BULLETIN of the FLORIDA STATE MUSEUM Biological Sciences Volume 20 1976 Number 3 LATEST PLIOCENE MAMMALS FROM HAILE XV A, ALACHUA COUNTY, FLORIDA JESSE S. ROBERTSON 0 e UNIVERSITY OF FLORIDA GAINESVILLE Numbers of the BULLETIN OF THE FLORIDA STATE MUSEUM, BIOLOGICAL SCIENCES, are published at irregular intervals. Volumes contain about 300 pages and are not necessarily completed in any one calendar year. CARTER R, GEBERT, Editor RHODA J. RYBAK, Managing Editor Consultants for this issue: CLAYTON E. RAY S. DAVID WEBB Communications concerning purchase or exchange of the publications and all manu- scripts should be addressed to the Managing Editor of the Bulletin, Florida State Museum, Museum Road, University of Florida, Gainesville, Florida 32611. This public document was promulgated at an annual cost of $2824.29 or $2.824 per copy. It makes available to libraries, scholars, and all interested persons the results of researches in the natural sciences, emphasizing the Circum-Caribbean region. Publication date: July 7, 1976 Price: $2.85 LATEST PLIOCENE MAMMALS FROM HAILE XV A, ALACHUA COUNTY, FLORIDA JESSE S. ROBERTSON, JR.1 SYNOPSIS: The mammalian fauna of Haile XV A is the first known from the Gulf Coastal Plain during Blancan ( Latest Pliocene) time. The mammals are repre- sented by 8 orders, 17 families, and 18 genera, of which two are new species and three are new to North America. Kraglievichia, a giant, extinct armadillo previously known only from Late Miocene and Pliocene deposits of South America, is reported for the first time in North America and a new species is described. A brief review of the subfamily Chlamytheriinae suggests that Plaina Castellanos is a synonym of Krag- lievichia Castellanos and that Holistetteria Castellanos is a synonym of Pampatherium Ameghino. Chlamytheres are shown to have been present in North America con- tinuously since the establishment of the Iate Cenozoic land bridge between North and South America. Das!/pus balus C Simpson ), a smaller extinct armadillo, is reported for the first time in the Blancan, marking its earliest occurrence anywhere. Glossotherium chapadmatensis C Kraglievich), a small mylodont ground sloth pre- viously known only from the early Pleistocene of Argentina, is reported from North America for the first time. This species may haye been ancestral to both G. harlani ( Owen) and G. robustus ( Owen). The Old World Flying Squirrel, Cryptopterus, is reported for the first time in the New World and a new species is described. The nearest related species occurs in the Late Pliocene deposits of West Germany. Close alliance of several Haile XV A taxa with South American Plio-Pleistocence forms strengthens the previously suggested correlation between the Chapadmalalanstage of South America and at least the early part of the Blancan stage of North America. The abundance of aquatic non-mammalian vertebrates, together with the largeextinct otter, Satherium, and the beaver, Castor, indicate that the environment ofdeposition was a permanent stream. The terrestrial community includes the tropical or subtropical edentates Kfagtievichia and Dasyptis. Probable forest indicators in-clude Mt/lohyus, Castor, Cryptopterus, Clossotherium, and Tapirus. The zooge6-graphy of several exotic forms is discussed. 1 The author is an Associate Professor in the Department of Biology, Jacksonville University,Jacksonville, Florida 32211. Most of this study was done. in partial fulfill*tent of the Ph.D.degree at. the University of Florida. Manuscript acc-epted 15 July 1974. ROBERTSON, JESSE S., JR. 1976. Latest Pliocene Mammals from Haile XV A, Alachua County, Florida. Bull. Florida State Mus., Biol. Sci. Vol. 20(3):pp. 111-186. 112 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 3 TABLE OF CONTENTS SYNOPSIS 111 INTRODUCTION - 112 ACKNOWLEDGMENTS 113 GEOLOGY - 118 SYSTEMATIC MAMMALIAN PALEONTOLOGY 114 Order Insectivora Cryptotis parea 114 Scalopus aquaticus 115 Order Edentata Glossotherium clwpadmatensis 116 Kfaglievichia new species 124 Review of the Chlamytheriinae 185 Dasypus bellus 142 Order Lagomorpha Syldlagus Sp. 146 Order Rodentia Cryptopterus new species 147 Castor canadensis 158 Sigmodon medius 158 Order Carnivora Canidae 154 154Satherium sp. 157Smilodon gracilis Order Proboscidea Gomphotheriidae 158 Order Perissodactyla 158Nannippus phlegon Equus (Dolichohippus) simplicidens 158 Equus C Hemionus ) calobatus 160 168Tapirus sp. Order Artiodactyla 168Mylohyus floridanus Hemiauchenia cf. macrocephala 167 Odocoileus u#ginianus 169 AGE AND CORRELATION 170 PALEOECOLOGY 176 178ZOOGEOGRAPHY SUMMARY 179 BIBLIOGRAPHY 181 INTRODUCTION Florida has long been noted for its rich Late Pleistocene fossil verte- brate deposits, including the Vero site in Indian River County ( Weigel 1962), the Melbourne Bone Bed in Brevard County ( Gazin 1950, Ray 1958), Seminole Field in Pinellas County ( Simpson 1930), Reddick ( Ray et aL 1963) in Marion County, and many others. In recent years a few 1976 ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 113 sites have yielded fossil vertebrates of older Pleistocene and even Latest Pliocene age. These Blancan sites are of particular interest, inasmuch as they constitute the Erst samples from eastern North American ( Webb 1974a). This report is the first systematic study of one of the Latest Pliocene sites. The Haile XV A locality is situated on the property of Parker Brothers Limestone Products Incorporated, near Haile, T9S, R17E, Section 25, NW 1/4 of SW 1/4, Alachua County, Florida. In 1964 the site was dis- covered and some specimens collected by Phillip Kinsey of Jacksonville Beach, Florida. Further excavations were carried out in the same year by S. David Webb, Robert Allen, and myself with the support of National Science Foundation Grant GB 3862 ACKNOWLEDGMENTS AND ABBREVIATIONS Fossil material was examined from the American Museum of Natural History (AMNH), the University of California, Museum of Paleontolegy ( UCMP), the Florida State Museum ( UF), the University of Houston ( HCT), the University of Kansas ( UK), and the United States National Museum of Natural History ( USNM). I wish to thank the persons in charge of these collections for the opportunity to study the materials in their care. I also wish to thank S. David Webb, Thomas H. Patton, and H. K. Brooks for their encouragement and help during the course of this study. GEOLOGY The Haile XV A deposit lies at an elevation of about 90 ft above sea level and is a filled fissure in the Ocala ( Eocene) Limestone. This fssure measures approximately 10 m along its east-west axis, and its north-south dimensions range irregularly from about 3 to 5 m ( Fig. 1). The bottom of the Bssure-filling sequence lies at a depth ranging from 1 to 7 m below the land surface. It consists of a stratum of dark brown, compact, iron-stained clays ( or "hardpan"), with a thickness ranging from 2 to 6 m. The fossil-bearing matrix overlies the "hardpan," and consists of an alternating sequence of sands and clays. The lowerniost unit is a coarse gravelly sand, containing calcareous cement and concretions. It is the thickest unit in the fossiliferous sequence, ranging from 2 to more than 6 m in depth, and is the most productive fossiliferous h6rizon. Many of the included larger bones are encrusted with cemented sand and gravel. The clay units are pure, greenish in color, massive in texture, and vary from 10 to 20 cm. The sand layers decrease'in thickness upward through the sequence, with the uppermost sands being less than 1 m thick. The clays vary little in thickness, except that the uppermost unit in the fos- siliferous section is nearly a meter thick. 114 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 3 >E - 1 E 1 1 03 33 A O D " C I 1 1 E- 1 1 Scal* in meten D A+-*7+- 1 /Il ll 012 Ll'Ill'Ill FIGURE 1. Geologic section at Haile XV A: (A) sandstone alternating with gray clay; (B) massive greenish clay; (C) coarse, poorly sorted, gravelly sandstone; (D) dark brown hardpan; and (E) Ocala Limestone ( Eocene ). In several instances fossil bones in a sand layer have been found pro- jecting partly into the overlying clay layer. In each instance the pro- jecting part of the bone shows no noticeable wear or weathering. Evi- dently each clay unit was deposited rapidly over the preceding sand unit, with little or no intervening period of erosion. Overlying the fossiliferous sequence of sands and clays is a layer of thin, brown sandstone that is nonfossiliferous and grades upward into the soil profile. SYSTEMATIC PALEONTOLOGY Twenty mammalian taxa, representing a minimum of 34 individuals, were collected at Haile XV A ( Table 1). ORDER INSEC[TVORA FAMILY SORICIDAE Cryptotis paroa ( SAY 1823 ) MATERIAL.-UF 17466, right mandible with Mi-M3• This specimen is indistinguishable from Recent Florida specimens of C. parva, except for its slightly larger size ( Table 2). The only other Blancan record of Cryptotis is from the Rexroad fauna ( Hibbard 1937, 1941), Meade County, Kansas. 1976 ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 115 TABLE 1.-MAMMALIAN FAUNAL LIST AND MINIMUM NUMBER OF INDIVIDUALS (MNI). Species MNI Cryptotis parua 1 Scalopus aquaticus 2 Glossotherium chapadmalensis 1 Kraglievichia Borid#nus 3 Dasupus bellus 1 St/luitagus sp. 3 Cryptopterus webbi 1 Castor canadensis 1 Sigmo,ion medius 2 Canidae 2 Satherium sp. 1 Smilodon gracilis 1 Comphotheriidae 1 Nannippus phlegon 2 Equus ( Dolichohippus) simplicidens Equus ( Hemionus) cf calobatus 2 Tapirus sp. 1 Mylohgus floridanus 1 Hemiauchenia cf. macrocephda 1 Odocoileus virginianus 5 TABLE 2.-MEASUREMENTS ( IN MM) OF THE LOWER DENTITION AND MANDIBLE OF UF 17466 AND OTHER F6sSIL AND RECENT Cryptotis parva C._parual (20 ) C, Erval ( fossil) UF 17466 X OR X2 OR) Condyloid to mental foramen 6.8 6.1 5.9-6.7 6.6 6.1-7.0 (8) Condyloid to Ml 6.7 6.9 6.7-7.4 7.1 6.8-7.6 (8) Condyloid to M2 5.3 5.7 5.5-6.0 5.8 5.54.1 (8) Mi-M• 2.8 2.6 2.4-2.7 2.7 2.5-2.9 (8) Ml-M3 4.0 3.4 3.3-3.6 3.6 3.5-3.6 (2) M2-MB 2.5 2.2 2.1-2.4 2.3 2.2-2.3 (2) ' Measurements after Martin ( 1974). 2 X=mean 8 0. R.=observed range; number of specimens in parenthesis FAMILY TALPIDAE Scalopus aquaticus LINNAEUS 1758 MATERIAL.-UF 17467, 2 left M2; UF 17465, right humerus, A comparison of the three specimens from Haile XV A with corre- sponding elements of modern S. aquaticus from the southeastern United States reveals no significant differences in size or morphology. Scalopus is also present in the Blancan Rexroad fauna of southwestern Kansas ( Hibbard 1941) and in the Sand Draw local fauna of north- central Nebraska ( Hibbard 1972). Another mole ( Hesperoscalops) from the Rexroad fauna is based on a lower dentition. A direct com- rp. 116 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 3 parison could not be made with that genus, because only upper teeth are known from Haile XV A. As Hibbard ( 1941) points out, however, mole humeri are quite diagnostic at the generic level, so it seems unlikely that a Scalopus-like humerus could belong to any other genus. ORDER El)ENTATA FAMILY MYLODONTIDAE Glossotherium ( Glossotherium ) chapadmalensis C KRAGLIEVICH 1925 ) Eumt/lodon chapadmatensis Kraglievich 1925 Clossotherium ( Clossotherium) chapadmalensis C Kragl.) Hoffstetter 1952 MATERIAL.-UF 10922, partial skeleton. The Haile XV A specimen probably represents one individual, be- cause there is no duplication of elements and the left and right elements are very similar in all measurements. Most, if not all specimens were taken from the basal sand layer of the fossiliferous sequence, and some parts of the skeleton and dermal ossicles were nearly articulated. CRANIUM AND UPPER DENTITION.-The upper dentition of the Haile XV A specimen ( Fig. 2) is complete except for the left upper caniniform. The right upper caniniform is triangular in cross-section and very well developed. The stoutness of the caninform is reflected by great trans- verse expansion of the anterior portion of the maxilla, as is also true of the holotype from Argentina. This contrasts strildngly with G. harlani, which has a reduced or absent caniniform and transversely narrow maxilla. The first upper molariform tooth is oval in cross-section. As in G. harlani and G. robustus, it is the longest anteroposteriorly of all the upper teeth but is relatively narrow in transverse diameter. Because the well- developed lower caniniform occludes in part with this tooth, wear on the anterior oblique surface in G. chapadmalensis is much more accentuated than in G. hadani. In G. robustus specimens observed, this tooth was worn off smoothly just above the alveolus. A very short diastema sepa- rates the upper caniniform tooth from the first molariform tooth in the specimen from Haile XV A. The second upper molariform tooth of UF 10922 is triangular in cross- section. The lingual side is the narrowest and contains a deep inflection. The transverse diameter is greater than the anteroposterior diameter in both the Florida and the Argentina specimens of this species, whereas the opposite is true in G. harlani. Of ·the two G. robustus specimens for which measurements are provided, one resembles G. chapadmatensis, whereas the two diameters in the other specimen are nearly equal. The third molariform tooth of UF 10922 is similar in shape to the second and morphologically similar to the corresponding teeth in both G. harkni 1976 -ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 117 1 FIGURE 2. Glossotherium chapadmalensis: upper dentition X 0.85. and G. robustus. The fourth molariform is relatively small compared to the Argentina specimen of G. chapadmalensis, although it is identical in shape. The palate is very constricted in the area of the last upper teeth and considerably expanded at the anterior end of the maxilla. Both of these 118 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 3 characters are more extreme in UF 10922 than in the holotype or in speci- mens of G. harlani and G. robustus. The remainder of the cranial material consists of part of the occipital bone and portions of the left and right periotic bone. The occipital frag- ment consists of the left condyle and a portion of the otic region. The hypoglossal canal, jugular foramen, and auditory meatus show no observ- able differences from those of G. harlani and G. robustus. A portion of the left squamosal bone is attached to the occipital, Part of the right squamosal is present, including the zygomatic process. MANDIBLE AND LOWER DENTITION.-The lower caniniform of UF 10922 ( Fig. 3,4) is well developed and directed dorsolabially. This tooth is relatively larger than the corresponding tooth in the South American specimen of G. chapadmalensis. The tooth has a chisel-like tip, the result of dual occlusion with the upper caniniform and first molariform teeth. The second molariform tooth is rectangular and has its long axis oblique to the tooth row. Stock ( 1925) measured the long axis as the anteroposterior length; Kraglievich ( 1925) used the shorter axis of the tooth. For purposes of comparison, Kraglievich's method is used here for the Haile XV A specimen and Stock's method for measuring G. robustus. ( For consistency one must reverse the measurements of G. robustus provided by Kraglievich [1928].) It is obvious from Table 3 that the orientation of the teeth, not their basic morphology, probably accounts for the interspecific differences in this tooth. The third lower molariform tooth is of a slightly different shape in G. chapadmalensis. The two main columns are separated by a thin bridge, while in G. robustus this bridge is thicker. Unfortunately, the mandibular symphysis is not preserved in the Haile XV A specimen, but the portion of the mandible present is similar to that of the other two species. VERTEBRAE.-The atlas of UF 10922 is the only well preserved verte- bra. According to Stock ( 1925) the atlas of G. harl(mi differs from that of G. robustus in the more posterior position of the lateral process. The location of the dorsal foramina also varies; in G. harlani they are well separated as compared to those in G. robustus. In the expression of these two characters, UF 10922 resembles G. harlani. LIMB BONES.-The right humerus and both ulnae are present, as are the left femur, both tibiae, and the left flbula. Hind foot bones repre- sented are the astragalus, calcaneum, and various metatarsals and pha- langes. The small size of this species is indicated by the measurements of these limb bones ( Table 4). Several bones of the manus warrant a brief description. The cunei- form appears to be somewhat different in G. chapadmalensis than in G. 0 1976 ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 119 4..+4 FIGuRE 3. Glossotherium chapadmatensis: right mandible, lateral view, X 0.60. 120 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 3 FIGURE 4. Glossotherium chapadmalensis: right mandible, occlusal view, X 0.60. 1976 R O B E R TS O N : H A ILE X V A LAT E S T P LIO C E N E M A M M A LS 121 TABLE 3.