourxosd wNWIxeyy (IZ) pb’ LI-9'67 LL peoy Jo Jo}OURIp 19RD (IZ) b' 17-8°97 1°07 SUONP[NIIPL IOAO JaIWeIP 1d}BIID (IZ) 0:9€-¥ 6 7'SE yisuay wnwixeyy isu WLOJISI - - (OZ) 96°ZEI-90'OL1 - 28 °8ZI (WAI yiaue7q (Z) 90° €E€1-O'SST ¢ trl (9€) 7 PEI-T7 89I 28°67 eZ rd ysueq (9) 79€-9 TP SCE (SZ) O'E7Z—-98 TE 99°07 ey 17 ela] Wisusy] 0981 LM Ise i azis ajdures = (N) azis ajdures = (N) Sexo], su YI apis ysu Ys YI su YI SUOISUSWIG ‘davQ UYyeyUasoLy SOIT WSNA €0SSZ WSNN Sexo], OS pure Lp dSNV tb dSNV pue ay} WOI] peAWap = BYSLIGAN ‘OSI = eYseIgaN ‘JoWeMpeoIg ‘uokURD BUD = Mad ‘eIe[e], pue “eIWIO}ITeD elueaA[Asuudag ‘Ssulaaed Apouuay YWoOg suewelq wniayjOuo HY] ‘ds quofosnsd _juojsuyol,, ‘ealg v7] OyoueY Wo peAuap Mdadsau (°§) “FW ‘syiovas (°S) “W JOJ SdBdueI IZIS SN]MUSOAKYIST SN] NUSOAKYIS] SNUSOAKYIST UOPO]IWG IO} SABuvI IZIS suourloads (uopojilus) snposanyovp ‘podojaaap 30U SOUIAWIOS SI Ja}ORIeYO ay) Jey) ,.JUASGe,, PUL ‘a[/qeINSeSWUUN SI UOISUSWIP dy) IVY) ,,-,, “YJOO] B UO UMOIO UIOM B , .M,, SUOTJBUIT)SA UB SIJBSIPUL , 2, , ‘UOpoOpUG YIM SUsUIdads ApouUsyY WOg ay) Jo sasues SurddepiovA0 dJedIpU SYS9}Se ‘SOTJOUWI[[IW UT oe s}UIWIINSeIYY “JOYING 9y) Aq Udy) 919M SJUIWOINSeIW JOYyIO {WIWOD ‘siad ‘(7861 ) USUNyY wolg “Iq Wo ‘ds SN]NUSOAKYIST pure NUO{MSN4I *] JOY asou) puke ‘(S96T) AQueyy Wo Ape 1uojsuyor , .“],, OJ 9souy “(ZEGT) JOOS puke Weis] Wo poauap Ajued ae Uopopiusgs 10} eye (6681) edog Aq sjuswansvoW JnoYIM pojOU puke anSopeywd dSNV ay) Ul 149249 (“§) “Py JO S9dAjOID9] DY} JO aUO se payeusisap ‘,g] ‘Ie[owlod psy) Joddn ya] ay} S1 OS dSNV *(SdLI9S-¢¢6 WAL) SBXa] ‘9ABD UYeYyUaSIIIy IY) WI, wnsas wNiayjoMmoy jo puke ‘(SQLT WSNA) ‘ds smpimusosdy as] yo atqes JYyslI pue (EQSszZ WSNA) Yuofpsnsd *] Jo aiqes yay [eyed ayy Jo ‘suaudads 1ayj0 pue (QORT LM) UOJOd [e1IDRy dy) WOIy TUOJSUYOr , SnprUsOAKYIST,, JO ‘NAdIg ‘BAL[V], puw ‘eUAOJTED ‘vag ey] oysueY WoIy UOpoznUs yo ‘e1uBA[ASUUdg ‘SUIZARD ApauUdy 310g WOIJ suautisads J9y}0 pue I1aI4aU (*§) “Py JO ad AjopOY 9y) puke 81719013 (UOPO]IUG) SnpOsaDY IV YO adAjOaU JY) JO BJVULA}SeIP puke SUOIHUApP JUIUBUTIAEd JaddN ay) JO sjUsWIaINSBI «¢ BGR (LZ) €'07-6'97 € 7 pUd [BISIP JO JO}OWTeIP ISIOASUPI], (LZ) 871-691 CE yeys Jo Ja}0uIeIp IejuL|doss0q (87) 6'PI-7'81 rel eYS JO JO}OWeIP asIaAsueI], Yel IW II] Tesreqeaj (07) Z'07-S'b7Z ha 6 06! 7 07 pUd [PISIP JO JO}OWIeIP ISIOASULI (SZ) €'ZI-L'91 = CZ - yyeys Jo Jaj}oureIp Iejueldosiog (SZ) 6 €1-0'07Z = Lean = YeYs JO JaJaWeIP asoAsueL] (pZ) 8°77Z-T7 67 = O'bZ = pue jeuxoid yo yidap mjueldosiog 12] Yel 12] Al ll Il (panunuos) etjedieoe ayy ee es ee a azis ojdues = (N) L8L11 HNWV Z6L11 HNWY 98L11 HNN niag ‘eieyey, pue SeSUBYIY suOISUdWIG “PIWIO}I[eD “eoalg eT] OyouRY “ainssiy preuod) sesueyly ‘oinssiy preuod) sesueyIYy “olnssiy pieuo) pue WOJj PoALIap Uopojlus uopopius 1papuo) sisdojuopo]nus sajkpo]soay sisdojuopojmus sjuswia[q JO} Sdduevl 9ZIS (penunuos) Z aqey ATB8E—Alep'¢ (Z) AlepS—ATeg "9 QQ IEG S| (pb) ML'EI-8°L1 (€) ME*LI-€ IZ (p) 0'OI-8'€l juasqe (8) O'OI-L ‘TI (9) 0'7E-9'8E (Z) 78-16 (p) ATeg'S—9¢°9 (€) ATS 6-7 61 (p) S€I-v'SI (p) O'87-8 'ZE (S) S19-S°LL (S$) $°8S-L'9L (S) 9 11-6'F1 (S) 9'0€-0'8E (€) L°L7-S$'SE (9) L'LI-1 SZ (61) 8°6-6 TI ID) OSS (0 (S) 8°91-7'0@ (91) L:OI-*7'€l (6) 0'0I-7 TI (9) MUTZI-MS ‘LI (L) 76-8 TI (p) M9°L-6'6 (Z) O'8h-P7 8h (9) 0'8-0°S7Z (S) 0°98-r'801 (bp) 6 L-9'IT 98 Oe! ATRL ATe3 CT S8P99 dWon (L) gy*1'€-0°9 (6),,*€ O1-9'S1 (8) gxL'b-Z'L (LZ) gxS TI-9'LI (9€) €91-9'I7 (O€) 9'07-€ LZ (LE) ZLI-9'bZ (gp) ,x1¥asqe—al ‘rI (LZ) V'1I-8'81 (97) , xh €€-€ 9b (97) 9x7 8-9 TI (€Z) 0'8-9'01 (ZZ) _*0 C1-6'07 (81) €'LI-6°7Z (61) L’S€-6'8P (pI) €8ZI-L'IST (SI) 90'€ZI-F IPI (PI) €°SI-Ip7Z (91) L’€€-9b' br (81) 0'9€-O'IS CO) SIS 1 (OS) 8°€1-90°61 (Z€) 6'6-T El (Or) p'9I1-1°2Z (pS) 101-9’ (O€) €°L-L'6 (S€) 9° 11-9€'81 (8p) 8°8-L ‘ZI (67), #0°S-9'L (Z€) 98°8h-6'S9 (rE), 48°L-9°EZ (p€), x0'L6-0'I€1 (0S) 1*6°9-L'LI Ite o'8 61 EXCL Ll 691 Cael 19 Lel te cel vy 0E O'LT Lp dSNV Ly dSNV c°8 Ee BSI OS dSNV MOE ISI ac el °6'L6 20'8 26 OE M6'8 8 6b 26°71 90°L6 20°8 1YsI9y UMOID Ja}oWeIp [enduljooong JQ]OUWPIP [RISIPOISAW AW ape[g quooR}OW Jo yWsUTT QuOdR}OUW Je YSIOY UMOID auooried ye }YysIaYy UMOID ajnuooesed ye yysIay UMOID ajAjsesed ye jYysIoy UMOID Jo}OWIeIP Jenduljoosong JOJOUIEIP [LISIPOISAW 4d auosojoid ye yYysIay UMOID JO}aWIeIp jenduljooong JOJIWLIP [PISIPOISA, ed JOO1 JO Ja}OWeIp [eNsuljooong wWNwWIxeyW JOOI JO JOJOWLIP [PISIPOISOW WINWIXe| }YSIaY UMOIO [PISIG }YSIOY UMOIO [RISA UMOIO JO JO}OWIeIP [eNsul[ooong UMOIO JO IOJOWIP [eISIPOISI|] SN[OATL JO IOJOWIP [PISIPOISA ,O WYysIay UMOID JO}OWeIP [enduljooong JOJOWEIP [BISIPOISA ¢] 1YsIay UMOID JO}OWeIP ;ensurljooong JOJOWEIP [BISIPOISA| z] IYsIay UMOID JO}OWeIpP [ensurljooong JOJIWLIP [BISIPOISA| {| rd-ed Wsur] BUID|SeIP duIURd}sod jo ysUIT rd-1D ysuey] BUId}SeIP duIuRoaid jo SueT] fable 4 Measurements of mandibles and lower permanent dentitions of Machaerodus (Smilodon) gracilis and M. (S.) mercerii from Port Kennedy Caverns, Pennsylvania, of [schyrosmilus ischyrus, I. idahoensis, I. johnstoni and I. crusafonti from the western States, of Smilodon gracilis from Florida and of Smilodon from Rancho La Brea, California, and Talara, Peru. Measurements were taken by the author except those from Merriam (1918) indicated by ‘‘1’’, Merriam (1905) indicated by **2°°, from Schultz and Martin (1970) indicated by **3’’, or by Dr Bjorn Kurtén (1982, pers. comm.) indicated by ‘‘4’’. Measurements of the teeth of the holotype of /. idahoensis are taken from roots and damaged crown. Measurements are in millimetres. “‘e’’ indicates an estimation; ‘‘w’’ indicates a worn crown on a tooth; ‘‘d’’ direct measurement across both dentaries; ‘‘alv’> measurement based on alveolus; ‘‘ +’ indicates a minimal measurement; and ‘‘—’’ that the dimension is not available. Numbered asterisks in right-hand column indicate overlapping size ranges with Smilodon from Rancho La Brea and Talara, and are referred to in the text. Elements Machaerodus (Smilodon) specimens Ischyrosmilus Ischyrosmilus and Port Kennedy Caverns, Pennsylvania ischyrus idahoensis Dimensions M. (S.) gracilis, ANSP 44 (neotype), 45 and 47 Merriam, 1905 Merriam, 1918 M. (S.) mercerii, ANSP 49 and 50 (neotype) Holotype ucmMP 8140 Holotype ucmp 22343 ANSP 44 ANSP 44 aANSP49 ANSP 49 ANSP 45, 47, 5O Cast AMNH 32797 left right left right Length Ii-Is 15.6e 15.0e 16.le 16.6e - dose Length I1-C1 - 22.6e 22.0e 23.9e - 36.8e Length Ih-Mi 129.1 MIZESe 123.0! (155.0e) Length I2-M1 126.7 112.2e 116.7 110.1 - 152.5e Length I2—Ps - - 96.1 99.2 - 120.9e Length Ci-Ps 91.1 - - - - (107. le) Length Ci—Mi 115.9 - - - 107.0! (3522) Length of postcanine diastema 58.0 - 48.2 48.7 3352 39E" 46.41, 46.3 Length Ps—Mi 59.9 70.0e - - 61.7, 62.4e 726; 73.265 Length of diastema post Ps 8.5e 11.2e 3.6 3.9e - 4.2 Length Pa—Mi 45.8 48.3e 42.8 42.4 49.8e3 57.3e, 58.53 I: Mesiodistal diameter 3.9 353 - - 4.0! - Buccolingual diameter Ss) oo) - - - = Crown height 6.0w 7.8w - - - - Iz Mesiodistal diameter Sai) 533 4.9 5.2 6.5! 8.5e Buccolingual diameter 6.9 7.0 5.1 6.7 - VAI Crown height 9.0 8.7w 8.5w 8.2w - = Is Mesiodistal diameter 6.5 6.3 6.1 6.2 - 11.Se Buccolingual diameter 8.9 8.3 8.2 8.2 - 14.0e Crown height 11.5w 11.9w 11.4w 12.2w 5 = Ci Greater crown diameter 10.6 - 11.5 NQES [4eSea ise 19.21, 18.3 Lesser crown diameter 7.6 - 7.4 8.0 152 W255 WEES Buccal crown height 14.8e - 17.2 16.9 = : P3 Mesiodistal diameter Us} = 9.Salv - TOL Has 1025 10%5e2 Buccolingual diameter 5.0 - 5.0alv - 6.53 5:9%,'6:5 Crown height 4.9 - - - = C ANSP 50 Ps Mesiodistal diameter 20.7 20.5 20.0 20.0 20.1 20:14, 20:5° PBX), HS Buccolingual diameter 8.7 8.7 8.8 8.9 8.9 12252 10.4, 10.5% Crown