-MEASUREMENTS ( IN MM) OF THE UPPER AND I.0WER DENTTIONS OF THREE SPECIES OF Glossotherium. G. chapadmalensis G. harlani G. robustus Haile XV A Argentinal California: Argentina3 UPPER DENTITION Anteroposterior length, tooth 1 14.7 14.0 17.9 17.9 Transverse length, tooth 1 10.6 10.5 16.0 18.2 Anteroposterior length, tooth 2 22.4 22.0 33.3 22.3 Transverse length, tooth 2 12.0 15.0 17.4 16.7 Anteroposterior length, tooth 3 15.6 20.5 27.1 22.8 Transverse length, tooth 3 22.4 24.0 23.0 23.0 Anteroposterior length, tooth 4 17.9 15.0 22.6 20.3 Transverse length, tooth 4 20.5 24.0 23.8 22.4 Anteroposterior length, tooth 5 21.4 24.0 24.3 23.3 Transverse length, tooth 5 13.6 18.0 19.4 19.2 LOWER DENTTION Anteroposterior length, tooth 1 20.5 14.5 21.4 20.4 Transverse length, tooth 1 12.4 10.5 16.0 15.0 Anteroposterior length, tooth 2 18.7 18.0 28.3 23.8 Transverse length, tooth 2 - 14.5 22.0 22.5 22.8 Anteroposterior length, tooth 3 14.5 14.0 28.8 25.0 Transverse length, tooth 3 20.5 25.0 18.2 19.0 Anteroposterior length, tooth 4 40.3 43.0 51.9 50.2 Transverse length, tooth 4 17.3 18.0 23.8 20.2 1 Measurements after Kraglievich ( 1925). 2 Measurements after Stock ( 1925). 0 Measurements after Owen ( 1842). 122 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 3 TABLE 4.-LIMB MEASUREMEXrs ( IN MM ) OF Glossotherium chapadmatensis AND Glosotherium hartani. C. chapadmalensis G. harlanil UF 10922, Haile XV A Rancho La Brea left right HUMERUS Greatest anteroposterior distance of head 70.0 70.0 123.8 Greatest width of shaft at deltoid ridge 71.3 122.6 Width of distal articular surface 86.0 133.2 ULNA Greatest length 285 395.9 Width of distal articular surface 34 55.0 FEMUR Total length 355 546.4 Transverse diameter of head 79.3 127.8 Least width of shaft 99.5 164.6 Greatest width across distal tuberosities 140 234.8 Width of distal condyles 111 188.5 Width of intercondyloid space 27.1 47 Width of inner condyle 47.2 47.2 88.7 Vertical extent of inner condyle 67 69.8 120 TIBIA Total length 181 176 247.3 Greatest width, proximal end 114 185.2 Greatest width, distal end 90 88 142.5 Anteroposterior distance, distal end 64.8 66.6 101.9 FIBULA Total length 183 263 Width, proximal end 50.5 103.6 Width, distal end 39.5 73.5 , After Stock ( 1925). The measurements represent the mean of large samples. The number of specimens varied from element to element and in some cases was not given. harlani; in the UF specimen it has a square palmar outline, whereas in Stock's figure ( 1925: Fig. 72d) of G. harlani it is rectangular. This dif- ference is reflected by the measurements of the cuneiform in Table 5. In UF 10922 the proximal-distal dimension is greater than the distance across the ulnar articular surface, while the opposite is true for 39 speci- mens measured by Stock. The cuneiform appears to be a deeper element in G. chapadmalensis than in G. harlani. The remaining elements of the manus preserved in the Haile XV A specimen are the left scaphoid, right pisiform, left metacarpal III, and left phalanx II, digit III. DISCUSSION.-This specimen so closely resembles Eumylodon ( - Glos- sotherium) chapadmalensis that it is here ascribed to that species. G. 1976 ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 123 TABLE 5,-MEASUREMENTS ( IN MM ) OF THE BONES OF THE MANUS AND PES OF Glossotherium chapadmatensts FROM HAILE XV A AND Giossotherium harlani. G. chapadmatensis left right G. harlanil SCAPHOID Greatest distance across articular surface 46.2 75.3(39)2 CUNEIFORM Greatest distance across dorsal surface from inner side to outer 42.1 68.5(42) Greatest proximal-distal distance 34.3 50.5 Distance across articular surface for ulna 31.2 61.3 PISIFORM Greatest length 25.3 55.6(22) Greatest depth 18.9 36.4 GreateSt width 20.9 34.7 METACARPAL III Greatest length 71.7 103.1(30) Width, proximal end 51.9 75.2 Width, distal end 30.5 50.9 Depth, proximal end 42.0 66.0 PHALANX II, DIGIT III Length 40.5 48.9(42) Depth, inner condyle 23.5 39.9 Width, proximal end 26.2 47.9 CALCANEUM Greatest width, anterior end 57.4 91.8(21) Greatest depth 83.4 125.4 ASTRACALUS Anteroposterior diameter 76.1 140.2(41) METATARSAL IV Length 84.3 119.3(29) Width, proximal end 41.2 41.5 60.7 Depth, proximal end 34.3 37.1 45.5 Width, distal end 38.5 50.1 Depth, distal end 35.3 43.5 PHALANx II, DIGIT III Length 30.2 31.0 35.4(32) Depth, inner condyle 20.4 21.5 21.8 Width, proximal end 22.6 22.5 35.1 UNGUAL PHALANX, DIGIT III Length 104.4 174.1 Proximal-distal distance of ungual base 41.0 73.5 Width, proximal end 32.4 55.0 1 Measurements after Stock ( 1925). 2 Number in parentheses refers to the sample size from which the mean (the numbers in the right-hand column) was calculated. 124 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 3 chapadmalensis was described frcim the Chapadmalal fauna ( Early Pleistocene) of Argentina and is here reported from North America for the Brst time. It is a small species that appears to be very closely re- lated to both G. harlani and G. robustus, but it differs from these two species in the nature of the anterior dentition and overall size. It re- sembles both in the morphology of most of the postcranial elements, with minor exceptions in the manus and pes. G. chapadmaZensis is thus tem- porally, geographically, and morphologically a plausible ancestor of both G. harlani and G. robustus. Much confusion has arisen concerning proper usage of the names Mylodon and Glossotherium. In his thorough review of the literature, Kraglievich ( 1928) concluded that Mylodon darwinii is the genotype of Mylodon, and the species described by Owen ( 1840) as Mylodon ro- bustus should actually be included in a distinct genus, Glossotherium. However, Kraglievich retained Paramylodon C Brown 1903) as a valid genus, even though Stock ( 1925) included it in the synonomy of Mt/lodon harlani. Simpson ( 1945) correctly argued that if the North and South American Late Pleistocene forms are not generically distinct, they should all be referred to the genus Glossotherium. Hoffstetter ( 1952) retained Paramylodon as a subgenus of Glossotherium, this distinction being based on the narrow muzzle, the long cranium, and frequent absence of the upper caniniform tooth. Following this usage, the Haile XV A species should be called Glossotherium ( Glossotherium) chapadmalensis. FAMILY DASYPODIDAE Kraglievichia floridanus new species HOLOTYPE.-UF 10902, partial skull collected by P. E. Kinsey, S. D. Webb, R. R. Allen, and J. S. Robertson in 1964. TyPE LOCALITY AND HORIZON.-Haile XV A, T9S, R17E, Sec. 25, NW 1/4 of SW 1/4, AIachua County, Florida, Blancan. DIAGNOSIS.-Differs from Kragliectchia paranensis in having the fourth upper tooth reniform rather than peg-like, with the long axis anterolingual instead of parallel to the tooth row. REFERRED MATERIAL.-UF 10902, mandible and postcranial elements ( probably the same individual as the type skull); UF 17474, right ulna, also from Haile XV A; UF 10432, right humerus, Santa Fe I; UF 9354, right humerus, Santa Fe I; UF 10830, right radius, Santa Fe I; UF 16371, left metatarsal IV, Waccasassa River; UF 17475, right metatarsal IV, Santa Fe II; UF i7472, right metatarsal III, Santa Fe II; UF 17476, right femur, Haile XIII; UF 17568 and UF 17569, right naviculars, Haile XV A. DESCRIPTION.-The Haile XV A skeleton and isolated postcranial material from other sites in Florida provide the Brst adequate samples of the posteranial skeleton of this genus. The following description is based primarily on the Haile XV A speci- 1976 ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 125 men; but because certain elements are lacking, it has been supplemented by material from other Florida Blancan localities. CRANIUM.-The nasal bones ( Fig. 5b) of UF 10902 compare favor- ably in size and morphology with those of the South American specimen of Kraglievichia paranensis described by Castellanos ( 1927). They differ, however, in having tiny protuberances on the anterior ends near the medial surface. These structures are not present on any of the other specimens or figures of Pampatherium or Kraglievichia studied. Only the anterior halves of the nasals are present in UF 10902, because the skull had been eroded away, and only parts of it recovered. The premaxilla ( Fig. 5a) contains only one alveolus. The premaxil- lary-maxillary suture forms the posterior border of the first alveolus ( or tooth socket), as in Pampatherium. Holmesina ( from North America) was originally thought to be distinguished by the presence of a single premaxillary tooth; two such teeth were supposed to occur in Pampa- therium. A single alveolus occurs in the neotype of «Holmesina" ( AMNH 26856; Simpson 1930) as well as in UF 889 ( P. septentrionalis). However, it now appears that the true Pampatherium also has only one tooth in the premaxilla ( G. Edmund, pers. comm.), and, therefore, the supposed difference between the North and South American forms is probably not valid. The major features of the maxilla from Haile XV A compare favorably with those in Pampatherium. The infraorbital foramen ( Fig. Se) is located directly above the sixth tooth in both genera. The anterior palatal foramina ( Fig. 5c) are located between the posterior edges of the fourth teeth in the Haile XV A specimen, whereas their position varies somewhat in Pampatherium. The maxillary process of the zygo- matic arch lies directly above the seventh tooth in both Kraglievichia and Pampatherium. Only a portion of the zygomatic arch is present in UF 10902 ( Fig. 5d-e). The zygomatic process of the maxilla turns posteriorly and down- ward as it leaves the skull. The anterior portion of the jugal, which borders the zygomatic process laterally, then turns upward and expands posteriorly to accept the squamosal process. A well-developed suture, located at the posterior end of the jugal, indicates that the zygomatic arch is complete in Kraglievichia, as it is in Pampatherium. The base of the zygomatic process of the maxilla is expanded by sinuses, as in Pampatherium UPPER DENTITION.-There are nine teeth in both the upper and lower jaws of Kraglievichia. One of the upper teeth is located in the premaxilla. ( Because the incisors of most mammals occur in the premaxilla, it is tempting to refer to the first tooth in Kraglievichia as an 126 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 3 prnx V 7: 1 - 0· 414~ R' . i- FE i , : Y. *9,41 6 4Y '1 4»~ A A, el, 8 1. C f --4*:x -- NekN.42x ~w- 1\ ju, , A», 'Y y A- G---- 8 9 ' Zi f 2979/ &nD ~6* 3*. , 1 1, ,'Ar--'of 7- P 0 8--IT-5 89 aR~8:-#1# 11!81 4V HI - 126 mrn G FIGuRE 5. Kraglievichia 8oridanus, cranial material: (A) right premaxillary; (B) nasal; (C) palatal portion of maxillary; (D) zygoma, ventral view; (E) zygoma, lateral view; (F) right mandible, occlusal view; (G) right mandible, lateral view; (H) first lower tooth; (I) fourth lower tooth. Abbreviations: iof=infraorbital foramen, ju = jugal, mx= maxillary, mx-pmx = premaxilla, n = nasals, numbers refer to tooth sequence. 1976 ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 127' incisor; however, I have continued to use the non-committal terminology "tooth one" through "tooth nine" in this report.) The fourth tooth pro- vides critical evidence for the phylogenetic position of the Haile XV A species. The alveolus for this tooth suggests that it was incipiently bilo- bate ahd nearly parallel with the more posteri6r teeth, a condition inter- mediate between earlier Kraglievichia and later Pampatherium. Measurements of the cranium and upper dehtition are presented in Table 6. MANDIBLE AND LOWER DENTITION.-Partial left and right mandibles are preserved in UF 10902. In the right mandible ( Fig. 5f-g) the 7th tooth is complete, the 8th is broken, and alveoli for the 6th and 9th teeth are present. Unfortunately, the anterior part of each mandible is lacking in UF.10902. Other than size, no differences are apparent between Krag- lievichia and Pampatherium with regard to the posterior parts of the mandibles ( see Table 6). Vertebrae.-The vertebral material from the Haile XV A skeleton consists of 7 thoracic, 5 lumbar, and 4 caudal vertebrae. Five of the thoracic vertebrae are cemented together in the proper sequence by coarse sandstone. Comparison of these specimens with the vertebrae of a well-preserved skeleton of Pampatherium from Branford IA, Suwan- nee County, Florida ( Rancholabrean), shows no significant morphologi- cal differences except for size. FRONT LIMBS.-In this description the major elements of the forelimb are compared with those of both Dasypus and Pampatherium. Kraglie- vichia #oridanus is approximately the same size as Rancholabrean speci- mens of Dasypus bellus ( a large extinct Pleistocene armadillo) and con- ceivably the limb elements of these two species could be confused. The podial elements and hind-limb elements should never be confused, how- ever. The right humerus of UF 10902 is well preserved, except for the distal end ( Fig. 6a-b). The humerus of Kraglievichia is greatly expanded laterally (as seen in the specimens from Santa Fe I). The supracondy. lar foramen is relatively larger than in Dampus. The articular surface for the radius is concave in Kraglievichia, whereas it has a slight con- vexity in Dasypus. The supinator ridge is relatively narrow in Kraglie- uichia, and the deltoid ridge is wider. Comparison of the proximal end of the humerus in Kraglievithia and Pampatherium shows no distinct differences, except for size. Three ulnae are known from the Haile XV A site; two of which are probably from the same individual because they are similarly preserved, are from opposing sides, and agree closely in measurements ( Table 7). The ulna of Kraglievichia ( Fig. Bc-d) is laterally flattened and has a long 128 B U LLE T IN FLO R ID A S TATE M U S E U M Vol. 20, N o. 3 TABLE 6.-CRANIAL AND DENTAL MEASUREMENTS ( IN MM ) OF Kraglieutchia AND Chlamytherium. Kraglievichia Kraglievichia Pampatherium Pampatherium UF 10902 AMNH 26856 UF 889, 890 Haile XV A Argentinal Florida Florida CRANIUM Width of nasals at anterior border of premaxilla 25.9 27.6 39.0 Width of palate at center of tooth 4 15.0 17.4 28.02 29.6 Anteroposterior length, tooth 4 12.3 12.6 - 16.7 Anteroposterior length, tooth 5 15.4 12.7 23.7 20.5 Anteroposterior length, tooth 7 16.4 15.0 22.6 23.6 Width, tooth 1 6.8 6.5 7.6 MANDIBLE Distance, top of ninth alveolus to bottom of jaw 37.8 32.8 60.2 51.4 Length, tooth 1 5.4 Width, tooth 1 3.7 Length, tooth 3 8.2 12.3 Width, tooth 3 5.2 . 6.9 Length, tooth 6 ( alveolar) 16.0 14.8 25.1 25.1 Length, tooth 7 14.3 13.9 23.3 22.1 Width, tooth 7 7.1 11.2 9.2 Width, tooth 8 ( alveolar, anterior lobe) - 11.9 10.5 10.0 Length,. tooth 9 10.3 . 13.8 Width, tooth 9 5.5 7.8 1 Measurements after Castellanos ( 1927)- 2 Estimated. 1976 ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 129 r •.Di · * f 4 4 A B 1/ 4 EF C D FIGURE 6. Kragliecichia Boridanus, front limb elements: ( A) right humerus, medial view; (B) right humerus, anterior view; (C) left ulna, medial view; (D) left ulna, anterior view; (E) left radius, posterior view,(F) left radius, anterior views; X 0.45. 130 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 3 TABLE 7.-MEASUREMENTS (IN MM) OF LAMB ELEMENTS OF Kraglievichia floridanus, UF 10902, HAILE XV A. Left Right HUMERUS Lateral width, proximal end 36.7 Anteroposterior width, proximal end 36.8 Greatest anteroposterior diameter of shaft 25.5 Lateral width of shaft at same location 18.2 Distance from proximal end to top of entepicondylar foramen 100.5 ULNA Total length 129.0 Lateral width, proximal end 17.2 Lateral width, distal end 16.1 Lateral width at semilunar notch 25.5 25.7 Anteroposterior width at semilunar notch 29.4 30.5 RADIUS Total length 89.2 87.2 Lateral width, proximal end ' 22.6 22.3 Anteroposterior width, pr6ximal end 11.7 11.4 Lateral width, distal end 23.0 22.3 Anteroposterior width, distal end 17.4 16.5 METACARPAL II Total length 32.8 Width, proximal end 10.9 Depth, proximal end 14.5 Width, distal end 10.5 Depth, distal end 11.2 METACARPAL III Total length 34.4 36.3 Width, proximal end 13.8 14.5 Depth, proximal end 12.8 13.0 Width, distal end 12.6 13.4 Depth, distal end 10.5 11.0 UNCUAL PHALANX, DIGrr V Total length 29.1 Width, proximal end 11.4 Depth, proximal end 10.3 FEMUR Total length 193.5 Lateral width, proximal end 60.9 Anteroposterior thickness of greater trochanter - 41.7 Lateral width of shaft at third trochanter 41.3 Anteroposterior thickness of shaft at same point 23.7 Greatest lateral width of articular facets 48.4 Greatest anteroposterior width, distal end 51.5 TIBIA AND F,BULA Total length 121.0 Lateral width, distal end 46.7 Ahteroposterior width, distal end 25.1 1976 ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 131 TABLE 7,-CONTINUED CALCANEUM Total length 63.3 Width of articular facets for astragalus 28.0 Depth of facet for cuboid 12.2 Width of facet for cuboid 11.5 NAVICULAR Greatest lateral width 39.5 Greatest anteroposterior distance 23.3 Greatest depth 22.5 MrrATARSAL II Total length 33.2 32.8 Width, proximal end 12.2 12.5 Depth, proximal end 13.8 14.3 Width, distal end ( articular surface) 13.2 13.7 Depth, distal end ( articular surface) 11.3 10.2 METATARSAL III Total length 36.5 34.5 Width, proximal end 15.0 14.9 Depth, proximal end 14.7 14.4 Width, distal end ( articular surface) 13.2 12.3 Depth, distal end ( articular surface) 11.3 11.5 UNGUAL PHALANX, DIGrr V Total length 24.6 Width, proximal end 17.6 Depth, proximal end 10.7 olecranon process. The articular facets for the radius and medial condyle of the humerus are combined in this form, whereas in Dasypus they are partially divided. Another striking difference is a lateral groove that runs the entire length of the ulna in Dasypus and terminates at the upper border of the semilunar notch in Kraglievichia. Size appears to be the only difference between the ulna of Kraglievichia and Pampatherium. In both Kraglievichia and Dasypus the distal end of the radius is massive compared to the proximal end, although it is less flattened in Kraglievichia ( Fig. Ge-f), The proximal portion of the shaft is relatively thicker and less curved in Kraglievichia. Comparison with Pampather- ium shows only a size difference. The left metacarpal II is present in the Haile XV A skeleton, and generally agrees with that of Pampatherium in shape and proportion. The facet for articulation with the trapezoid, however, shows an im- portant difference; when viewed laterally, this facet is smoothly curved in Pampatherium and has a sharp V-shape in Kraglievichia. The magnum facet is oblong in Pampatherium and round in Krag- lievichia. The trapezoid facet is wider dorsally in Pampatherium. To 132 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 3 compensate for this, the facet for metacarpal III bulges out proximally to form a pocket for the trapezoid facet. In Kraglievichia less of the trapezoid articulates with metacarpal III. The left and right third metacarpals are preserved in UF 10902 from Haile XV A and also in UF 10722 from Santa Fe I. In this element. the facet for metacarpal II and the trapezoid shows a relatively greater as- sociation with the trapezoid than with the adjacent metacarpal. In both Pampatherium and Kraglievichia the magnum facet is convex dorsally and concave ventrally. In Kraglievichia the greater portion of the facet is convex, whereas the opposite is true in Pampatherium, When viewed from below, the ventral portion of this facet lies oblique to the long axis of the bone, whereas in Pampatherium it is perpendicular to the long axis. HIND LIMBS.-The femur ( Fig. 7a-b), the tibia, and the fibula ( Fig. 7c-d) show no morphological differences from those of Pampatherium except for size. The calcaneum of KragUevichia is less expanded distally and less robust than that of Pampatherium ( Fig, 8). In Kraglievichia the two astragalar facets are connected to form a bilobed facet, whereas in Pam- patherium they are separated by a central valley. In Kraglievichia the facets are nearly equal in size, whereas in Pampatherium the lateral facet is much the larger of the two. There are no apparent differences in the naviculars of Kraglievichia and Pampatherium. It may be noted that the three right naviculars from Haile XY A exhibited little variation in the relative shapes and sizes of the facets in which the cuneiform bones articulate. In metatarsal II the mesocuneiform facet is relatively more narrow ventrally in Kraglievichia than in Pampatherium. There is an indenta- tion at the proximal end, at the point where it meets the proximal end of the metatarsal I; in Pampatherium there is no such indentation, the medial border forming an unbroken line. The metatarsal I of Kraglie- uichia presumably had a corresponding projection; no such feature oc- curs in metatarsal I of Pampatherium. In lateral view the proximal articular surface of metatarsal III appears rounded in Kraglievichia, while in Pampatherium it forms a straight line perpendicular to the long axis of the element. Because the proximal end of this element is rounded in Kraglievichia, the articular surface can also be seen when the element is viewed from above. This would appear to permit more dorsoventral movement of the toes of Kraglieuichia. The facets for metatarsal III and the cuboid are united on metatar5al IV in Kraglievichia, but separate in Pampatherium. Two ungual phalanges of digit V ( UF 10902) have been recovered 1976 ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 133 4 0, A B C D Fi,GuRE 7. Kraglievichia 8oridanus, hind limb elements: (A) right femur, anterior view; (B) right femur, lateral view; (C) right tibia and fibula, anterior view; and ( D) right tibia and fibula, lateral view; X 0.45. 134 B U LLE T IN F LO R ID A S TATE M U S E U M Vol. 20, N o. 3 E A B C FIGURE 8. Kraglievichia and Pampatherium, calcanea, anterior views: (A) Kraglievichia ~loridanus, right calca- neum, Haile XV A; (B) Kraglievichia sp., right calcaneum, Inglis IA; and (C) Pampatherium septentrionalis, right calcaneum, Branford IA; X 0.80. 1976 ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 135 from Haile XV A; one ( presumably from the forefoot) is rather narrow and pointed, and the other ( presumably from the hind foot) blunt and broad. They agree in all features except size with the corresponding phalanges in Pampatherium. Measurements of the limb element in other examples of Kraglievichia from various Florida localities are pre- sented in Table 8. REVIEW OF THE CHLAMYTHERIINAE The genus Kraglievichia is reported here for the first time in North America. In order to place the genus in a meaningful context, the sub- family Chlamytheriinae is reviewed briefly. The nomenclatural history of the various chlamythere genera will be discussed, followed by a brief characterization of the various genera, beginning with the oldest. The Brst remains of a chlamythere were discovered in a Brazilian cave deposit in 1836 by Peter Wilhelm Lund and described by him as Chlamytherium humboldtii ( Lund 1838 ). Lund consistently used his original generic name in his early works, but later ( beginning about 1840) he emended this to Chlamydotherium, calling the earlier name a lai)ms. Several later authors followed this secondary spelling. Bronn ( 1838) meanwhile had given the name Chlamydotherium to a genus of glyptodonts. Ameghino ( 1875) proposed Pampatherium as a substitute for Lund's supposedly preoccupied generic name, Chfamgdotherium, but he later discontinued use of the new name after checking the spelling in Lund's origin al description. Paula Couto ( 1956) has revised the use of Pampatherium, which has come back into general usage. The first North American record of Pampatherium was reported by Leidy ( 1889a), although he Brst designated the new species Glyptodon septentrionale. Leidy ( 1889b) later referred the same material to the South American species Chlamytherium humboldtii. Sellards ( 1915) believed that the North and South American forms represented different species and resurrected Leidy's original specific name septentrionale. Ameghino ( 1902) described Machlydotherium from the Eocene of Patagonia. This still stands as the earliest record for a chlamythere. Castellanos ( 1927) named two new genera of chlamytheres: Vassallia, based on an edentulous mandible and several dermal plates ; and Krag- lievichia, based on two skulls, a mandible, and a small amount of post- cranial material. The type species of Vassallia is Chlamytherium minu- tum ( Moreno and Mercerat 1891). Kraglievichia was erected to include C. paranesis, C. intermedia ( Ameghino 1887), and C. subintermedius Rovereto 1914), with C. paranends as the generic type. Simpson ( 1930) established the genus Hollnesina, including in it the 136 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 3 TABLE 8.-MEASUREMENTS (IN MM) 07.LIMB ELEMENTS OF Kraglieulchia FROM VARIOUS FLORIDA LOCALITIES. UF 10432 ( right) UF 9354 ( right) Humerus ~ Santa Fe I Santa Fe I Greatest anteroposterior diameter of shaft 28.8 Lateral width of shaft at same point 19.5 Lateral width, distal end 60.2 54.2 Greatest width of distal articular surface 37.8 36.8 UF 10449 (left) UF 17474 ( right) Ulna Santa Fe I Haile XV A Total length 116.5 Lateral width, proximal end 17.6 Lateral width, distal end 10.4 Lateral width at semilunar notch 20.5 25.1 Anteroposterior width at sen*lunar notch 23.4 31.8 UF 10830 ( right) Radius Santa Fe I Total length 105.2 Transverse width, proximal end 24.5 Anteroposterior width, proximal end 13.3 Transverse width, distal end 23.1 Anteroposterior width, distal end 20.6 UF 17476 (left ) Fernur Haile XII B Lateral width, proximal end . 55.3 Anteroposterior width, proximal end 34.2 Lateral width of shaft at third trochanter 38.8 Anteroposterior width of shaft at same point 16.5 UF 17473 (left) Calcaneum Santa Fe II Total length 64.1 Width of articular facets for astragalus 24.5 Depth of facet for cuboid 14.9 Width of facet for cuboid 10.6 UF 17568 ( right) UF 17569 ( right) Navicular Haile XV A Haile XV A Greatest lateral width 37.3 33.6 Greatest anteroposterior distance 22.6 18.3 Greatest ' depth 30.6 26.6 UF 10902 ( left) Metatarsal II Haile XV Total length ~ '30.8 Width, proximal end 11.3 Depth, proximal end , 13.6 Width, distal end ( articular surface) 12.3 Depth, distal end ( articular surface) 10.2 1976 ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 137 TABLE 8.--CONTINUED UF 17472 ( right) Metatarsal III Santa Fe H Total length 34.2 Width, proximal end 16.5 Depth, proximal end 15.2 Width, distal end ( articular surface) 13.5 Depth, distal end ( articular surface) 11.4 UF 17475 ( right) Metatarsal IV Santa Fe II Total length 29.5 Width, proximal end 11.2 Depth, proximal end 12.2 Width, distal end ( articular surface) 15.1 Depth, distal end ( articular surface) 10.5 species septentrionalis, after studying well-preserved material from the Seminole Field in western peninsular Florida. Subsequent authors tended to refer· the North American forms to Holmesina and those from South America to Chlamytherium. Castellanos ( 1937) later proposed a new genus, Plaina, the type species of which ( C. intermedius) he had earlier placed in Kraglievichia. His rationale for establishing this new genus was based largely on his interpretation of the lineage of the chlamytheres. Castellanos believed that logically there should be a form intermediate in size between Krag- lievichia and Chlamytherium, and because C. intermedius is larger than the other material referred to Kraglievichia, he saw it as representing this intermediate form. The genus Ho#stetteria Castellanos ( 1957) was based upon a skull collected in Ecuador, which had been described previously as a new species, C. occidentalis, by Hoffstetter ( 1952 ). Machlydotherium is.the oldest of the various. chlamythere genera, its remains being known from the Eocene, of Patagonia. Its precise re- lationship to later chlamytheres is not clear ( Simpson 1945 ), a situation that may be attributed partly to the absence of any Oligocene records of the subfamily. Vassallia occurs in Araucanean ( Pliocene ) deposits in Argentina ( Castellanos 1946) and is also known ( a skull; UCMP 40401) from the Late Miocene La Venta fauna of Colombia. This identification is based upon the nature of the anterior dentition (the Erst Bve teeth are peg-like and rotated lingually) and the small size of the specimen. Porta (1962) reported Kraglieuichia from the La Venta fauna of Colombia, but this record was based only upon dermal plates. It now seems more likely that they represent Vassallia instead. 138 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 3 Castellanos ( 1927) listed two major morphological differences be- tween Vassallia and Kraglievichia. In Vassallia the first five teeth tend ~ to be peg-like and rotated lingually, whereas in Kraglievichia only the first four teeth show this condition. Vassallia is also significantly smaller than Kraglievichia. I suspect here, as did Castellanos, that Vassallia is the ancestor of Kraglievichia. So far as the status of Plaina is concerned, Castellanos ( 1927) was probably correct in his initial judgment when he assigned the material on which this genus was based ( three isolated dermal plates) to the genus Kraglievichia. He considered these plates to be intermediate in size and sculpturing between Kragliedchia and Pampatherium, al- though they actually fall within the smaller size range of Kraglievichia. Furthermore, it seems illogical to interpret slight variations in plate rugo- sity as generically significant. I therefore propose that Plaina be re- garded as a synonym of Kraglievichia. The genus Ho#stetteria Castellanos ( 1957) is based upon supposed differences in shape and measurements of the teeth. These differences are minor, however, and may be attributed to specific.variation. Holl- stetteria is thus considered to be a synonym of Pampatherium In Simpson's ( 1930) description of Holmesina, he listed a number of characters by which this new genus differed 'from Pampatherium. James ( 1957) subsequently showed that the characters given by Simpson are not sufEcient to separate the North and South American forms generi- cally, and that Holmesina is a synohym of Pampatherium. I support James' conclusion. Simpson ( 1930) argued also that if Holmesina was not valid, then all South America forms should be placed in the genus Pampatherium, as they are no more different from each other than is Pampatherium from Holmesina. This argument is not without merit. However, new characters discovered during the present study, consid- ered together with those previously recognized, provide ample justifica- tion fOr recognition of Kraglievichia and Pampatherium as distinct genera. To summarize, the genera of chlamytheres recognized here are: Machlydotherium ( Eocene, South America); Vassallia ( Miocene and Pliocene, South America); Kraglievichia, including Plaina ( Pliocene, South America and Plio-Pleistocene, North America) ; and Pampather- ium, including Ho#stetteria and Holmesina ( Pleistocene, North and South America). The most obvious morphological trend in chlamythere evolution since the Miocene has been a consistent increase in size. Vassallia is slightly smaller than Kraglievichia, which in turn is substantially smaller than 1976 ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 139 Pampatherium. This increase in size may be seen by comparison of the two femora in Figure 9. Inasmuch as the increase in size noted above occurred gradually, many intermediate stages logically would be expected. A series demon- strating the changes in size between Krailievichia and Pampatherium appears in Figure 10. The dermal plates illustrated are from various Plio-Pleistocene stages in Florida. Plates from the Early Blancan Haile XV A site are the smallest, followed by those from the Early Irvingtonian Inglis IA site. Still larger plates occur in the later Irvingtonian Coleman IIA site, and the largest plates of all are those from the Rancholabrean Branford IA locality. Another major trend in chlamytheriine evolution is the shift from peg-like to bilobate tooth shape and from an oblique to a parallel orienta- tion. This change seems to have progressed in an anterior direction, inasmuch as an increase in the number of anteriorly-situated bilobate and parallel teeth may be seen in the transition from Vassallia through Pampatherium C Castellanos 1937). The Haile XV A specimen differs from previously described speci- mens of Kraglievichia and Pampatherium in the nature of the anterior upper dentition. In South American specimens of Kraglievichia, the first four teeth are oval in cross-section rather than reniform and have their long axes turned lingually. In Pampatherium the Brst three teeth show this condition, but the fourth is usually bilobate and parallel with the rest of the tooth row. In James' ( 1957) specimen the first four teeth are oval, but the long axis of the fourth tooth parallels the rest of the tooth row, rather than being canted lingually. The peg-like appearance of the fourth tooth is presumably a primitive condition that recurred in an aberrant individual. The incipiently bilobed condition in the fourth upper tooth of Kraglievichia jiorid(inus represents the transitional condition between Kraglievichia and Pampatherium, although the fea- tures of the feet remain primitive in several respects. Chlamytheres apparently originated in South America, because all pre-latest Pliocene records of this group are restricted to that continent. The oldest genus is the Eocene Machlydotherium. No chlamytheres are known from the Oligocene, but Vassallia is present in Miocene deposits, and both Vassallia and Kraglievichia occur in the Pliocene of South America. Kraglievichia is here reported in Blancan deposits of North America. Pampatherium is present throughout the Pleistocene in both North and South America. This distribution presents some interesting geographic and phylogenetic problems. Until now the evolution of the chlamytheres was thought to have occurred exclusively in South America, with only the end product, 140 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 3 B A FIGURE 9. Pampatherium and Kraglievichia, femora, anterior views: (A) Pam- patherium septentrionalis, right femur, Branford IA; and (B) Kraglievichia #or- idanus, right femur, Haile XV A; X 0.45. 