height 13.8 12.8w 13.2 12.8 1355) - - Ischyrosmilus Johnstoni Mawby, 1965 Holotype wr 1239 Ischyrosmilus crusafonti Schultz and Martin, 1970 Holotype UNSM 25493 Smilodon gracilis Inglis IA, Florida Cast UF 20065 Smilodon gracilis Bass Point Waterway, Florida ROM:B numbers Size ranges for Smilodon derived from Rancho La Brea, California, and Talara, Peru left right (N) = sample size = 25e 23.5—-15.0e*?! (34) - 33.5e 35.7-25.0e (34) - 142e 158.8e-124.6*? (26) E 157.6—-123.3e*? (33) - 112.0 §2+ 122.8-95.8 (11). - 128.5 95+ 148.7-113.5**4 (36) 36.5alv 37.0alv 26.93 32.5e 80.2—43.5*5 (46) 69.8 71.Salv 59.5e3, 60.04 52e 65.5 (1) 6.4 8.5 3.9 - 48.9 48.5 43.83 38.7 60.9-48.3 (46) - 5.5e - 5.6-4.0 (29) - 6e - 9.7-7.0 (30) = - - 13.5e-10.5 (29) ROM:B 4249 - 6e - 6.0 7.8—5.5*® (40) - 8e = Uo 10.3-8.3 (40) - - - 9.2 16.8e-10.1 (34) A t i 9.6-7.2 (48) i : < 12.9-10.1 (47) : = 2 21.4e-13.5 (45) ROM:B 4236 - 16.Salv 1B e771 4204 11.1 17.6-12.1*7(19) - 10e 10.03, 11.54 - 7.9 13.8-8.5 (19) - = 16.8 24.8-17.9 (15) 13.0 14alv 10225954 9alv 10.2-6.5*8(25) 8.0 6alv 63 6alv 7.1-5.7 (25) lle - = - 8.3-5.8 (25) ROM:B 4229 22.0 21.8 18.73, 19.54 18.6 - 30.9-22.5 (S50) 10.3 9.9 9.63, 10.04 8.1 8.8 14.6-10.5 (52) = 12.6w - 22.2-16.5 (42) 49 Table 4 (continued) a Elements and Dimensions Machaerodus (Smilodon) specimens Port Kennedy Caverns, Pennsylvania M. (S.) gracilis, ANSP 44 (neotype), 45 and 47 M. (S.) mercerii, ANSP 49 and 50 (neotype) Ischyrosmilus ischyrus Merriam, 1905 Holotype ucmp 8140 Ischyrosmilus idahoensis Merriam, 1918 Holotype ucmp 22343 ANSP 44 ANSP 44 ANSP 49 ANSP 49 ANSP 45, 47, 50 Cast AMNH 32797 left right left right ANSP 45 Mi: Mesiodistal diameter 24.6 24.8 22.8 2D 25.7 Xoo PADS 332 SoN Buccolingual diameter 11.4 10.8 10.3 10.5 11.6 ISOM WLS. IGS IBS, 1B Crown height paraconid 14.lw 12.7w 12.3w 13.3w 14.6 - - Crown height protoconid 13.8w 14.9w 11.8w 12.2w 14.9 - . Greater diameter of symphysis - - 55.8 50.0 - 70.0 Lesser diameter of symphysis - - IS}5 7/ 16.5 29.5 ANSP 45 Breadth across canine alveoli - - 38.2e 36.4e 38.2ed - 76.2 Breadth across genial flanges - - 39.0e 36.7 34.1d 52.01 Sy20- Minimum depth in diastema 3355 34.4 323 30.1 - - Minimum width in diastema 12.9 9.6 WES 1291 - = Depth beneath P3 29.8 3357 34.0 30.6e 41.4! 52:05 Depth mesial to P4 B75 32.8e 32.5 32.4 41.21, 39.53 49.01, 48.8 Width mesial to Pa 15.6 12.2e 14.8 14.7 - DDD ANSP 47 left dentary Depth at P4—Mi: contact 32.6 29.3e 32.8 33.2 Dee 37-535, 36102 = Width Psa—M: contact 16.4 12.7e 15.5 15.8 15.3e - DAES Length from Mi to coronoid - - - 38.2 - 21.5 Length from I: to coronoid - - - 146.4 - - Height from coronoid to angle - - - 50.0e - - 2 EEE Ischyrosmilus Ischyrosmilus Smilodon gracilis Smilodon gracilis Size ranges for Smilodon johnstoni crusafonti derived from Mawby, 1965 Schultz and Martin, 1970 Inglis 1A, Bass Point Waterway, Rancho La Brea, California, Holotype wt 1239 Holotype UNSM 25493 Florida Florida and Talara, Peru Cast UF 20065 ROM:B numbers left right (N) = sample size SS... SS. eee ee ROM:B 4260 28.5w 28.3w 25.83, 25.94 20.8 23.0alv 32.1-24.5*® (55) 13.8 2.7) 12253 12-44 10.5 10.0alv 17.0-11.8 (57) . - - 14.5w - 22.2-17.6 (50) - - = 15.0w - 24.9-16.0 (36) 74e - 46e - 78.5—48.7* 1° (48) 19.5 - 14.6 - 36.6—20.9 (50) 64e - 32e - 64.6-47.8 (44) 6Se - 44e - 63.0—40.2 (50) 47.8 46.5 - 28e - 40.4-27.3*1! (52) 20.0 18.8 - 3} - 21.8-13.8 (52) 42.2 44.5 - 27Te - 41.8-33.3*!? (26) 43.3 46.5 32.3 25e - 49.9-32.6*!3 (51) 20.0 20.6 - 13.0 - 25.2-16.5 (51) 44.7 46.0 - 28.7e - 54.0-33.7 (52) 19.8 20.6 - 14.7 - 25.8-18.8 (52) - 38.5 . - - - : Z = = - 209.9-151.0 (30) 83 79.0 - - 77.9-57.1 (42) (€Z) bp Or-b' PS BPE ZT PE WIN[NDeJUIISNS JOAO YIPIM UWINUITXe|Y (IZ) S°t€-T bb 8'bT 6'€7Z JOB} PlOgNd ssolOe YIPIM WINWIxeY] (GO) EGEAG Az L'0€ OIE JOR} IL[ESeINSE ssO1Oe YIPIM WINWIXE| (0%) +'6L-8'90I - Z'€8 yisuay wnuwixeyy 1y3u 1Y3u wnouesjea (SI) 161-6 IZ (11) § 61-6 '7Z = PSI > = pus [BISIP JO JNJQWeIP ISIOASURIT, (IZ) 801-191 (0%) 7'7ZI-9 91 5 901 = 101 YeYspIu UT Id}OWeIP JOLIa}sodo1ajUy (IZ) 6'7I-6'LI (61) €7I-€ 91 zZ c'6 : vO YeEYSPIW Ul IO}OWeIP asIOASUeL], (IZ) €'€7-T ZE (61) 9'07-L 87 C9 Ll 6'SI = J}IWeIP [eUSAOSIOP [eWIXOId WNWIXe| (IZ) 107-887 (61) 681-997 CSI CsI OE! 891 JoAWeIP asJaasuey) [PWIXOId WNWIxeEy (1Z) 9° 19-9'98 (LI) p'6L-v LOI : 109 = e yisug] UINWIXe isu YP] wysu Ye] A AI A A AI Al A pure Aq eypedivsejayy (07) Z'07-S FZ (LZ) 6 €I-1 EZ Psi 2 = - pud [eISIp JO Ja]QWeIP asIOAsUPIL, (COMTI aAON (LZ) 6 €I-8'LI 601 C6 = - YeYSpIU UT JAJIWeIP JOLIa)sodo1jjUy ($7) 6 €1-0'07 (LZ) O'SI-8'81 Cll 901 = YEYSPIW Ul JOJOWeIP ISIOASUBI], (pZ) 8°77-T 67 (pZ) p'97-O'PE 6'SI 5 8°61 0'€7 JOOWIP [eNUAOSIOP [RWIXOId WNWIXeEYy (97) I'p7-E'1€ (v7) Z8I-I v7 P's - 8°07 OLI JOJOWIP ISIOASUB) [PWIXOId WNUWIxe|| (SZ) 0'€8-9'601 (SZ) 9L-0'SOI ULL = ; 2 yisusy WnuWIxeyy 12] YI 1ysu 1ysu Ill ll Ill II Il ll Il pue yy eyedieoniayy (97) O'LE-O'rr GEGG Ja\9WeIp [euxoid wnuUIY, (9%) 7 Ip-s's¢ 9'PE Jo}oWeIp [eWIxold wINWTxe| Yel snipey (v2) O'O€-b' Er L'0@ i £4 yeys JO JOWeIP WNWIUT (OI) 7 8€-0'19 0'S7Z cle Saploy[ap Jaqn) JOAO JeYs JO JaIWeIP WNUIXeE| (SZ) 7 S7-0'9€ 8 17 607 SUONE|NOUe JUN Jo Ja}OWeIP UMUTTUTA (PI) 2p pS-P' PL 0°99 LL9 JQ IWeIP IYyooH [eysip wnwixeyy 143 ne) smouiny] azis ajduies = (N) IS dSNV 8b dSNV Nidg “eie[e], pue ‘eruoyIyED eruea[Asuudg erueaA[Asuudg SUOISUDWIG ‘valg eT] OyouRY Woy poatiap ‘suiaaeg Apouusy Wog ‘suiaaeg Apouusy og pue UOPO]IMG 10} SABueI IZ1S MdaI4IU (°§) SNPOLADYIDW sijlova8 (°§) Snposavyovw Suda] q “Q[QVINSBOWUUN SI UOISUSWIP SY) JY) SOIBOIPUL ‘ Pr -,, ‘SONOUNT[IW Ul Ie s}UdAINSBaYY “JOYINE oy) Aq s}USWOINSeOUW WOIJ ‘niog “eleye yl, WoOIJ pue (Z7EG[) YOOIS puve WILD] WO PSALap oe “PIUIOJI[RD ‘Bolg BT OYOURY WO UVOPO]NUG IO} Bed ‘UOpo]IMS pue ‘i4adIaU (*S) “Py ‘S1pP1IDAT (UOPO]IUG) SNPOLADY IVP JO SJUIWI]a [BLURII}SOd JO SJUIWIAINSBIPA, = AGey $6 ST HWOU LETp A WOU L6 8 tl 10 c'8 Yel 9 OI SETp BNO Tl 0'€7 1ysu 87Tb FOU 6 6€ LSI 701 1°6 wysu CET 1: WOU BWO\seIp duluRd\sod jo ysuey Jajaweip [enduljooong JOJQWPIP [BISIPOISA 1) yisua] JOOI-UMOID 1YsIay UMOID JoJOWeIp [ensuljooong Jd}9WeIP [PISIPOISa|y el uonnuag 19ddq Nq uRsgof [A BL[IXBW Y2] “OOP8 #WOx SI}1IDAB. SNJWUSOAKYIS] | AemiayeAy JUIOg sseg sioquinu €:WOu SIPIIDAS. SN]USOAKYIS] sjeliajeyy jejueg ‘y Leg SJUIWIOINSBOU PIJBUUT}S9 UB SOJBOIPUI . ‘ 5). “Q[QRIIBACUN SI JUSWAINSBIW JY) JEU) -,, Puke ‘YyJ00) B UO UMOIO UIOM B ee MA SSNJOIATL JO JUIWIINSPIW S2}BOIPUl SSONOUT[[IW UT JIB s}UDWOINSeIP] *(668) 2doD wo se gp dSNV JO} SsasayjUoied Ul S]UdDINSeII] °(QL6[) UOS}1IQOY Wor ‘epliopy ‘AjuNOD enysery ‘WAX Ay eIo'T apley wos pue ‘eruealAsuuag ‘Apauusy j10g Woody esesjse ay) JO pue seplopy ‘AjUNOD 9}0[1eYD ‘iq UBIgOL [q pue ‘AjUNOD BjOSeARS ‘]T AEMIIIVAA JULOG SSVY WIOAJ SI1IDIT SNJNUSOAKYIST JO S[RLAIJOUL PIAIAOIIA A[JUIIIA JO SJUIWIANSeITA] q pue y syieg 9 qe L (97%) €°SI-$ 07 (%€) 901-6 91 (1€) 1'OI-€'vl (SE) GLIS9I8C (97) 8°81-O'LZ (Z€) 8'OL-8'b6 A (O¢) (1€) (€€) (T€) (ZZ) 891-802 (O€) €'ZI-8'SI (1€) S' 11-091 CONG SCSVCE (87) S€1-0'61 (LZ) L’€L-$ ‘96 L'9C-0'8t O'1C-$'67 8: SP-T 19 0°9¢-c 19 ISz O17 Z'8P 0'8b 1ysu Pud [SIP JO JOJQWeIP ISIOASURI YRYSPIL Ul JOJOWIP [PIUZAOSIOG YEYSPIW Ul JQJQWeIP aIsJOAsueI JQIWeIP [PIIUZAOSIOP [RWIXOId WINWIXRYy JQOWeIP ISIOASUR [PWIXOId WINWIXeyy yisua] wnwixeyy A pur “[]] ‘II eiesreieiayy peoy jo Ja}QWeIP WNWIXe|y yoou JO Ja}awWeIp WNW yisue] wnwixeyy yIpPIM WNWIXeyy snjesensy ATPO'ET JOIWIP [PISIPOISI|] su WW 60th WOU ‘ }YSIOy UMOID 8°8 Ja\awWeIp [ensuljooong a JQIWIP [BISIPOISI 1ysu Fd 6CTp @WOU 28P bp 8P yysug] JO01-UMOID = 891 doejins yeoong uo WYsIaYy UMOID C8 6L JQ9WeIP UMOIO ([eNsuI[OIONg{) Jessa] TIT aaa JOJOWLIP UMOID ([BISIPOISAW, ) 19}B2IH Yel el fe) OSCh aWOu 9th FWOU 9°0€ yISUa| JOOI-UMOID 86 1YySlay UMOID OL JaaWeIp [ensduljooong 09 JO}OWIP [BISIPOISI| Ye] cy 6bCh FWOU uonnuUEg 19M0T cal O'rl ape[q euodR}OW JO YysUeT MLS THI auOdeIOW JO JYSsIay UMOID MO'TT 161 auooevsed jo yysIay UMOID Mo'6 8 El gjnuooesed jo yy3iay UMOID MLS juasqe ajAjsered Jo 1Yysloy UMOID Lcl +611 Ja}aWeIP fensulpooong S'0€ THE JQJIWLIP [LISIPOIsa|y su rd O€TP s:WOU Mo'9 16 00! WYslay UMOID VL vL OL ajaweip jensurjooong Ll Ol O'LI JQJIWLIP [BISIPOISa 1ysu YI ed epeTp GINO S@Tb WOU (panuyuos) uo ued Jaddy Ud ueogof [7 ] AemiajeA JUlog sseg BI[IXeW Yo] “OOPS aWOU SI1LIDAS SNJTUSOAKYIS] sioquinu €:WOu SIIIDAS SN] NUSOAKYIST sjeliayey jequag ‘y wed EE en nn nn nn Te (panunuos) 9 aqey a LOl uone[noiue [PISIP JO JOJQWIP ISIOASUPIL 06 Pud [BISIP JO JO}OWPIP [eIJUBAOSIOG Pl : pud [PSIP JO JOJOWIPIP ISIOASUPI 0'8 YPYSPIUW Ul JBJIWPIP [PIUBSAOSIOG vil YeYS JO JOJOWIeIP ISIOASUPIL GEhI pua [ewixoid jo 1a]aWeIP [eUZAORSIOG Isl pues jewlxoid jo 10}QWeIP ISIOASURIL ILe yisury] 1ysuy, xue[eyd [BwIxOld OOP a:Wou sel = i uonenoime [easuejeyd jo 1a}awWeIp asiaAsuel] 8 rl = - pud [eISIP JO JO}QWeIP [eIJUZAOSIOG 6SI 9°81 € Sl PpUd [PISIP JO JOJOUIEIP ISIOASUBI] 8 TI OCI IG)! YeYSPIW Ul JOJOWeIP [eIJUSAOSIOG S01 8 Cl pl YeEYSPIW Ul JOJOWIP IsIOASURI] (GCG €°67 pS7 pu yewixoid jo 1a}9WeIp [e1]UZAOSIOG (S/E| GGG 8°07 pua [ewixoid jo 1a}9WeIP asJOAsUeI] 6°76 a : yisur] Yel 1ysu Yel AI pue II] B![esreqeloyy Al Ill lil ZETH LWOU PLS9 a:Wou €€Tp LWOU (ZZ) 0°97 LIZ Jea[YOON JO UONBAg|[a [BUIAIXY CIC Li pesy fo Ja}aweIp WINWIUI COMES? $97 9°97 peoy jo Ja}OWeIp WNWIxe| (LI) O'1Z r'6l - yoou JO Ja}aWeIP WNWIxey (LZ) €°97 L'8Z TeI[YIOI JO JOJOWeIP asJOAsues | (8€) O'8Pr bv yy a Sb JQJOWIP IsIOASUBI] WNWIXeYy (8b) 78h vty IP ysusy WNUWIxe|] Yel 8ecp AWOU snjesensy e1uva[Asuuag ‘sulaaeg Apauusy od J Aemiaqe A JUlOg sseg VAX 2leH snjeselse ys snyeseyse iysu ‘Sp dSNV ‘SIJ1oD4I3 ssoquinu €:WOu ‘O6PLI An (‘S) snposapyoow SI1Iv4d SN]WUSOAKYIST syjl9v4s8 UOpopIMs sJUdWIa]q [BlUeID}SOg *g 118g DEE - piuosojoid jo yYysIay UMOaID - piuooesed Jo yYysIay UMOID AleO 01 Ja}QWeIp yensulpooong a /7:°00€_O§[(— _ (0%) €'€€-S Ih (€) 0'O€-0'8E 0-9€ siskydida [e\sip Jo Ja}aWeIpP Jo1Iasodo1g}Uy (pl) O'Sh-€'€9 (€) 1'OS-9°SS O'7S sisXydido |eysip Jo JaJQWIeIP asIOASUeI], (91) 1°S7-O'€E (€) 7 97-0'87Z COE YeEYSPIW Je JO}OUeIP asIOASUBI], (pl) 7SL-¥'06 (Z) 9 IL‘90°9L L’8L Jajqweip jewrxoid asiaasuesy (v1) 0°6€7—-0'SOE (€) 0'$97-0'L67Z OSE yisua] wnuixeyy 68799 dWon PIqLL 00! yojou Je[Apuoosajqul JO Ja}OURIP aSJOASURI) WINUTIUT] (61) S'‘pI-€°€z ($) O'€I-S'SI OSI yojou se[Apuoosajqul JO Ja}9WeIpP asJoAsueI) WNWIXeE| (81) 6'PZ-0°9€ (S) O1Z-S°ZE 687 a[Apuod [eIpow JO INJOWIP asIOAsUeI] 8'7Z a[Apuod [esa}e] JO JOJIWIP as1dASURI], (81) 6 Ip-0'09 pre J0BJINS (Je[N}OI) Iepjayed Jo IDOWwIP asIaAsuel] (67) 6°€9-€'08 (S) O'L9-O'IL CGL sisXydida |e sip Jo 1a}0WeIP JOLIa}sodolmUYy (61) 7 S9-7'06 (S) pL9-S'E€L 6 EL sisXydida [e}sip Jo 1a}QWIeIP asJOAsueI], (07%) 8°97-F'SE (QI GEGeSi8C P°6C P67 YeYspIu! Je IQJAWeIP 10LIA}sodo1g)UY (0%) 9°6@-¥ Or (S$) O'O€-L' TE 91 °9€ LLE YEYSPIW Je JO}OWeIP asIoAsuRL] (07%) 1°6€-L'0S (S) O'Ip-8 Ur Gay 6 Cr suoway jnded jo aJaWeIp IOLIa}sodo1ajuy (81) $'6L-8'80I (S$) 1°68-L°S6 266 Jaqurysoy 19)R0I3 pure jndeo JaA0 JajaweIp yeulxoid WnUIxepy 260b a[Apuod [eipouw 0} soWay yndeos wo yWuUeT (L1) O'LI€-0'80r (S) O'8EE-O' ESE Sa[Apuod [k}SIP 0} IAJURYSON 1A}B9I3 I9AO0 YBUYT 88799 dWwon 9901 LA nwa O'CE yojou [erpes jo yur] BEE yojou IeunyIWias Jo yur] (tZ) €°LS-8'8L P8s $sa001d ploUOIOS J9A0 JaIWeIP JOLIa}sodo1gjUy (%) OSP' TSP Lae Yd}]OU JeUN[IWs JO I[PpIw UI Ja}QWeIP 1OLIa}sodo1gUYy (LI) 9'€€-0'8r € OE UOURIDIIO JO JOJIWIP ISIOASURT) LINWIXey 879 LM euln (SZ) €°S$7-0'9E (€) $°67-L‘0E ELC OST O'L7 IeITYOOI JO JI}9WeIP JOMA}sodo1ajue WNUTUTAY (ZZ) L€6-€'87I (€) $'78-¥'P8 LIL v'OL sisXydida Jeis1p Jo 1a}9WeIP asJoasueL], (pZ) 7 8€-O0'19 (bp) -°6€-0'6P 6'SE YeyYspIuU ye JOJ}AWeIP JOLIAsOdolaUYy (PZ) O'O€-F EF (p) S'87-O'EE SSG YeYSPIW Je JDJOWIeIP asIaAsueI], (SI) 0'60€-0'S8E (€) O'OrE-O'8SE O'6IE g[Apuoo [eipaw 0} yndeo wos ysueq 1691 LAM L8>99 dwon S79 LM sngoumny azis ajdures = (N) azis ajdwes = (N) nid “Ble[ey, pue “eIWoj[ed SPXd] ‘9ARD UYyeYyUasoLy sexay ‘uoXkued eid sUOISUaWIC] ‘eolg eT] OyoUeY WOIj poauap Winds WN1AIYJOWOH pue UOPO]IMGS OJ SIBuUeI IZI¢ JO} Saduel 9Z1¢$ tuojsuyol snjmusoscyos] sjuawiayq ‘JuoWoInseoW ayewIXosdde ue sayeoipul , ,d,, ‘Joy Ne oy) Aq Udye) sS}UIWIINseIW WIJ pue (ZEH]) YOOIS puv WILD, WO PIALAp d1OM s]USWTOINSRaT\y ‘Nidg ‘BBTV pue ‘eusOseD ‘eaig ey oyouey ‘sexay ‘aaeDd uYyeyUasoliy WOAl ([96] ‘apBayA) Wn4asS UNAaYJOWOP] JO sasuBs IZIs puB (S9GT) Aquey, Aq 1UOJsUYOr SnjNUSOsKYIST OF PIUBISSE SJUIWI]a [eIURID}SOd JO sSjudUIAINSeAyY = L,: BIGE] Literature Cited ADAMS, G. I. 1896 The extinct Felidae of North America. American Journal of Science, Ser. 4, 1:419-444. BARBOUR, E. H. and H. J. COOK 1915 A new saber-toothed cat from Nebraska. Nebraska Geological Survey 4:235-239. BERTA, A. and H. GALIANO 1983 Megantereon hesperus from the Late Hemphillian of Florida with remarks on the phylogenetic relationships of machairodonts (Mammalia, Felidae, Machairodon- tinae). Journal of Paleontology 57:892-899. BIELAWSKI, J. B. 1905 Gisement et faune de Rocca Negra a Perrier. Revue d’Auvergne 1905:297-304. BOVARD, J. F. 1907 Notes on Quaternary Felidae from California. Univer- sity of California Publications, Bulletin of the Depart- ment of Geology 5:155-170. BROWN, B. 1908 The Conard Fissure, a Pleistocene bone deposit in northern Arkansas: with descriptions of two new genera and twenty new species of mammals. Memoirs of the American Museum of Natural History 9:158-208. CHURCHER, C. S. 1966 The affinities of Dinobastis serus Cope 1893. Quater- naria 8:263-275. GCOPES ESD: 1880 On the extinct cats of North America. American Naturalist 14:833-858. 1893 A new Pleistocene saber-tooth. American Naturalist 27:876-879. 1895 The Fossil vertebrata from the fissure at Port Kennedy, Pa. Proceedings of the Academy of Natural Sciences of Philadelphia for 1895:446-451. 1899 Vertebrate remains from Port Kennedy bone deposit. Journal of the Acaderny of Natural Sciences of Philadelphia, Ser. 2, 11:193-267. CROIZET, J.-B. and A.-C.-G. JOBERT 1828 Recherches sur les ossements fossiles du département du Puy-de-Dome. Paris, (n.p.). 224 pp. DE BONIS, L. 1976 Un félidé a longues canines de la colline de Perrier (Puy-de-Dome): ses rapports avec les félines machairodontes. Annales de Paléontologie, Vertébrés 62:159-198. EVERNDEN, J. F., D. E. SAVAGE, G. H. CURTIS, and G. T. JANE 1964 Potassium-argon dates and the Cenozoic mammalian chronology of North America. American Journal of Science 262:145-198. FABRINI, E. 1890 I Machairodus (Meganthereon) del Valdarno superiore. Bollettina’ Comitato Geologico d'Italia 21:121-144, 161-177. GERVAIS, P. 1878 Sur la dentition des smilodons. Comptes Rendus Hebdomadaires des Séances de |’Académie des Sci- ences de Paris 87:582-583. GILL, I: 1872 Arrangement of the families of mammals and synopti- cal tables of characters of the subdivisions of mammals. Smithsonian Miscellaneous Collections 1121-98. GRAY, J. E. 1821 On the natural arrangement of vertebrose animals. London Medical Repository 15:296-310. GREGORY, J. T. 1942 Pliocene vertebrates from Big Spring Canyon, South Dakota. University of California Publications, Bulletin of the Department of Geological Sciences 26:307-446. HARRISON, J. A. 1983 The Carnivora of the Edson Local Fauna (Late Hemphillian), Kansas. Smithsonian Contributions to Paleobiology, No. 54: 1-42. HAY, O. P. 1920 Description of Pleistocene vertebrates found in the United States. Proceedings of the United States National Museum 58:83-146. 