1976 ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 141 Pampatherium, having migrated to North America during Ranchola- brean time. Study of the Haile XV A fauna and other pre-Ranchola- brean sites of Florida indicates instead that Kraglievichia migrated to D 0 C FIGURE 10. Kraglievichia and Pampatherium, dermal plates, dorsal views: (A) Kraglievichia flon'danus, Haile XV A; (B) Kraglieuichia sp Inglis IA; (C) Pam- patherium septentrionalis, Coleman IIA; and (D) Pampatherium septentrionalis, Branford IA; X 0.30. 142 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 3 North America shortly after the establishment of the Late Cenozoic land bridge between the two continents, and that chlamytheres evidently continued to evolve in Florida and the Gulf Coastal Province throughout the Pleistocene. Three hypotheses can be offered regarding the Pleistocene zoogeo- graphy and evolution of the chlamytheres. The first suggests that from a common Kraglievichia origin North and South American Pleistocene chlamytheres had separate but parallel histories. If so, Holmesina would be the correct name for the later North American chlamythere. A second hypothesis suggests that Kraglievichia spread from South America into North America near the end of the Pliocene, after which it became extinct in South America. This would account for the absence of chlamythere records in the Early Pleistocene of South America. ( A dermal plate that was questionably attributed to the Chapadmalalan stage by Kraglievich [1934] actually came from a locality near the major deposit, and its horizon is unknown. Since each chlamythere shell pro- duces a high number of bony plates, they are usually abundant if present at all.) Later in the Pleistocene, then, Pampatherium spread back into South America. The third hypothesis suggests that chlamytheres were continuously distributed through tropical America during most of the Pleistocene. Thus the evolution of Pampatherium from Kraglievichia was a single continuous event that spread through populations from the Gulf Coastal region and mesoamerican parts of North America into the northern parts of South America. The absence of Early Pleistocene deposits in tropical South America makes it difficult to test this hypothesis. This last hy- pothesis, here favored, may be partly tested by careful comparisons of later Pleistocene chlamytheres. Dasypus bellus SIMPSON 1929 MATERIAL.-UF 16698: left nasal, right maxilla and mandible, right astragalus, and 51 isolated dermal plates from Haile XV A, UF 10449: right ulna from Santa Fe IA. The dentigerous specimens probably represent one individual, be- cause they articulate closely. Most of the plates also were found closely associated. The present material has been compared with other speci- mens Of Dasypus bellus and with its closest living relative, Das!/pus nouemcinctus. The only other known D. bellus specimens containing teeth are the excellent skeleton from Crankshaft Pit, Missouri ( UK 15544; Oesch 1967), and a fragmentary jaw from Melbourne ( Ray 1958 ). Com- parisons have been made with posteranial material from numerous Pleis- tocene sites in Florida. 1976 ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 143 A B FIGURE 11. Das!/Pus beUus, mandible, UF 16698, Haile XV A: (A) occlusal view; and (B) lateral view, X 1.45. The nasal bone is represented by the anterior portion only and is larger than, but morphologically similar to the nasal bone of D. no- uemcinctus. In discussing the dentition, teeth will be designated as follows: Ti will refer to the Erst upper tooth, T: the second upper tooth, etc. Simi- larly numbered subseripts will be used in referring to lower teeth. The maxilla is broken posteriorly and contains only T)-T: Except for size, the lateral portion of this specimen compares closely with the corresponding position of the Crankshaft Pit specimen. In the Haile XV A specimen T3 is transversely flattened, but all the rest are round and peg-like. The preceding two teeth of the Haile XV A Dasypus ( UF 16698) were probably also transversely Hattened, as this is the usual con- dition in D. bellus. In D. novemcinctus the Brst three teeth tend to be less Rattened laterally than in D. bellus. The remaining upper teeth were probably round and peg-like as is usually the case with T4-T: Talmage and Buchanan ( 1954), as well as others, have pointed out that tooth number and structure are variable in Dast/pus. In four D. novemcindus skulls examined, the total number of teeth varied from seven to nine. In UF 16698 ( Fig. 11), as in the Kansas mandible, two foramina are present-one large and one small. In UF 16698 the larger of the two foramina occurs between Tz and'Ta. The smaller of the two foramina is located 38 mm anterior to Ti. In UK 15544, the larger foramen is anterior to the smaller and is located directly beneath T". The smaller is located beneath Ta. The number of these mental foramina in D. nouemcinctus varies from one to four. The symphysis of the Haile XV A specimen is weak, as is characteristic of the genus. The mandible of UF 16698 lacks T#-Ts. The preceding teeth, Ti-T), are laterally flattened, but the succeeding teeth, L and T„are round in 144 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 3 TABLE 9.-DENTAL MEASUREMENTS ( IN MM ) OF Dasvpus beNUS. UF 16698 UK 15544 Haile XV A Crankshaft Pit Florida Missouri left right MAXILLARY Total distance, teeth 3-6 19.9 23.81 23.7 Anteroposterior distance, tooth 3.5 4.61 4.5 Transverse distance, tooth 3 2.1 4.61 4.5 Anteroposterior distance, tooth 4 4.4 Transverse distance, tooth 4 3.6 Anteroposterior distance, tooth 5 4.0 4.1 Transverse distance, tooth 5 4.2 5.1 Anteroposterior distance, tooth 6 4.2 5.22 4.5 Transverse distance, tooth 6 4.1 4.8 MANDIBLE Depth of ramus at tooth 1 6.8 12.4 Width of ramus at tooth 1 2.8 3.7 Depth of ramus at tooth 6 8.4 15.2 15.5 Width of ramus at tooth 6 4.8 8.2 7.8 Depth of ramus at tooth 8 7.2 13.5 12.2 Width of ramus at tooth 8 5.2 7.1 6.4 Length of tooth row 38.8 45.6 Anteroposterior distance, tooth 1 3.0 2.88 Transverse distance, tooth 1 1.3 1.83 Anteroposterior distance, tooth 2 3.3 2.2 2.1 Transverse distance, tooth 2 1.4 2.7 2.3 Anteroposterior distance, tooth 3 3.3 4.0 4.5 Transverse distance, tooth 3 2.2 3.2 3.2 Anteroposterior distance, tooth 4 4.81 4.7 4.6 Transverse distance, tooth 4 2.81 4.5 4.2 Anteroposterior distance, tooth 5 4.01 4.1 4.4 Transverse distance, tooth 5 3.32 4.5 4.5 Anteroposterior distance, tooth 6 4.5 4.3 4.5 Transverse distance, tooth 6 ' 3.7 4.7 4.7 Anteroposterior distance, tooth 7 4.6 4.6 4.2 Transverse distance, tooth 7 3.5 4.7 4.6 Anteroposterior distance, tooth 8 4.1 3.5 3.4 Transverse distance, tooth 8 4.2 3.5 3.4 1 Alveolar. 2 Alveolar, estimated. 3 Not completely erupted. cross-section. In this feature the Haile XV A specimen differs from the lower dentition of UK 15544, in which all the teeth are round. Except for smaller size and this tendency toward flattening of the anterior teeth, the mandibles and dentition of the Haile XV A specimen differ little from other Dast/pus material. Dental measurements appear in Table 9. In the ulna from Santa Fe I the, channel on the medial side is deep and persists to the proximal end of the element. This distinguishes it from the otherwise very similar ulna of Kraglieuichia ( p. 127). 1976 ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 145 The astragalus of the Haile XV A D. bellus, along with other speci- mens of D. bellus from Florida Pleistocene sites, differs little from Recent specimens of D. nouemanctus, except for size. A comparison of the astragali of Dasypus from various stages of the Pleistocene reveals a pro- gressive size increase similar to that of the chlamytheres. The dermal plates of UF 16698 do not differ from any other speci- mens of Dasypus, except for size. Martin ( 1974:41) measured samples of plates from several Florida sites and showed little overlap in plate size between D. bellus and D. novemcindus. His measurements indicate that the size of the plates does not follow the trend through the Pleisto- cene that is evident in other parts of the D. bellus skeleton. The Cole- man IIA plates are notably small and overlap slightly in size with those of D. novemcinctus. However, this is probably a function of the small sample size used; there are approximately 2500 plates of diverse sizes in a single armadillo carapace, and there is no satisfactory method for determining from which parts of the shell the sample may have come. Size trends based on plates thus should involve large samples in order to insure accuracy. The morphological differences between the Haile XV A material and later D. bellus material are very slight and presumably have a variational basis. Consequently, the Haile sample is referred to Das!/pus bellus. Das!/pus bellus has been considered a southern North American species of Rancholabrean age. However, Martin ( 1974) extended the temporal range of the species back into Irvingtonian time, and the pres- ent study places the earliest record in the Blancan. A question arises regarding the relationship between D. bellus and D. not)emeinctus. These animals are clearly more closely related to each other than to any other armadillos ( Auffenberg 1957), and it is uncertain whether D. nouemcinctus was derived from D. bellus or existed allo- patrically with that species during the Pleistocene, replacing it during the last few thousand years. If D. novemcinctus did evolve from D. bellus, it appears to have done so quite rapidly. In Miller's Cave ( Pat- ton 1963) a date of ca 8000 BP was given for the Travertine stratum bearing D. bellus remains, as compared to ca 3000 BP for the overlying Brown Clay deposits containing D. not)emcinctus. Thus, the change oc- curred in that area within a span of only 5000 years. Another possibility is that D. novemcinctus existed somewhere else during the Pleistocene and replaced D. bellus very late in the Pleistocene. If so, it would be expected earlier in some Central or South American Pleistocene deposits. To date, the only fossil record of D. nouemcinctus is from Miller's Cave, and the only Late Pleistocene armadillos from Mexico are referable to 146 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 3 D. bellus. Until more tropical fossil records are found, the problem must remain unresolved. In view of the long residence of Dasypus bellus in Florida, it seems surprising that no pre-Rancholabrean records are available elsewhere in North America. Possibly Dasypus was restricted to the Gulf Coastal region during most of the Pleistocene. Slaughter ( 1961) cited two ecological conditions that may have been necessary for this species: (1) winters no more severe than those occurring in North-central Texas today, and ( 2) rainfall more than 20 inches per year. Thus the distri- bution of such climatic conditions may have restricted D. bellus to Flor- ida and the Gulf Coastal Plain during most of the Pleistocene. ORDER LAGOMORPHA FAMILY LEPORIDAE Syluilagus sp. GRAY 1867 MATERIAL.-UF 17561: 2 Ma; UF 17562: M3; UF 17563: lumbar vertebra; UF 17564: 3. innominates; UF 17565: 2 femora; UF 17566: 1 tibia; UF 17567: 2 metatarsals. Hypolagus, Alurobgus, and Syluilagus are the smaller lagomorphs common in North American Blancan deposits. Unfortunately, Ma and M~, the only teeth preserved in the Haile XV A fauna, are not diagnostic ( Dawson 1958, Downey 1970). When compared with the Hypolagus innominates ( Dawson 1958), the Haile XV A specimens possess a less elevated ilial crest, which is probably a more advanced condition. The remainder of the Haile XV A material resembles Sylvilagus in size and morphology, but one cannot completely discount reference of this ma- terial to Alurolagus C Downey 1968, 1970) from the Benson and Curtis Ranch Faunas of Arizona. A species determination is not possible from the available material. Two living species of Sylvilagus occur widely in Florida, S. palustris and S. Boridanus. The genus is presently distributed throughout North America, and also ranges into South America where it is represented by S. /loridanus. and S. brasiliensis (Hall and Kelson 1959). Previously, the principal Blancan records of Syluilagus were from the Broadwater fauna ( Barbour and Schultz 1937) and the Sand Draw fauna ( Skinner and Hibbard 1972) of Nebraska. Numerous Irvingtonian and Rancholabrean records of this genus are also known throughout North America. 1976 ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 147 ORDER RODENTIA FAMILY SCIURIDAE Cryptopterus webbi new speciesi FIGURES 12-13 TYPE.-UF 12353: partial right mandible with third molar. TypE LoCAL]TY.-Haile XV A, T95, R17E, Section 25, NW 1/4 of SW 1/4, Alachua County, Florida. AGE.-Blancan DIAGNOSIS.-Large sciuropterine; low-crowned, subovate Ma.not tapered posteri- orly; protoconid connected to weak anteroconid; large anterolabial cingulum as in Cryptopterus tobieni; Rat trigonid basin bearing "chaos" of low oblique ridges; strong metaconid and metaconid-metastylid crest; distinct entoconid; faint hypolophid; broad posterolophid bearing vestigial hypoconulid; posterolabial flexid absent. DESCRIPTION.-The unique flying-squirrel specimen from Haile XV A consists of the posterior part of the right mandible bearing a moderately- worn third molar. Much of the ascending ramus, especially the condyle, is well preserved. The distance between the last molar and the condyle is about 15 mm. A large dental foramen lies 6.1 mm posterior to the last molar. The angular region is massive, concave lingually, and descends to a depth of more than 12 mm below the alveolar level, at which point it is broken. The coronoid region is also broken. The third molar measures 4.91 mm in length, 3.70 mm maximum width across the metaconid and protoconid, and 3.29 mm posterior width across the entoconid and hypoconid. This is by far the largest flying squirrel in the New World, rivaling the extinct Eurasian species of Cryptopterus and Petauria and some living Asian species of Petaurista. The Haile XV A molar has a subovate outline and is only slightly tapered toward the posterior end. This readily distinguishes it from specimens of Miopetaurista and Pliopetaurista, in which the posterior end of the tooth tends to be long and narrow. The crown is low, the protoconid rises only 1.1 mm above the crown base; it thus contrasts with most species of Petaurista and especially with Eupetaurus ( McKenna 1962). On the labial side of the molar the strong protoconid and hypoconid are separated by a moderately well-developed mesoconid. The meso- conid is more robust than in Pliopetaurista and does not produce a labial spur, such as Sulimski ( 1964 ) noted in the specimens from Wdze, Poland, The protoconid is joined to a small anteroconid by a short, anteriorly curved crest, and thence weakly to the prominent metaconid. The weak anteroconid does not extend toward the labial wall of the molar as in 1 Named for Professor S. David Webb in honor of his contributions to the vertebrate paleontol- ogy of Florida. 148 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 3 FIGURE 12. Cryptopterus webbi type, right mandible, anterior end upwards, UF 12343, Haile XV A; X 5.40. 1976 ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 149 4 mm 1 1 FIGURE 13. Cryptopterus webbi, right M3, anterior end upwards, UF 12353, Haile XV A. most sciuropterines. Instead, a large cingulum occupies the anterolabial corner of the molar, recalling the arrangement found in the third lower molar of Cryptopterus tobieni ( Mein 1970: fig. 39 ). The anterolingual part of the third molar is greatly expanded as in most sciuropterines. It supports a very large prominent metaconid from which a posterior crest connects broadly back to a low metastylid ( or mesostylid according to Mein's 1970 terminology), and a labial crest reaches the anteroconid. From the robust metaconid, the enamel surface slopes gradually downward into a very shallow trigonid valley. In its center are several low sinuous ridges (lophulids) that trend in a postero- lingual direction. Such lophulid complexes are characteristic of many sciuropterines; the homologous but higher ridges in Petaurista xanthotis were appropriately termed "metaconid-metastylid chaos" by McKenna ( 1962). These lophulids presumably provide drainage canals for the juices produced when pulpy foods are crushed in the trigonid basin. Similar features for similar purposes are found in such unrelated mam- mals as Ailuropoda and Pongo. A distinct entostylid occupies the midlingual part of the molar from Haile XV A. A weak ridge, the hypolophid, extends labially from the entostylid, but is soon lost in the broad continuity between the trigonid and talonid "valleys" in the center of the tooth. A minor depression, 150 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 3 separating the entoconid from the strong posterolophid, encircles the posterior margin of the tooth. A thickening of the enamel in the middle of the posterolophid evidently represents a vestigial hypoconulid. Al- though faint, this structure is notable inasmuch as Mein ( 1970:22) characterized all other Cryptopterus as having a smooth posterolophid without any hypoconulid. There is no sign of a posterolabial inRection in the molar from Haile XV A. DIscussION.-Although the affinities of the Haile XV A flying squirrel might reasonably be sought among other North American sciuropterines, such comparisons prove unsatisfactory. The North American record consists of Glaucomys of Pleistocene and Recent age and a late Miocene sample of two small species from the Cuyama Valley of California de- scribed as Sciuropterus by James ( 1963), but assigned to Cryptopterus by Mein ( 1970). James ( 1963) suggested that the North American Mio- cene species might have been ancestral to Glaucomys, or at least that no closer ancestry is known. Mein ( 1970) was more skeptical of a direct relationship. In either event the Haile XV A specimen, far from bridging that gap, differs in a number of basic features from what would be ex- pected of a Latest Pliocene Glaucomys ancestor. Whereas Recent Glaucomys is even smaller than the Cuyama Miocene fossils, the Haile XV A specimen is much larger. Neither an anteroconid nor an antero- lingual cingulum ( of any size) occurs in either the California Miocene specimens or in Glaucomys, yet both features are found in the Haile XV A specimen. The protolophid directly connects the protoconid to the metaconid in both the Miocene Cuyama specimens and in Recent Glaucomys specimens, whereas it is interrupted by an anteroconid in the Haile XV A specimen, The talonid valley is not crenulated in the Haile XV A specimen, and the trigonid lophulids are heavier than in other North American flying squirrels. Turning to the Old World fauna, we find a much greater variety of b6th fossil and Recent genera with which to make comparisons. These include several taxa whose size equals that of the Haile XV A specimen, and some that exceed it. It seems evident that the phylogenetic de- velopment of the flying squirrels was centered in Eurasia, and it is fortu- nate that a relatively complete fossil record of the group is known from Europe. Mein ( 1970), who recently published the most complete phylogenetic review of the sciuropterine squirrels, recognized three broad groups of fossil and Recent genera based on lophule development and enamel crenulation. His arrangement is as follows: GROUP I ( Enamel smooth; lophules absent) Recent Genera: Glaucomys, Eoglaucomys, Iomys. Fossil Genera: Cryptopterus, Petauria. 1976 ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 151 GROUP II ( Enamel smooth; lophules present) Recent Genera: Pteromys, Trogopterus, Ptero- myscus, Belomys, Aeretes, Petaurista, Eupetaurus. Fossil Genera: Miopetaurista, Forsythia, Pliope- taurista. GROUP III ( Enamel crenulated; lophules rare or absent) Recent Genera: Petinomys, Hylopetes, Aeromys. Fossil Genera: Blackia, Pliopetes. Of these diverse genera, European fossil specimens of Cryptopterus and Petauria bear the closest resemblance to the specimen from Haile XV A. As in those genera, the enamel in the Haile specimen is not finely crenulated. The pattern of cuspids and lophids agrees closely with that in some species of Cryptopterus. Unfortunately, the question of whether the lophules in the upper cheek teeth were elaborately de- veloped cannot be directly answered by the Florida specimen, but if the detailed resemblances of Cr~/ptopterus lower molars are correct, loph- ules would be absent from the upper molars just as they are from the lowers. Several features of the lower molar from Haile XV A rule out its relationship to Group II genera. The nearly ovate shape of the last molar is a reliable distinction from molars of the extinct taxa Miope- taurista and Pliopetaurista. These taxa also bear a distinct posterolabial fiexid that is barely ( if at all) recognizable in the Florida molar. As Sulimski ( 1964) noted in his material from Poland, the mesoconid ex- tends labially as a spur in Pliopetaurista, in contrast to the round meso- style in the Florida specimen. Finally, the metastylid ( mesostylid of Mein 1970) is very prominent in Miopetaurista and Pliopetaurista, whereas in the Florida specimen it is largely submerged by the broad crest connecting it to the metaconid. Each of these features indicates a closer relationship of the Haile XV A specimen with Group I taxa. The flying squirrel tooth from Haile XV A most closely resembles the third lower molars of Cryptopterus. It also resembles homologous teeth of Petauria helleri from Early Pleistocene red earth fissure Bllings in the Solenhofen Limestone near Schernfeld, Bavaria ( Dehm 1962), but that genus completely lacks a mesoconid on its lower molars, and the hypo- lophid is more strongly developed than in the Florida specimen. This general resemblance of the Florida specimen to the Bavarian specimen led to its preliminary assignment to "Petauria sp." in Webb ( 1974a). It is now evident, however, that the Haile XV A specimen is perhaps more accurately recognized as a new species of the genus. Cryptopterus. 152 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 3 When one compares Cryptopterus webbi with other species of that genus, the closest resemblance is to those from the very Late Pliocene. In view of the Blancan age of the Florida species, this is what one might expect. The youngest European species of Crytopterus is C. tobieni from the lignites of Wolfersheim-Wetterau, West Germany (late Perpi- gnan equivalent). The slightly older species, C. thaleri, resembles C. tobieni in size and many other features but is not presently represented by any lower molars. Special features shared by lower third molars of C. tobieni and C. webbi are the broad anterolabial cingulum, the weak, anteriorly-directed anteroconid, the short protolophid, the mestastylid little distinct from the metaconid ridge, the distinct entoconid, and the weakly developed hypolophid. C. webbi is slightly smaller than C. tobieni and perhaps C. thaleri, and it differs from all other known Crt/ptopterus specimens in the presence of a vestigial hypoconulid. Lower third molars of the Miocene species of Cryptopterus differ from those of C. webbi and C. tobieni in having a narrow anterolabial cingulum, a strong labially directed anteroconid, a complete protolophid, a more distinct metastylid, and the entoconid attached both to the posterolophid and to a strong hypolophid. An exception to this is the presence of a distinct entoconid and a weak hypolophid in Gaillard's ( 1899) original figured specimen from La Grive, which is now referred to C. gaillardi ( Mein, 1970). Evidently this late Miocene variant embodied some of the progressive features that were developed more definitely during the Pliocene in C. tobieni and C. webbi. Cryptopterus mathewsi from Late Miocene de- posits in the Cuyama Valley of California also differs in some of these same features from C. webbi and C. tobieni; however, the lower third molar is poorly known ( James 1963). No lower dentition has been re- covered in C. uphami material from the same area. The anteroconid is essentially absent, as is the anterior cingulum. I am not yet convinced, in fact, that the Cuyama species are referable to Cryptopterus. The itinerary by which Cryptopterus reached Florida in the Blancan must remain in doubt, but the general pattern of distribution may be deduced. Since the nearest afBnities of C. webbi are with C. tobieni ( Late Pliocene of Europe), it seems probable that the genus reached North America from the Old World during that epoch. We may safely say that the date of arrival in North America was within late Hemphil- lian or Blancan time. Furthermore, the only likely dispersal route was via the Bering Land Bridge between Asia and North America. Flying squirrels do not truly «fly," and their present distribution patterns do not indicate great vagility. This New World occurrence of Cryptopterus indicates continuity of subtropical rain forest from Eurasia into eastern North America during the Late Pliocene. 1976 ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 153 TABLE 10.-MEASUREMENTS ( IN MM) OF THE FE:MORA OF FOSSIL AND RECENT CastOr canadensis. C. canadensis C. canadensis UF 17489 Recent Haile XV A N X OR Width, distal end 38.7 6 36.6 34.1-38.4 Anteroposterior depth, distal end 29.2 6 29.7 28.5-32.5 Width, external condyle 13.5 5 14.0 12.2-15.5 Width, internal condyle 12.5 5 12.1 11.6-12.5 Width of shaft at third trochanter 25.2 6 29.1 27.0-30.5 Anteroposterior diameter of shaft at third trochanter 13.9 6 11.5 10.2-13.5 FAMILY CASTORIDAE Castor canadensis LINNAErs 1758 MATERIAL.-UF 17489: left femur. This element, which lacks the proximal end and represents a young individual, shows no morphological differences from the living species. A comparison of measurements ( Table 10) with examples of young, Recent Castor canadensis shows no significant size differences. Other Blancan records of Ca#or in North America are from the San Joaquin locality in California ( Kellogg 1911, Stirton 1935) and the Hagerman fauna of Idaho ( Stirton 1935, Zakrzewski 1969). FAMILY CRICETIDAE Sigmodon medius GIDLEY 1922 MATERIAL.-UF 17489: left Mi; UF 12341: right Ml; UF 12337: left MZ; UF 12339-12340: right M3; UF 12334, UF 12338: left Ml; UF 12336: left MZ; UF 12342: right Ml and Ma (both unworn). This species characteristically possesses only two or three roots on the Ml· If accessory roots are present, they are centrally located and are very small, peg-like structures. The labial root is always better devel- oped than the lingual. The two specimens from Haile XV A ( UF 12338) exhibit these characteristic features. Sigmodon medius is nearly identical morphologically with a closely allied species, Sigmodon minor. According to Martin ( 1974), the only difference between the two species, other than size, is that the reentrant folds are deeper and narrower in S. minor than in S. medius. This char- acter is not obvious in the relevant Haile XV A specimens because they show signiBcant wear. Discrimination between these two species must 154 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 3 TABLE 11.-DENTAL MEASUREMENTS (IN MM) OF Sigmodon medius AND Sigmodon minor.1 S. medius S. minor N X OR N X OR Length Ml 2 2.25 2.18-2.32 45 1.89 1.72-2.19 Width Ml 2 1.24 1.23-1.25 48 1.31 1.17-1.48 Length M2 1 1.49 49 1.40 1.22-1.62 Width M2 1 1.53 46 1.36 1.20-1.50 1 Re-calculated from Martin ( 1970). depend on size alone. The larger measurements of the Haile XV A material ( Table 11) indicate that it belongs to S. medius. Sigmodon medius is one of the most common small rodents in the Blancan of North America; its occurrences include the Benson local fauna of Arizona ( Gidley 1922), the Vallecito fauna of California ( Downs and White 1968), the Hudspeth and Red Light local faunas of Texas ( Strain 1966, Akerston 1970), and the Sand Draw fauna of Nebraska ( Skinner and Hibbard 1972). ORDER CARNIVORA FAMILY CANIDAE GRAY 1821 MATERIAL.-UF 17492-17493: 2 tibiae. Two poorly preserved tibiae indicate the probable presence of a small canid. They are about the size of a fox, but positive identification is not possible as only the much worn proximal ends are preserved. FAMILY MUSTELIDAE Satherium sp. MATERIAL.-UF 17487: humerus; UF 17491: right metatarsal II; UF 17490: right metatarsal III; UF 17494: medial phalanx. The humerus and podial elements closely resemble those of both Pteronura, the living giant, flat-tailed otter of South America, and Sather- ium, the extinct giant otter, from various Blancan sites in North America. Despite the numerous close resemblances between these two genera and th6 probable ancestral relationship of Satherium to Pteronura, I follow Gazin ( 1934) and Bjork ( 1970) in recognizing the extinct genus as dis- tinct. These two giant genera are surely more closely related to one another than to Lutra. The humerus from Haile XV A ( Fig. 14) is more robust than are 1976 ROBERTSON: HAILE XV A LATEST PLIO(JENE MAMMALS 155 4 % FIGURE 14. Satherium sp., humerus, anterior view, UF 17487, Haile XV A; X 1.95. 156 B U LLE T IN FLO R ID A S TATE M U S E U M Vol. 20, No. 3 TABLE 12.-MEASUREMENTS ( IN MM) OF THE HUMERUS OF Satherium AND R~CENT Pteronum AND Luton. Satherium Pteronura Hagerman, Id. Haile XV A brasitiensis Lutra canadensis USNM 23266 UF 17487 Recent, South America Recent, North America N X OR N X OR Greatest width, distal end 33.4 33.0 3 36.3 33.5-40.1 7 25.1 24.1-27.2 Greatest width, distal condyle ( measured from anterior side) 21.0 21.0 3 23.3 21.5-26.5 7 15.1 11.8-17.3 Length of entepicondylar foramen 5.5 5.8 3 5.9 5.4-6.5 6 4.4 3.7- 4.7 Width of entepicondylar foramen 3.3 3.9 3 4.2 4.0-4.4 6 2.0 1.5- 2.3 Greatest anteroposterior thickness of shaft 13.5 17.4 3 16.1 14.7-18.5 7 13.1 11.3-14.8 Transverse width at same location 9.9 10.3 3 10.2 9.5-11.4 7 7.9 7.5-8.8 Greatest anteroposterior thickness of distal condyle 15.4 16.1 3 14.6 12.7-16.2 7 11.2 10.2-12.4 1976 ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 157 TABLE 13,-MEASUREMENTS ( IN MM ) OF THE AErRAGALI OF Smilodon gracilis FROM HAILE XV A AND PORT KENNEDY.1 Haile XV A Port Kennedy UF 17496 Length 41.6 48 Width 45.2 38 Width of trochlea 28.7 27 Vertical diameter of head 18.7 17 Transverse:diameter of head 26.6 25 External elevation of trochlea 21.1 22 1 Measurements after Cope ( 1899). humeri of Satherium from the Hagerman local fauna of Idaho ( Bjork 1970). In this respect the Haile specimen agrees more closely with sped- mens of living South American Pteronura ( Table 12). In the dimen- sions of the metatarsals and in most other respects, however, the three samples are remarkably homogeneous. The distal width of the Haile humerus likely would be greater were it not for the pathological condi- tion of the medial side; the bone has turned distad and narrowed and is associated with much osteoporotic tissue. This anomaly does not seem to have affected either the articular surface nor the entepicondylar fora- men, which are located laterally and proximally, respectively, to the affected area. The most impressive features that distinguish the Haile XV A specimen from the Hagerman skeleton ( USNM 23266) are the much greater medial dimensions of the shaft and the deeper and wider ectepicondylar ridge. In all respects, however, there are strong similari- ties among the Haile sample and the genera Satherium and Pteronura. FAMILY FELIDAE Smilodon gracilis COPE 1880 MATERIAL.-UF 17496: right tibia ( distal end) and right astragalus; UF 17498: right metacarpal II. The Haile material presumably is all from a single medium-sized animal. All three bony elements were found together, the tibia and astragalus articulate well, and the material is similarly preserved. The material is referred to the genus Smilodon on the basis of two characters ( see Merriam and Stock 1932 ) : (1) the presence of the astragalar fora- men; and (2) the medial facet for the calcaneum and the facet for the navicular are merged. The Haile XV A sabercat was very small and was similar in size to Smilodon gracilis. A comparison of the measurements of the astragalus from Haile XV A with those of the astragalus of S. gracilis from the Port Kennedy Bone Bed ( Cope 1899) shows a close correlation in size (Table 13). Inasmuch as S. gracilis is the only small species of Smilodon 158 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 3 known from North America and is of relatively early Pleistocene age, the Haile XV A specimen is referred to that species. Two species of small sabercats are known from South America. Smilo£lon cruciens is known only from a single mandible ( Ameghino 1904). Smilodontidion riggsi ( Kraglievich 1948), which is slightly larger than the Haile XV A form, is present in the Chapadmalal fauna and is known only from posteranial material.1 The astragalus of this specimen ~ is preserved but in poor condition, and the critical characters of the facets cannot be verified from the figures. Although other species occur both in the Chapadmalal fauna and the Haile XV A fauna, it does not follow that the two cats are necessarily related. Nonetheless, the overall faunal resemblance, together with the similarity in morphology of the South American samples, suggest that possible synonymy of the Chapadmalalan form with Smilodon gracilis ought to be considered. ORDER PROBOSCIDEA FAMILY GOMPHOTHERIIDAE CABRERA 1929 MATERIAL.-UF 17464: a small portion of a molar and two small pieces of ivory. The complicated nature of the cheek tooth clearly indicates that this specimen is a gomphotheriid. Further identification is not possible be- cause of the inadequate material available. ORDER PERISSODACTYLA FAMILY EQUIDAE Nannippus phlegon (HAY 1899) MATERIAL.-UF 17484-17485: both upper cheek teeth; UF 17547: metapodial; UF 17548: proximal phalanx, UF 17549: four ungual phalanges. A nearly unworn upper cheek tooth 51 mm high indicates that this form was strongly hypsodont. The other well-worn upper tooth ( Fig. 15) and the posteranial material agree in all respects with the typical Blancan species, N. phlegon. Measurements are provided in Table 14. Equus ( Dolichohippus) simplicidens CopE 1892 MATERIAL.