1927 The Pleistocene of the Western Region of North America and its vertebrated animals. Carnegie Institu- tion of Washington, Publication 322:1-346. HIBBARD, C. W. 1952 Vertebrate fossils from late Cenozoic deposits of central Kansas. University of Kansas Paleontological Contributions, Vertebrata, Article 2:1-14. 1955 Notes on the microtine rodents from the Port Kennedy cave deposit. Proceedings of the Academy of Natural Sciences of Philadelphia 107:87-97. JOHNSTON, C. S. and D. E. SAVAGE 1955 A survey of various late Cenozoic vertebrate faunas from the Panhandle of Texas. Part 1, Introduction, description of localities, preliminary faunal lists. University of California Publications in Geological Sciences 31:27-49. KAUP, J. J. 1833 Description d’ossements fossiles de mammiferes in- connus jusqu’a présent qui se trouvent au Muséum grand-ducal de Darmstadt. Second cahier: carnassiers fossiles. Darmstadt, Heyer. 31 pp. KLEIN, J. 1971 The ferungulates of the Inglis Ia local fauna, early Pleistocene of Florida. M.Sc. Thesis, University of Florida. 115 pp. KITTS, D. B. 1957 A Pliocene vertebrate fauna from Ellis County, Oklahoma. Oklahoma Geological Survey, Circular a5 27) KURTEN, B. 1962 The sabre-toothed cat Megantereon from the Pleis- tocene of Java. Zoologische Medelingen 38:101—104. 57 1963 Notes on some Pleistocene mammal migrations from the Palaearctic to the Nearctic. Eiszeitlater und Gegenwart 14:96-103. 1965 The Pleistocene Felidae of Florida. Bulletin of the Florida State Museum, Biological Sciences 9:215-273. KURTEN, B. and E. ANDERSON 1980 Pleistocene mammals of North America. New York, Columbia University Press. 442 pp. KURTEN, B. and M. CRUSAFONT PAIRO 1977 Villafranchian Carnivores (Mammalia) from La Puebla de Valverde (Teruel, Spain). Commentationes Biologicae 85:1-39. LEIDY, J. 1868 Notice of some vertebrate remains from Hardin County, Texas. Proceedings of the Academy of Natural Sciences of Philadelphia for 1868:174-176. 1889 The sabre-tooth tiger of Florida. Proceedings of the Academy of Natural Sciences of Philadelphia for 1889:29-31. LUND, P. W. 1842 Blik paa Brasiliens Dyreverden for sidste Jordom- vaeltning. Danske Videnskabernes Selskabs Skrifter, Naturvidenskabelig og Mathematisk Afhandling 9:137-208. LUNDELIUS, E. L., JR., C. S. CHURCHER, T. DOWNS, C. R. HARINGTON, E. LINDSAY, G. SCHULTZ, H. A. SEMKEN, S. D. WEBB and R. J. ZAKRZEWSKI In press The North American Plio-Pleistocene sequence. Ber- keley, University of California Press. LYDEKKER, R. 1884 Indian Tertiary and Post-Tertiary Vertebrata. Siwalik and Narbada Carnivora. Palaeontologia Indica, Ser. 10, 2:178=-355. MARTIN, L. D. 1980 Functional morphology and the evolution of cats. Transactions of the Nebraska Academy of Sciences 7:141-154. MARTIN, L. D. and C. B. SCHULTZ 1975 Cenozoic mammals from the Central Great Plains. Part 5, Scimitar-toothed cats, Machairodus and Nim- ravides, from the Pliocene of Kansas and Nebraska. Bulletin of the University of Nebraska State Museum 10:55-63. MATTHEW, W. D. 1918 Contributions to the Snake Creek fauna; with notes upon the Pleistocene of Western Nebraska; American Museum Expedition of 1916. Bulletin of the American Museum of Natural History 38:183-229. MAWBY, J. E. 1965 Machairodonts from the late Cenozoic of the Panhan- dle of Texas. Journal of Mammalogy 46:573-587. MEADE, G. E. 1961 The saber-toothed cat, Dinobastis serus. Bulletin of the Texas Memorial Museum 2:23-60. MERRIAM, J. C. 1905 A new sabre-tooth from California. University of California Publications, Bulletin of the Department of Geology 4:171-175S. 1915 Extinct faunas of the Mohave Desert, their significance in the study of the origin and evolution of life in America. Popular Science Monthly 86:245-264. 