-UF 16909: partial skull; UF 17556: partial maxilla with deciduous molariform tooth; UF 10894: 15 cervical vertebrae; UF 10877: 37 thoracic verte- brae; UF 10895: 13 lumbar vertebrae; UF 10896: 2 sacral vertebrae; UF 10898: 3 humeri ( 2 left, 1 right); UF 10919: 1 right radius,· UF 10910: 2 left pelves; UF 10911: 4 femora (2 left, 2 right); UF 10915: 2 right metatarsals; UF 10921: 9 splints; UF 10916: 4 cuboids ( 2 left, 2 right); UF 10917: 3 calcanea; UF 10920: 1 Churcher's ( 1967) cautionary note suggesting that the specimen may have leaked down from younger debosits is noted. 1976 ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 159 2 astragali; UF 10917: 4 proximal phalanges; UF 10913: 4 distal sesamoids; UF 10914: 7 ungual proximals. The Haile XV A skull ( UF 10909) is in poor condition, and not enough of the fragments are present to allow complete reconstruction. FIGURE 15. Nannippus phlegon, left upper molar, occlusal view, UF 17484, Haile XV A; X 6.15. 160 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 3 TABLE 14.-MEASUREMENTS (IN MM ) OF Nannippus phlegon CRANIAL AND POST- CRANIAL MATERIAL FROM HAILE XV A. UPPER CHEEK TOOTH UF 17484 UF 17485 Anteroposterior distance 17.5 19.6 Transverse distance 20.4 Crown height 51.0 Metapodial Transverse width, distal end 25.7 Anteroposterior width, distal end 24.0 Proximal phalanx Width, proximal end , 24.4 Width, distal end 20.6 UNGUAL PHALANX UF 17549a UF 17549b UF 17549c Total length 43.4 43.6 43.1 Width, distal articular surface 22.7 23.9 22.8 Both left and right molariform tooth rows, however, are complete ( Fig. 16) and a portion of the premaxilla has been rebuilt. A small portion of the occipital region is preserved, but the rest of the skull consists of isolated fragments. This skull is referred to the subgenus Equus ( Dolichohippus) on the basis of the well developed parastyles and metastyles, the dilation of the fossettes, and the more rounded nature of the protocones. Assignment of extinct North American species, formerly referred to Pbsippus, to the same subgenus as the living Grevy's Zebra follows Skinner and Hibbard (1972). In dental morphology, UF 10909 resembles both E. ( D.) shoshonensis and E. ( D.) simplicidens. Gazin ( 1936) stated that the ranges of mea- surements of these two species may overlap, although E. ( D.) simplici- dens·tends to be smaller. Cheek teeth from Haile XV A ( Table 15) are smaller than any measured by Gazin ( 1936). They are close to the size of a specimen of E. ( D.) simplicidens described by Hibbard ( 1941) from the Blancan Rexroad fauna of Kansas, and they fall at the lower end of the range given by Howe ( 1970) for a large sample from the Broadwater local fauna of Nebraska. The postcranial material is also small ( Table 16), and falls within the range of measurements given by Gazin ( 1936) for a specimen of simplicidens. Equus ( Hemionus) cf. calobatus TROXELL MATERIAL.-UF 17750: upper cheek tooth; UF 17483: lower molar; UF 17482: Ma; UF 17551: left humerus; UF 17552: astragalus; UF 17553: proximal phalanx; UF 17554: medial phalanx; UF 17555: ungual phalanx. 1976 ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 161 r /2 'f 6 t FIGURE 16. Equus ( Dolichohippus) simpticidens, upper dentition, occlusal view, anterior end upwards, UF 10909, Haile XV A; X 0.80. 162 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 3 TABLE 15.-MEASUREMENTS. (IN MM) OF THE UPPER DENTrnON OF Equus (Doli- chohippus ) simplicidens FRoM HAILE XV A.1 Left Right Tooth Length Width Length Width Tooth row2 158 155 pl 12.2 6.4 pz 32.9 24.7 32.8 24.6 ps 24.1 26.2 24.0 25.2 pd 23,3 24.8 21.9 24.3 22.3 23.8 23.8 25.5 M~ 22.0 22.6 22.4 23.9 Ma 28.8 23.0 33.2 35.3 1 Measurements taken at grinding surface; width measurements do not include cement. 2 Not including P: TABLE 16.-MEASUREMENTS (IN MM) OF Equus ( Dolichohippus ) simplicidens PosT- CRANIAL MATERIAL FROM HAILE XV A. N X OR HUMERUS Transverse width, distal condyle 3 70.1 69.7- 70.5 Anteroposterior width, distal end 3 80.7 77.0- 84.0 RADIUS Total length 1 316.0 Anteroposterior width, proximal end 1 77.6 Transverse width, proximal end 1 36.3 Anteroposterior width, distal condyle 1 37.4 Transverse width, distal condyle 1 57.9 FEMUR Total length 1 368.5 Transverse width, proximal end 1 114.0 Anteroposterior width, proximal end 1 88.5 Transverse width, distal condyle 4 87.2 84.6- 89.7 Anteroposterior width, distal end 4 116.8 114.0-118.0 METATARSAL Anteroposterior width, proximal end 2 41.2 39.6- 42.8 Transverse width, proximal end 2 48.5 48.2- 48.8 CALCANEUM Total length 3 107.7 103.0-110.0 ASTRAGALUS Total length 2 58.0 57.0- 58.9 Lateral width 2 58.8 57.7- 59.8 PROXIMAL PHALANX Total length 3 78.5 76.5- 81.2 Width, proximal articular surface 2 43.7 42.8- 44.5 Width, distal end 4 36.4 33.7- 38.5 MEDIAL PHALANX Total length 6 45.5 44.1- 47.4 Width, proximal articular surface 6 40.7 39.1- 43.2 Width, distal end 6 40.9 38.8- 42.8 UNGUAL PHALANX Total length 6 54.4 50.6- 58.8 Width, proximal articular surface 6 38.5 34.1- 43.3 1976 ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 163 The three teeth listed above are morphologically different from those of Equus ( Dolichohippus ). The fossettes of the upper tooth are much more complicated, and the lower teeth ate quite distinct from those of E. (Dolichohippus) figured by Gazin ( 1936), Hibbard ( 1941), McGrew ( 1944), and Skinner and Hibbard ( 1972). In all the lower molars fig- ured by the above authors, the median valley divides the isthmus, as pointed out by Skinner and Hibbard ( 1972); in the two teeth dis- cussed here, the median valley approaches the isthmus but does not divide it. In this respect, the Haile XV A specimens resemble Equus fraternus, a later Pleistocene ass. Several postcranial elements in the Haile XV A fauna are probably also referable to E. ( Hemionus) calobatus. They are much smaller than specimens attributed to E. ( D.) simplicidens. Troxell's type and re- ferred material from the Arkalon ( Hibbard 1953) and the Cragin Quarry local faunas ( Schultz 1969) form the bases for comparison. Hibbard ( 1956) also reported the medial phalanx of an ass in the Meade Gravel Member in Meade County, Kansas, with which the Haile XV A medial phalanx agrees very closely ( Table 17 ). FAMILY TAPIRIDAE Tapirus sp. MATERIAL.-UF 17468: partial crown of a cheek tooth. Since only one genus of Tapir is known from post-Hemphillian de- posits of North America, it seems reasonable to ascribe this specimen to Tapirus. The material serves only to record the presence of this genus in the fauna. This genus is also recorded in Santa Fe I, another Blancan locality in Florida ( Webb 1974a). ORDER ARTIODACTYLA FAMILY TAYASSUIDAE Mylohyus #oridanus KINsEY 1974 HOLOTYPE MATERIAL.-UF 18002: left and right mandibular rami and upper left canine. This species was recently described by Kinsey ( 1974), based upon the material from Haile XV A. It represents the oldest ( and only Blancan) record for this genus. The material is mfigured ( Figs. 17, 18) and the diagnosis is included here in order to complete the study of the fauna. Measurements of the holotype are presented in Table 18. 164 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 3 TABLE 17.-MEASUREMENTS ( IN MM ) OF Equus ( Hemionus ) cf. calobatus CRANIAL AND POSTCRANIAL MATERIAL FROM HAILE XV A. UPPEh CHEEK TOOTH Anteroposterior distance 25.4 Transverse distance 23.4 M, UF 174822 UF 17482b Anteroposterior distance 28.7 Transverse distance 11.2 10.7 HUMERUS Total length 255.0 Transverse width, anterior end 88.01 Lateral width, distal condyles 88.1 Anteroposterior width, distal condyles 46.9 ASTRAGALUS Total length 50.4 Greatest width 50.0 PROXIMAL PHALANX Total length 60.6 Width, proximal end 34.8 Width, distal end 30.5 MEDIAL PHALANX Total length 40.0 Width, proximal articular surface 38.1 Width, distal end 34.8 UNCUAL PHALANX Total length 45.7 Width, proximal articular surface 33.1 1 Estimated. DIAGNOSIS.-Close to Mi/lohyus nasutus in size and elongation of the snout. Larger and more robust than Mylohgus fossilis and lacking the extreme constriction of the symphysis that is diagnostic of that species ( Lundelius 1960). Pre- and post-canine diastema longer than in other species of MI#ohyus. Symphysis much more slender than in Prosthenops. Coronoid process triangular with straight ( not convex) anterior margin. Total length of the Pz-Ma series less than in other Mylohyus, both absolutely and relative to the overall length of the jaw. M~ and Mz relatively shorter than in other Mylohyus, except M. exoticus ( USNM 8876). Anteroposterior diameter of C greater than in all other Myloh!/us, both absolutely and relative to the cheek teeth series. Width of C greater relative to width of P~ and Mi than in other Mylohyus ( Kinsey 1974). DISCUSSION.-Interestingly, Mylohyus is. exceedingly rare in most Pleis- tocene faunas of Florida. For example, a large Late Blancan collection from Santa Fe I contains only one specimen of this genus, whereas the 1976 ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 165 45 4. FIGuRE 17. Mylohyus foridanus type, right mandibular ramus, lateral view, UF 18002, Haile XV A; X 0.95. /.. )-0 1 FIG~E 18. Myloh!/us foridanus type, mandibular symphysis, dorsal view, UF 18002, Haile XV A; X 1.0. other peccary, Plat!/genus, is very abundant. Similarly, in the Coleman IIA fauna of Irvingtonian age, Platygonus outnumbers Myloht/us eleven to one. The presence of Mylohgus at Haile XV A suggests an ecological distinction, as will be discussed further in the paleoecology section ( be- low). 1976 ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 167 TABLE 18.-MEASUREMENTS ( IN MM) OF THE TYPE DEN·rmoN OF Mulohvus #Ori- danus. 1 Left Right Length, upper canine 15.1 Width, upper canine 11.1 Length, lower canine 14.0 Width, lower canine 10.6 Length, P2 8.5 Anterior width, Pz 4.9 Posterior Width, Pz 5.1 Length, Ps 12.0 Anterior width, Pa 9.0 Posterior width, Pa 9.1 Length, P* 10.6 13.0 Anterior width, P* Posterior width, 1?4 12.6 Length, Ml 13.2 13.5 Anterior width, Ml 12.6 12.8 Posterior width, Ml 12.8 12.7 Length, M2 15.2 15.5 Anterior width, M2 14.0 14.1 Posterior width, M2 13.9 14.1 Length, M3 21.1 21.0 Anterior width, Ma 13.1 13.1 Posterior width, M3 11.8 11.9 Length, premolars 33.1 Length, molars 49.9 50.3 Length, molar-premolar series 83.0 Post-canine diastema 80.0 Pre-canine diastema 14.0 13.0 Depth of jaw at Ml 41.2 41.0 Thickness of jaw at Mi 21.2 21.1 Width, jaw at canines 41.2 Width between canine alveoli 17.6 Least width, symphysis 26.4 Length, articular condyles to anterior end of symphysis 273.0 1 Measurements after Kinsey (1974). FAMILY CAMELIDAE Hemiauchet:ia cf. macrocephala ( COPE 1893 ) MATERIAL.-UF 10900: partial skull; UF 10894: 2 cervical vertebrae; UF 10899: phalanges ( 3 medial, 1 ungual). The Haile XV A skull ( UF 10900) ( from a juvenile individual) con- sists of the palate with upper dentition ( Fig. 19) and part of the cranium. The cranial fragment includes parts of the parietal, left squamosal, left exoccipital, and supraoccipital. This portion of the skull, particularly the external auditory meatus and the subsquamosal foramina, agrees very closely with the McKittrick material of this species Bgured by Stock 168 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 3 FIGURE 19. Hemiauchenia cf. macrocephaa, upper dentition, anterior end upwards, UF 10900, Haile XV A; X 0.93. ( 1928 ). The left DP: is present, as are the left and right DPs and DP,; left and right PJ and Pt are present but unerupted. The complete right molar series is present, while the left is represented only by the Ml. The Haile XV A specimen is a small 11ama referable to the genus formerly called Tanupolama, which is now synonymized with Hemi- auchenia ( Webb 1974b). H. macrocephala is the smallest of the North American species of Hemiauchenia and has a relatively wider P. than H, blancoensis. The Haile XV A specimen lacks the critical P4, but the 1976 ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 169 TABLE 19.-MEASUREMENTS ( IN MM ) OF THE UPPER DENTITION OF Hemiauchenia cf macrocephola FROM HAILE XV A. Left Right Length, Dpz 10.8 Width, DP2 5,8 Length, DPa 19.7 19.2 Anterior width, DPa 11.51 12.4 Posterior width, DPS 17.3 Length, DP# 19.6 18.8 Anterior width, DP# 17.7 17.8 Posterior width, DP# 19.2 19.1 Length, Pa 16.52 Width, Pa 9.9 Length, Pd 18.82 Width, P4 15.2 Length, Mi 26.2 25.7 Anterior width, Mi 20.2 21.2 Posterior width, Mi 19.8 20.4 Length, M2 22.7 Anterior width, M2 21.2 Posterior width, Ma 19.7 Length, Ma 26.6 Ariterior width, Ma . 14.2 Posterior width, M3 19.1 Length of deciduous premolar series 50.1 Length of molar series 75.0 Length of tooth row functional at time of demise 1251 Length of permanent tooth row 1101 Palatal width at DPB 34.5 Palatal width at DP* 37.5 Palatal width at Ml 47.5 i Estimated. 2.Unerubted. specimen is quite small ( Table 19) and tentative species allocation is based on size. The occurrence of H. macrocephala in the Haile XV A fauna marks its earliest record anywhere. A long continuous history for this species in Florida is indicated by its presence in the early Irvingtonian Inglis IA fauna ( Webb 1974b). Hemiauchenia was widespread in the Blancan and is known in most North American faunas of this age, as well as in the Chapadmalalan of Argentina ( Kraglievich 1946). FAMILY CERVIDAE Odocoileus UirginianUS ZIMMERMAN MATERIAL.-UF 10885: 6 partial upper dentitions; UF 17481: right mandible with Pz, DP*, M2• Ms; UF 17478: left P,; UF 17477: right MB; UF 10883: 19 TABLE 20.-MEASUREMENTS ( IN MM) OF POSTCRANIAL ELEMENTS oF Odocoileus Cirginianus FROM SEVERAL FLORIDA PLEISTOCENE LOCALITIES. 170 B U LLE TIN FLO R ID A S TATE M U S E U M Vol. 20, N o. 3 Haile XV A Santa Fe I Santa Fe 4A N X OR N X OR N X OR ~ HUMERUS Lateral width, distal condyles 3 31.0 30.6-31.4 2 28.4 26.2-30.6 1 28.4 METACARPAL Lateral width, proximal end 3 23.5 22.7-24.8 2 27.4 26.6-28.2 1 22.2 Anteroposterior width, proximal end 3 16.7 16.4-17.1 2 20.3 20.2-20.5 1 18.1 METATARSAL Lateral width, proximal end 1 23.2 2 22.4 21.5-23.3 1 20.8 Anteroposterior width, proximal end 1 23.7 2 24.8 22.9-26.6 1 23.2 FIRST PHALANX Length 7 41.0 37.743.2 10 43.5 39.4-48.9 Width, proximal end 7 13.4 12.2-13.9 10 13.3 11.6-15.2 Inglis IA Coleman IIA Reddick I N X OR N X OR N X OR HUMERUS Lateral width, distal con(lyles 4 32.9 31.3-34.4 7 30.4 28.2-33.2 3 29.6 28.6-31.2 METACARPAL Lateral width, proximal end 1 29.6 1 24.3 2 27.0 27.6-26.4 Anteroposterior width, proximal end 1 20.7 1 23.8 2 19.5 19.1-20.0 METATARSAL Lateral width, proximal end 3 25.2 24.6-25.5 7 23.6 22.5-24.8 1 23.1 Anteroposterior width, proximal end 3 26.5 26.0-27.4 7 25.2 23.6-35.8 1 21.9 FIRST PHALANX Length 1 46.6 7 43.4 40.3-44.4 4 42.5 40.6-44.4 Width, proximal end 1 15.4 7 13.0 12.2-13.8 4 14.7 14.3-15.5 X 1976 R O BER TSO N : H A ILE X V A LATE S T P LIO C E N E M A M M A LS 171 TABLE 20. CONTINUED Devil's Den Itchetucknee River N X OR NX OR HUMERUS Lateral width, distal condyles 8 35.0 33.5-36.7 7 31.4 28.6-34.0 METACARPAL Lateral width, proximal end 12 28.1 27.0-30.2 11 27.2 23.7-29.6 Anteroposterior width, proximal end 12 20.3 18.5-21.3 11 18.4 17.1-21.5 METATARSAL Lateral width, proximal end 11 25.5 24.3-26.9 14 25.6 21.0-26.7 Anteroposterior width, proximal end 11 26.9 25.7-27.8 14 25.7 22.5-29.6 FIRST PHALANX Length 14 41.5 26.145.8 Width, proximaI end 14 12.7 11.5-13.8 172 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 3 cervical vertebrae; UF 10891: 3 left humeri; UF 10893: right humerus; UF 17479: 3 metacarpals ( 2 left, 1 right); UF 10887: metatarsal; UF 17480: left astragalus; UF 10886: 3 podials; UF 10888: 15 phalanges. The genus Odocoileus is recognized in many of the Blancan faunas of North America. In most cases, however, the material has been scarce and species allocation has not been possible. The definitions of closely related species ( and genera) of deer are very similar, and the taxonomic problem is compounded by the large amount of individual variation in these forms. The nature of the antlers and the fourth lower premolars are considered important in cervid tax- onomy at the generic and subgeneric levels; unfortunately, these features are also extremely variable. In Simpson's ( 1928) discussion of Blasto- cerus extran€us, he pointed out that specimens of Odocoileus studied approached Blastocerus in character, and some Recent Blastocerus ma- terial diverged from his specimen toward Odocoileus. The subgenuus Procoileus ( Frick 1937 ) is based primarily on the characters of the P4· Examination of a large sample of Blancan Odocoileus teeth from the Santa Fe I fauna showed individual variation that approached the char- acters of Blastocerus, Procoileus, and even Craniocerus. Similar, though less pronounced, examples of individual variation may be observed in samples of Odocoileus teeth from several later Pleistocene and Recent sites in Florida. It appears that tooth morphology by itself is too variable to permit taxonomic assignments at the species level. Size also has been a criterion for species determination of fossil deer. Although the Haile XV A deer are relatively small, this is not considered taxonomically signiBcant, inasmuch as a size comparison of postcranial elements from later stages of the Pleistocene ( Table 20) shows wide size variation that encompasses the Haile XV A samples. Harlow and Jones ( 1965) also showed that in Florida the size of Recent deer fluctuates signifcantly from habitat to habitat. An element by element comparison of the Haile XV A deer material with specimens of 0. uirginianus from other Florida Pleistocene sites, and with Recent specimens, shows no consistent differences. Conse- quently, it is considered safest to refer the Haile XV A cervid to Odocoil- eus virginianus. AGE AND CORRELATION North American vertebrate paleontologists, including Schultz ( 1938), Wilson ( 1938), and McGrew ( 1944), have long recognized the Blancan age on the basis of the presence and absence of certain mammalian genera. McGrew ( 1944) summarized the criteria as follows: 1976 ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 173 1) the absence of typically Pliocene genera; 2) the presence of characteristically Middle and Late Pleistocene genera; 3) the presence of certain genera that survived from the Hemphillian but which did not live past the Blancan; 4) the presence of certain genera that were limited to the Blancan; and 5) the absence of certain genera that did not make their appearance until after the close of the Blancan. Each of the above criteria may be examined with regard to the Haile XV A fauna: 1) There are no typically Hemphillian genera represented in Haile XV A. Kraglievichia may be of Pliocene age in South America, but it is Blancan or later in North America. It should be pointed out here that such negative evidence is never conclusive and is particularly dangerous with regard to the Haile XV A fauna. The small number of taxa found probably indicates that the specimens collected represent only a small part of the overall fauna. 2) Several genera that range into the Middle and Late Pleistocene appear in this fauna; such tan include Odocoileus, M!/loht/us, Smilodon, and Sig- modon. 3) Of the genera that survived the Hemphillian but not the Blancan only Nannippus is definitely present. The presence in the fauna of a gom- photheriid may also fit this category, despite the fact that the specimen cannot be definitely assigned to a genus. 4) Only one subgenus, Dolichohippus ( genus Equus), in Haile XV A is re- stricted to the Blancan in North America, The species Sigmodon medius, Nannippus phlegon, and Equus ( Dolichohippus) simplicidens are wholly restricted to the Blancah and are widespread through most of North America. 5) The genera Dast/Pus and Mulohvus, which previously were thought to have appeared after the close of the BIancan, are present in Haile XV A. In view of the peculiar ecological and geographic features of the Haile XV A fauna, these age extensions are not unexpected and do not affect the hy- pothesis of an older age suggested by the weight of other evidence. In each case the species is more primitive than those occurring in younger faunas. The weight of the mammalian faunal evidence ( see Table 21), par- ticularly the presence of Nannippus, Equus ( Dolichohippus ), and Sig- modon medius, points conclusively to a Blancan age. A more refined correlation of the Haile XV A fauna with other North American Blancan faunas ( i.e. outside of Florida) is not feasible at present because of the geographic distances and ecological differences. Most Blancan faunas are located in western North America, but none, other than those in Florida, are known from the Gulf Coastal Plain. Consequently, the lack of a more detailed faunal comparison need not imply a great difference in time. The absence of Kraglieuichia and Dasypus from Blancan sites elsewhere in North America more likely results from geographic rather than temporal differences. Conversely, a closer similarity to other Blan- can faunas in Florida ( e.g. Santa Fe I) does not indicate a close temporal correlation. Although definite similarities exist between the Haile XV A and Santa Fe I faunas, enough differences are present to indicate a differ- TABLE 21.-FAUNAL COMPARISONS OF HAILE XV A AND OTHER BLANCAN LOCALrnES. m u io n IE D U .134inoS R nE H 174 B U LLE T IN FLO R ID A S TATE M U S E U M Vol. 20, N o. 3 .- 5 1- 0 9 1~ hi .09# 4 1 15 1 e f S an ta F e I, F lo rid a E B la nc o + C ita C an yo n, T ex as R ed L ig h t, Te xa s i 2~1.1 8 i 0 al ds pe th , Te xa s ns on , A riz on a· rt is R an ch , A riz on a d D ra w , N eb ra sk a j a :~ ~ 5~§ 6 3§ .s~1 a jj 3 1 e;jpl 11111,13Haile XV A AE&6 0819 'Ig ro u lp g d foic hm a Cryptotis X Scalopus X Glossotherium XXX X X Das!/pim X Kraglievichia X X Syludagus ?X X X X X Sigmodon XXXXX X Smitodon X X Satherium XXX Tapirus X Nannippus XXXXXX X Equus ( Dolichohippus) XXXXXX?XXXXXXXXXXXX? Mylohus X Odocoileus XX?XX X X X XX Hemiauchenia XXXXX XXXX X X XX 1976 ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 175 ence in time and probably ecology as well. The Santa Fe I camel be- longs to the typically Blancan,species Hemiauchenia blancoensis, whereas the Haile XV A form most closely resembles H. macrocephala. This probably indicates a younger age for Haile XV A, as H. macrocephala is characteristic of the Irvingtonian and Rancholabrean ages elsewhere. Certain taxa found in the Santa Fe I fauna that might be expected at Haile XV A do not occur there. One of the most notable absences is the typically Blancan hyaenoid dog, Borophagus, although in all like- lihood this represents a sampling error. The absence of Castoroides and Hydrochoerus from Haile XV A probably can be explained in the same manner. Other differences between the faunas of Haile XV A and Santa Fe I are most likely ecologically induced, as will be discussed in the next section. Among the best represented animals in the Haile XV A fauna are the edentates, a group of undoubted South American origin. Of the three genera at Haile XV A, only Glossotherium has been recognized pre- viously in North American Blancan faunas, being present in the Blancan faunas of Texas and Nebraska ( Table 21). Dasypus is known only from later deposits elsewhere in North America, and Kragli€vichia has not heretofore been reported from this continent, although its probable descendant, Pampatherium ( - Holmesina), has been found in Irving- tonian and Rancholabrean deposits. This early record of a major influx of edentates from South America is further indication of the late Blancan age of Haile XV A. The small chlamythere from Haile XV A shows a remarkable resemblance to the species Kraglievichia paranensis from Late Pliocene deposits of South America. Although no authentic Chapadmalalan ( Early Pleistocene) species is known, it probably lived in regions more tropical than Argen- tina and probably also was physically similar to the Florida material. Florida and Argentina apparently represent the northern and southern limits of the range for this form, and it seems safe to conclude that con- temporaneous populations occupying the middle portion of the range were broadly similar to those at the extremes. The close resemblance of the Haile XV A Glossotherium to that from the Chapadmalalan stage also supports this correlation. This late Blancan inHux of edentates into Florida adds a new dimen- sion to the previous correlation between the Blancan stage of North American and the Chapadmalalan stage of South America. This correla- tion had previously been based upon the appearance of a few Nearctic species in the Chapadmalalan and Uquian of South America ( Patterson and Pascual 1968, Simpson 1969). Haile XV A adds the obverse re- lationships. North America received an important contingent of Neo- 176 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 3 tropical taxa in the Blancan. The only Nearctic taxa in the Haile XV A fauna that closely resemble species in the Chapadmalal fauna are Smilo- don and Hemiauchenia. The combination of correlative taxa of both Nearctic and Neotropical origin makes the resemblance between the Blancan fauna of Florida and the Chapadmalalan fauna of Argentina particularly strong. EUSTATIC CONSIDERATIONS In addition to faunal correlation, the effects of sea level changes caused by Pleistocene glaciation may be employed as an aid in dating Florida Pleistocene deposits. Glacial eustatic fluctuations of sea level have resulted in "terraces" or "abandoned shore lines," which are domi- nant geomorphic features of the Florida landscape ( Cooke 1945, Mac- Neill 1949, White 1958, Alt and Brooks 1965). The location of a deposit that can be correlated with these features can be quite helpful in age determination. For example, a terrestrial deposit that is presently lying at or below sea level must have been laid down at a time when sea level was as low or lower than it is now. This would be an indication that the deposit was laid down during glacial times, as these were the times of low sea level. Once this is determined, the faunal elements may indi- cate which glacial stage is represented. Inland deposits at higher elevations present a problem in that they could have been deposited either during a glacial or an interglacial stage. If it can be shown that a particular inland deposit was coastal at its time of deposition, then it must represent an interglacial stage, when sea level was high. Haile XV A stands at an elevation of 90 ft above sea level. It contains shark vertebrae that are similar in preservation to the other vertebrate fossils in the site ( and therefore are probably not in- trusive). This indicates the close proximity of the sea during the time of deposition. Remains of other Bsh species also suggest at least estuarine conditions. Haile XV A could represent either an interglacial stage or a preglacial wet interval during the Blancan; a preglacial age seems prob- able on biostratigraphic grounds. PALEOECOLOGY Among the lower vertebrate remains from Haile XV A are numerous lamnoid type shark vertebrae. Their preservation does not differ fronn that of other vertebrate hard parts in the site, and thus they cannot be interpreted as intrusive from adjacent Eocene limestones. ( Sharks are known in the Ocala Formation, but not abundantly and mainly from teeth.) A variety of bony fishes of both freshwater and marine taxa await further study. These include Amia, Lepisosteus, Centropomus, 1976 ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 177 Bagre, Ictalurus, and numerous other as yet unidentified species. Am- phibians are represented by both anurans and urodeles. The turtles from Haile XV A include the aquatic forms Chrysemys platymarginata, Tri- onyx, Kinosternon, and Chely,ira, and the terrestrial box turtle Terrapene carolina. The material of Chrysemys platymarginata C Weaver and Rob- ertson 1967) includes numerous complete shells. They all occur in up- right position, but facing in various directions, and are most abundant in the uppermost green clay unit. Most of the birds at Haile XV A are aquatic and include a duck, three herons, and a grebe ( Campbell 1976). Although the lower vertebrate fauna has not been studied in detail, this preliminary analysis is enough to indicate an aquatic environment near the sea. The mammalian taxa may be separated for the most part into two ecological components, a smaller group of aquatic species and a larger group of terrestrial species. Castor inhabits medium to larger permanent bodies of flowing water. Pteronura, the probable descendant of Sather- ium, inhabits streams from the Guianas to Argentina ( Walker 1975). Its surrounding terrestrial habitat is probably secondary to its stream habi- tat, as it lives in suitable rivers that flow through a variety of terrestrial areas. These mammals are thus ecologically associated with the turtles, fishes, and water birds as a primary inhabitants of the aquatic site. The abundance of large terrestrial animals and the relative scarcity of smaller terrestrial ones strongly suggests a spring-head habitat. The presence of a few of the denser bones ( and especially teeth) of Sig- modon, Scalopus, Cryptotis, and Cryptopterus must be ascribed to for- tuitous burial. The only other small mammals in the fauna are Castor and Satherium, which are normally aquatic in nature and thus may be expected abundantly in the stream. All the other mammals in the Haile XV A fauna are large terrestrial forms, which were probably trapped as they drank by the treacherous slopes and abrupt limestone ledges that often surround Florida springs. We may next consider the nature of this terrestrial faunal sample. The terrestrial animals consist of Terrapene, the box turtle, one species of quail, one turkey ( Campbell 1976), a few small mammals, and a con- siderable variety of large mammals. The presence of eurytopic large forms, such as Smilodon, Hemiauchenia, two subgenera of Equus, Nan- nippus, Odocoileus, and of the small mammals Cryptotis, Scalopus, and Syluilagus tell us little about the local paleoecology. These forms are present in many Blancan local faunas of North America and probably ranged through many habitats. Several taxa in the Haile XV A fauna, however, strongly suggest the presence of a forest community adjacent to the site of deposition. The 178 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 3 new species of Myloh!/us lends strong support to this interpretation. Lundelius ( 1960) reviewed evidence that indicates a forest habitat for Mylohyus. He also included Castor, Tapirus, and Glaucomys ( among others) as forest forms. Both Castor and Tapirus are present in the Haile XV A fauna, and the flying squirrel Cryptopterus ( like Glaucomys ) certainly indicates a forest environment. Glossotherium may also have favored forested or partly forested areas. Das!/pus and Kraglievichia probably were restricted in North America to the Gulf Coastal Plain by their need for moderate to warm tempera- tures and high rainfall. This may explain their absence from all other North American Blancan faunas. In South America, Kraglievichia seems to have been largely restricted to the northern tropical part of the con- tinent. These edentates suggest a tropical or subtropical climate in central Florida during the Latest Pliocene. In summary both the geological and faunal evidence seem to indicate that the Haile XV A site lay at the springhead of a coastal stream that Eowed through a tropical or subtropical forest or forest savanna. ZOOGEOGRAPHY The Haile XV A fauna is the first Blancan fauna to be described from the Gulf Coastal Plain. Its unusual faunal assemblage ( compared to Blancan faunas in western North America) apparently is a reflection of its location. Few American Blancan faunas are from coastal areas, and no others are so close to the Middle American tropics. The most striking distinction of the Haile XV A fauna is the diversity of edentates. Evidently it was in contact with the tropical corridor through which South America emigrants moved. Of the edentates, only Glossotherium was n6t restricted to the tropical portion of North America, as indicated by its presence in the Blanco ( Texas) and Broad- water ( Nebraska ) faunas. This genus probably was also widely dis- tributed on the South American continent, as the presence of G. chap- admalensis in Argentina indicates. On the other hand, Kraglieuichia probably was restricted to the tropical or subtropical parts of North America during Blancan time, as indicated by its absence from all other North American Blancan faunas. Its presence in earlier Pliocene deposits of Argentina suggests that its range extended southward into more temperate climes in South America, but later it seems to have re- treated tropicad in that continent also. Apparently the later Pleistocene evolution from Kraglieuichia to Pampatherium involved, among other things, an increased tolerance to colder climate, which perhaps was cor- related with the greater size of Pampatherium. Pampatherium was able, in the course of the Pleistocene, to disperse northwestward from 1976 ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 179 the Gulf Coastal Plain throughout most of temperate North America; in South America it spread southward throughout Argentina. The genus Dasypus seems to have been even more closely restricted to tropical regions than either Glossotherium or Kraglieoichia. It also was able to move out of the Gulf Coastal Plain later in the Pleistocene (as Dasypus bellus ), reaching as far west as Texas and as far north as West Virginia ( Guilday and McCrady 1966), Tennessee ( Guilday et al. 1969), and Missouri ( Simpson 1949). The probable mechanism permit- ting its tolerance to colder climate may have been an increase in size, as was true of Pampatherium. Its drastic size decrease in latest Pleisto- cene time ( assuming D. bellus is closely related to D. nouemcinctus) could account for its present limited southern distribution ( Humphrey 1974). Dast/pus bellus is not known from South American fossil de- posits, although there is little doubt that the genus originated on that continent. This suggests that its range was similarly restricted to tropical areas on that continent, where fossil sites are rare