1917 Relationships of Pliocene mammalian faunas from the Pacific Coast and Great Basin provinces of North America. University of California Publications, Bulle- tin of the Department of Geology 10:421-443. 1918 New mammals from the Idaho Formation. University of California Publications, Bulletin of the Department of Geology 10:523-530. 1919 Tertiary mammalian faunas of the Mohave Desert. University of California Publications, Bulletin of the Department of Geology 11:437a—437e, 438-585. MERRIAM, J. C. and C. STOCK 1932 The Felidae of Rancho La Brea. Carnegie Institution of Washington, Publication 422:1-231. OSBORN, J. F. 1918 Equidae of the Oligocene, Miocene and Pliocene of North America, iconographic type revision. Memoirs of the American Museum of Natural History, n.s. 2:1—330. PILGRIM, G. E. 1931 Catalogue of the Pontian Carnivora of Europe in the Department of Geology. London, British Museum of Natural History. 174 pp. PIVETEAU, J. 1961 Carnivora. Jn Piveteau, J., ed., Traité de paléon- tologie. Paris, Masson, t.6, vol. 1, pp. 641-820. RAWN-SCHATZINGER, V. M. and R. O. COLLINS 1981 Scimitar cats, Homotherium serum Cope from Gassa- way Fissure, Annon County, Tennessee, and the North American distribution of Homotherium. Journal of the Tennessee Academy of Science 56:15-19. ROBERTSON, J. S. 1976 Latest Pliocene mammals from Haile XVA, Alachua County, Florida. Bulletin of the Florida State Museum, Biological Sciences 20:1-18. SCHAUB, S. 1925 Ueber die Osteologie van Machaerodus cultridens Cuvier. Eclogae Geologicae Helvetiae 19:225-266. 1934 Observations critiques sur quelques Machairodontides. Eclogae Geologicae Helvetiae 27:99—406. SCHULTZ, C. B. and L. D. MARTIN 1970 Machairodont cats from the early Pleistocene Broad- water and Lisco local faunas. Bulletin of the University of Nebraska State Museum 9(2):1-—56. SCHULTZ, C. B., M. R. SCHULTZ, and L. D. MARTIN 1970 A new tribe of saber-toothed cats (Barbourofelini) from the Pliocene of North America. Bulletin of the University of Nebraska State Museum 9(1):1-31. SIMPSON, G. G. 1941 The function of saber-like canines in carnivorous mammals. American Museum Novitates 1130:1-12. 1945 The principles of classification and a classification of mammals. Bulletin of the American Museum of Natural History 85:1—350. SLAUGHTER, B. H. 1960 A new species of Smilodon from a late Pleistocene alluvial terrace deposit of the Trinity River. Journal of Paleontology 34:486—-492. 1963 Some observations concerning the genus Smilodon with special reference to Smilodon fatalis. Texas Journal of Science 15:68-81. STOLL, N. R., R. Ph. DOLLFUS, J. FOREST, N. D. RILEY, C. W. SABROSKY, C. W. WRIGHT, and R. V. MELVILLE, eds. 1961 International code of zoological nomenclature adopted by the 1Sth International Congress of Zoology. London, Published for the International Commission on Zoological Nomenclature by the International Trust for Zoological Nomenclature. 176 pp. TEILHARD DE CHARDIN, P. 1939 On two skulls of Machairodus from the Lower Pleistocene beds of Choukoutien. Part 1, The Machairodus of Locality 13. Bulletin of the Geological Society of China 19:235-256. TEILHARD DE CHARDIN, P. and P. LEROY 1945 Les félidés de Chine. Publications de I’Institut de Géobiologie 11:1-58. TEILHARD DE CHARDIN, P. and J. PIVETEAU 1930 Les mammiferes fossiles de Nihowan. Annales de Paléontologie 19:1-134. WEBB, S. D. 1974 The status of Smilodon in Florida. In Webb, S. D., ed., Pleistocene mammals of Florida. Gainesville, University Presses of Florida, pp. 149-153. WOODWARD, A. S. 1898 Outlines of vertebrate palaeontology for students of zoology. Cambridge, Cambridge University Press. 470 Pp. ZDANSKY, O. 1924 Jungtertiare Carnivoren Chinas. Palaeontologia Sinica, Ser. C, 2:1-153. 59 e F Oo ee ISBN 0-88854-305-0 BY | ISSN 0384-8159 j