and little known. The unusual distributional pattern of Cryptopterus ( Latest Pliocene of Florida and trans-Beringean of West Germany) indicates a wide distri- bution of large Group I flying squirrels. Presumably this group was re- stricted in the New World to the eastern deciduous forests of North America. James ( 1963) reported a smaller simple-toothed group of sciuropterines in Miocene deposits of California, but no other large complex-toothed specimens are known from the New World. Presum- ably Cryptopterus ranged from the eastern deciduous forests of North America through similar forests in Alaska and Siberia into Central Europe. The presence of Pteromys at Choukoutien ( Young 1934) adds another very late Cenozoic record of a large Hying squirrel from a temper- ate region, but it is not closely related to Cryptopterus ( Mein 1970 ). The close resemblance between the Florida and West German specimens suggests rapid dispersal to the New World during late Hemphillian or Blancan time. Obviously this fortuitous record from the Blancan of Flor- ida is peripheral to the main theatre of sciuropterine evolution in the Old World tropics. SUMMARY The Haile XV A mammals are from one of the few Blancan faunas on the Gulf Coastal Plain. It is the only such fauna that has been ex- tensively studied to date and also the only one from a subtropical forest situation. Included among the mammals from this site is Glossotherium chap- admalensis, previously known only from Argentina, It is possible that this form gave rise to G. robustus in South America and G. harlani in 180 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 3 North America. Apparently, Glossotherium was able to spread through- out much of temperate North America, whereas the other edentates represented in the Haile XVA fauna were restricted to subtropical re- gions during most of the Pleistocene. The occurrence of a new species of Kraglievichia at Haile XV A and Santa Fe I is the first record of this chlamythere genus in Nort:h America. The presence of this species, which is closely related to K. paranensis, strengthens the correlation between North and South American faunas of Blancan and Chapadmalalan ages respectively. The oldest evidence of Dat/pus bellus occurs at Haile XV A. This species or its immediate ancestors undoubtedly originated in South America. The giant flying squirrel, Cryptopterus, gives the local fauna an un- expected link with Old World Tertiary forest faunas. The new species from Haile XV A was among the last ( Late Pliocene) dispersals of Old World subtropical forest dwellers to North America ( Repenning 1965). The presence of Satherium in the Haile XV A fauna marks the first Florida record of this large aquatic otter. It adds yet another link to the neotropical fauna, as it is closely related to Pteronura, the giant otter of South America. The presence of Mylohyus ~oridanus at Haile XV A marks the earliest record for this genus, although its ancestor, Prosthennops, is known from the Pliocene of North America. It probably occurred in subtropical forests, in contrast to the contemporary genus Platygonus. If the tentative assignment of the Haile XV A camelid to Hemiau- chenia macroceplulla is correct, it marks the earliest record for this species anywhere. The typical Blancan species, H. blan€oensis, occurs at Santa Fe I, suggesting an age difference between the Florida Blancan localities. The remaining members of the Haile XV A fauna are fairly typical representatives of the North American Blancan fauna. Such character- istic taxa as Equus ( Dolichohippus) simplicidens, Nannippus phlegon, and Sigmodon medius substantiates a Blancan age assignment for Haile XV A. The presence of sharks and brackish-water fishes indicates that the sea stood near Haile during the time that Haile XV A was deposited. The presumed sea level, which was 90 ft higher that that of today, is believed to represent a pre-glacial high sea level. Although the bulk of the vertebrate fauna at Haile XV A indicates stream conditions, an adjacent terrestrial community is also well repre- sented. Of the mammals, only Satherium and Castor represent the aquatic community. The terrestrial community includes several clear 1976 ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 181 indicators of a tropical forest situation, such as Mylohgus, Cryptopterus, three kinds of edentates, and Tapirus. The others are eurytopic small and large mammals. BIBLIOGRAPHY Akersten, W. A. 1970, Interpretation of sediments and vertebrate fossils in fll of Red Light bolson, southeastern Hudspeth County, Texas. Soc. Econ. Paleo. Miner., Permian Basin Sec., 1970 Field Trip Guidebook, Publ. 70-12: 82-7. Alt, D., and H. K. Brooks. 1965. Age of the Florida marine terraces. J. Geol. 73(2): 406-11. Ameghino, F. 1875. Notas sobre algunos f6siles nuevos de la Formaci6n Pampeana. Ameghino Obras 11: 13-7. 1887. Apuntes preliminares sobre algunos mamiferos extinguidos del yacimiente de Monte Hermoso existentes en el Museo de La Plata. Extracto, Entrada 1, Tomo 1, Bol. Museo La Plata Imprenata, El Censor 8: 339-53. 1902. Notices pr61iminaires sur des mammif6res nouveaux des terrains cr6tac6s de Patagonie. Bol. Acad. Cien. Cordoba Tome 17: 5-70. 1904. Nuevas especies de mamiferos, Cretacios y Terciarios, de la Republica Argentina. Ann. Soc. Cient. Age. 16: 193-208. Auffenberg, W. 1955. Glass lizards ( Ophisaurus) from the Pleistocene and Pliocene of Florida. Herpetologica 2: 133-36. 1957. A note on an unusually complete specimen of Dasupus bellus ( Simpson) from Florida. Quart. J. Florida Acad. Sci. 20(4): 233-37. 1958. Fossil turtles of the genus Terrapene in Florida. Bull. Florida State Mus., Biol. Sci. 3(2): 53-92. Bader, R. S. 1957. Two Pleistocene mammalian faunas from Alachua County, Florida. Bull. Florida State Mus., Biol. Sci. 2( 5): 53-75. Barbour, E. H., and C. B. Schultz. 1937. An Early Pleistocene fauna from Ne- braska. Amen Mus. Nov. 942: 1-10. Bjork, P. R. 1970. The Carnivora of the Hagerman local fauna ( Late Pliocene) of southwestern Idaho. Amer. Phil. Soc. Trans. 60: 1-54. Brisson, M. J. 1762. Regnum animale in classes IX distributum sive synopsis metho- dica. Editio Altera. Leiden, Theodorum Haak: 1-296. Brodkorb, P. 1963. A giant flightless bird from the Pleistocene of Florida. Auk 80( 2): 111-15. Bronn, M. 1838. Lethaea geognostica. Stuttgart, Vol. 2: 545-1350. Brown, B. 1903. A new genus of ground sloth from the Pleistocene of Nebraska, Paramytodon nov. gen. Bull. Amer. Mus. Nat. Hist. 19: 569-83. Cabrera, A. 1929. Una revisi6n de los mastodontes Argentinos. Rev. Mus. La Plata 32: 61-144. Campbell, K. C. 1976. The Early Pleistocene avifauna of Haile XV A, Florida. Wilson Bull. 88:--. Castellanos, A. 1927. Breves notas sobre los clamiterios. Publ. Cent. Estud. Ingen. Rosario: 1-8. 1937. Anotaciones sobre la linea Blogenitica de los clamiterios. Publ. Facul. Cien. Univ. Nac. Litoral. Serie Teen.-Cient. Publ. 8: 1-135. 1946. Una nueva especie de clamiterio Vassallia maxima n. sp. Publ. Inst. Fiso. Geol., Univ. Nac. Litoral 6( 26) : 5-47. 1957. Nuevos g6neros de clamiterio y de gervo y una nueva especie de Plohophorus. Ameghiniana 1(3): 5-14. Churcher, C. S. 1967. Smilodon neogaeus en las barrancas costeras de Mar del Plata, provincia de Buenos Aires. Publ. Museo Munic. Cien. Nat. Mar del Plata 1: 245-62. Cooke, C. W. 1945. Geology of Florida. Florida Geol, Surv. Bull. 29: 1-339. 182 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 3 Cope, E. D. 1880. On the extinct cats of America. Amer. Nat. 14: 838-58. 1892. A contribution to the vertebrate paleontology of Texas. Proc. Amer. Phil. Soc. 30: 123-31. 1893. A preliminary report on the vertebrate paleontology of the Llano Estacado. Geol. Surv. Texas, 4th Ann. Rept.: 1-136. 1899. Vertebrate remains from Post Kennedy bone deposit. J. Acad. Nat. Sci. Phila. 11: 193-286. Dawson, M. R. 1958. Later Tertiary Leporidae of North America. Univ. Kansas Paleo. Contr., Vertebrata 6: 1-75. Dehm, R. 1962. Altpleistocane sauger von Schernfeld bei Eichstatt in Bayern. Munich. Bayer. StaatE. Palaon. Hist, Geol., Mitteilungen 2: 17-61. Deperet, C. 1897. Les animaux Pliocdnes du Roussillon. Mem. Soc. Geol. France; Paris. Paleontologie 3: 1-164. Downey, J. S. 1968. Late Pliocene Lagomorphs of the San Pedro Valley, Arizona. U. S. Geol. Surv. Prof. Pap. 600 D 169-73. 1970. Middle Pleistocene Leporidae from the San Pedro Valley, Arizona. U. S. Geol. Surv. Prof. Pap. 700 B: 131-35. Downs, T., and J. A. White. 1968. A vertebrate faunal succession in superposed sediments from Late Pliocene to Middle Pleistocene in California. 23rd Intnatl. Geol. Cong. 10: 41-7. Frick, C. 1937. Horned ruminants of North America. Amer. Mus. Nat, Hist. Bull. 69: 1-669. Gaillard, C. 1899. Mammifdres Miocdnes nourieaux ori peu connus de la Grive-St.- Alban ( Isere). Arch. Mus, Hist. Nat., Lyon 7: 1-79. Gazin, C. L. 1934. Upper Pliocene mustelids from the Snake River Basin of Idaho. J. Mamm. 15: 137-49. 1936. A study of the fossil horse remains from the Upper Pliocene of Idaho. Proc. U. S. Natl. Mus. 83 ( 2985): 281-320. 1942. The late Cenozoic vertebrate faunas from the San Pedro Valley, Arizona. Proc. U. S. Natl. Mus. 92(3155): 475-518. 1950. Annotated list of fossil Mammalia associated with human re- mains at Melbourne, Florida. J. Wash. Acad. Sci. 40: 397-404. Gidley, J. W. 1922. Preliminary report on fossil vertebrates of the San Pedro Valley, Arizona. U. S. Geol. Surv. Prof. Pap. 131 E: 119-31. Gray, J. E. 1821. On the natural arrangement of vertebrose animals. London Med. Reposit. 15( 1): 296-310. . 1837. Description of some new or little known Mammalia, principally in the British Museum collection. Mag. Nat. Hist. New Series 1: 577-87. . 1867. Notes on the skulls of some hares ( Leporidae) and picas ( Lagomyidae) in the British Museum. Ann. Mag, Nat. Hist, 20: 219-25. Guilday, J. E., and A. D. McCrady. 1966. Armadillo remains from Tennessee and West Virginia caves. Bull. Natl. Speleol. Soc. 28: 183-84. Guilday, J. E., H. W. Hamilton, and A. D. McCrady. 1969. The Pleistocene verte- brate fauna of Robinson Cave, Overton County, Tennessee. Paleovertebrata, Montpellier 2: 25-75. Hall, E. R., and K. R. Kelson. 1959. The mammals of North America. The Ronald Press, New York, 1079 p. Harlow, R. F., and F. K. Jones. 1965. The white-tailed deer in Florida. Florida Game and Freshwater Fish Comm. Tech. Bull. 9: 1-240. Hay, 0. P. 1899. On the nomenclature of certain American fossil vertebrates. Amer. Geologist 24: 345-349. . 1917. Vertebrata mostly from stratum No. 3, Vero, Florida, together with descriptions of new species. Florida Geol. Surv., 9th Ann. Rept.: 43-68. 1928. Again on Pleistocene man at Vero, Florida. J. Wash. Acad. Sci. 18 (9): 233-241. 1976 ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 183 Hibbard, C. W. 1937. An Upper Pliocene fauna from Meade County, Kansas. Trans. Kansas Acad. Sci. 40: 239-265. 1941, Mammals of the Rexroad fauna from the Upper Pliocene of southwestern Kansas. Trans. Kansas Acad. Sci. 44: 265-312. 1953: Equus ( Asinus) calobatus Troxell and ass6ciated vertebrates from the Pleistocene of Kansas. Trans. Kansas Acad. Sci. 56: 111-26. 1956. Vertebrate fossils from the Meade Formation of southwestern Kansas. Pap. Michigan Acad. Sci. 41: 145-260. , and W. W. DaIquest. 1962. Artiodactyls from the Seymour Forma- tion of Knox County, Texas. Pap. Michigan Acad. Sci. 48: 83-99. Hirschfield, S. E., and S. D. Webb. 1968. Plio-Pleistocene megalonychid sloths of North America. Bull. Florida State Mus., Biol: Sci. 12(5): 213-296. Hoffstetter, R. 1952. Les mammifdres Pleistocdnes de la R6publique de l'E:quateur. Mem. Soc. Geol. France 66: 1-391. Howe, J. A. 1970. The Range of Variation in Equus ( Plesippus ) simplicidens Cope from the Broadwater Quarries of Nebraska. J. Paleo. 44: 958-68. Hugueney, M., and P. Mein. 1966. Les rougeurs Phocdnes du Roussillon, dans collections Lyonnaises. Lyons Univ., Fac. Sci., Lab. Geol., Travaux, New Series 13: 243-66. Humphrey, S. R. 1974. Zoogeography of the nine-banded armadillo ( Dasypus novemcinctus) in the United States. Bioscience 24: 457-62 James, G. T. 1957. An edentate from the Pleistocene of Texas. J. Paleon. 31 (4): 796-808. 1963. Paleontology and nonmarine stratigraphy of the Cuyama Valley badlands, California. Part I. Geology, faunal interpretations, and systematic descriptions of Chiroptera, Insectivora, and Rodentia. Univ. Calif. Publ., Geol. Sci. 45: 1-154. Kellogg, L. 1911. A fossil beaver from the Kettleman Hills, California. Univ. Calif. Publ., Bull. Dept. Geol. 6( 17): 401-2. Kinsey, P. E. 1974. A new species of peccary, Mylohvus, from the Early PIeistocene of Florida. In Webb, S. D., ed., Pleistocene Mammals of Florida. Univ. Florida Press, Gainesville, 270 p. Kraglievich, L. 1925. Cuarto nuevos gravigrados de la fauna araucana Chapadma- tense. Anal. Mus. Nac. Hist. Nat. 33: 215-35. 1928. Mylodon darwinii Owen, es la especie genotipo de M!/lodon Owen. Rectificacidn de la nomenclatura gen6rica de los milodontes. Physis 9: 169-85. 1934. La antiquedad Pliocena de Ias faunas de Monte Hermoso y Chapadmalal, dedicadas de su comparaci6n con las que le precedieron y sucedi- eron. Montevideo, 136 p. 1946. Sobre camelidos Chapadmalenses. Inst. Mus. Univ. Nac. La Plata, Not. Mus. La Plata, Tomo 11, Paleontologia 93: 317-29. 1948. Smilodontodon riggsi, n. gen. n. sp., un nuevo y pequefio esmilo- donte en la fauna Pliocena de Chapadmalal. Rev. Mus. Argentina Cien. Nat., Cien. Zool. 1( 1 ) : 1-44. Kretzoi, M. 1959. Fauna und faunenhorizont von csarnota. Jahr. Ungar. Geol. Anst., 385 p. Leidy, J. 1889a. Description of the vertebrate remains from Peace Creek, Florida. Trans. Wagner Free Inst. Sci. 2: 19-32. 1889b. Fossil vertebrates from Florida. Proc. Acad. Nat. Sci. Phila.: 96-7. Linnaeus, C. 1758. Systema Naturae per regna tria naturae, secundum classes, ordines, genera, species cum characteribus, differentiis, synymis, locis. Editio decima, Reformata. Stockholm, Laurentii Salvii 1: 1-824. Lund, P. W. 1838. Blik paa Brasiliens Dyreverden for Sidste Jordomvaeltning. Overs. K. Danske Videnskebernes Selskabs., Forh. 8: 29-144. 184 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 3 Lundelius, E. L. 1960. Mt/Zoh!/us nasutus, long-nosed peccary of the Texas Pleisto- cene, Bull. Texas Mem. Mus. 1: 9-40. MacNeill, F. S. 1949. Pleistocene shorelines in Florida and Georgia. U. S. Geol. Surv. Prof. Pap. 221 F: 95-107. Martin, R. A. 1974. Fossil mammals from the Coleman IIA fauna, Sumter County. In Webb, S. D., ed., Pleistocene Mammals of Florida. Univ. Florida Press, Gainesville, 270 p. MeGrew, P. 0. 1944. An early Pleistocene ( Blancan) fauna from Nebraska. Field Mus. Nat. Hist., Geol. Ser. 9(2): 33-66. McKenna, M. C. 1962. Eupetaurus and the living petauristine sciurids. Amer. Mus. Nov.: 2104: 1-38. Mein, P. 1970. Les Sciuroptdres ( Mammalia, Rodentia) ndogends d'Europe Occi- dentale. Geobios 3( 3): 7-77. Merriam, J. C., and C. Stock. 1932, The Felidae of Rancho La Brea. Carnegie Inst. Wash. Publ. 422: 3-231. Moreno, F. P., and A. Mercerat. 1891. Exploraci6n arqueoldgica de la Provincia de Catamarea. Paleon. Rev. Mus. La Plata 1: 222-34. Oesch, R. D. 1967. A preliminary investigation of a Pleistocene vertebrate fauna from Crankshaft Pit, Jefferson County, Missouri. Natl. Speleo. Soc. Bull. 29( 4): 163-85. Owen, R. 1840. The zoology of the voyage of H.M.S. Beagle ... Part I: Fossil Mammalia. London: 13-111. . 1842. Description of the skeleton of an extinct gigantic sloth, Mt,lodon fobustus Owen ... London: 1-176. Patterson, B., and R. Pasqual. 1968. Evolution of mammals on southern continents. V. The fossil mammal fauna of South America. Quart. Rev. Biol. 43(4): 409-451. Patton, T. H. 1963. Fossil vertebrates from Miller's Cave, Llano County, Texas. Bull. Texas Mem. Mus, 7: 141. Paula Couto, C. de. 1956. Resumo de mem6rias de Lund s6bre as cavernas de Lagoa Santa e seu conteudo animal. Univ. Brasil Mus. Nac., Publ. avulsas 16: 1-14. Porta J. de. 1962. Edentata del Mioceno de La Venta ( Colombia). I. Dasypo- doidea y Glyptodontoidea. Bol. Geol., Univ. Indus. Santander 10: 5-23. Ray, C. E. 1957. A list, bibliography, and index of the fossil vertebrates of Florida. Florida Geol. Surv., Spec. Publ. no. 3: 1-175. . 1958. Addition to the Pleistocene mammalian fauna from Melbourne, Florida. Bull. Mus. Comp. Zool., Harvard 119: 42149, , S. J. Olsen, and H. J. Gut. 1963. Three mammals new to the Pleisto- cene fauna of Florida, and a reconsideration of fve earlier records. J. Mamm. 44(3): 373-95. Repenning, C. A. 1965. Palearctic-Nearctic mammalian dispersal in the Late Ceno- zonic, In Hopkins, D. M., ed., The Bering Land Bridge. Stamford Univ. Press, 511 p. Robertson, J. S. The fossil bison of Florida. In Webb, S, D., ed., Pleistocene Mam- mals of Florida. Univ. Florida Press, Gainesville, 270 p. Rovereto, C. 1914. Los estratos Araucanos y sus f6siles. Anal. Mus. Nac. Hist. Nat. Buenos Aires 25: 1-236. Savage, D. E. 1951. Late Cenozoic vertebrates in the San Francisco Bay region. Univ. Calif. Publ., Bull. Dept. Geol. 28( 10): 215-314, Say, T. 1823. Account of an expedition to the Rocky Mountains. Volume I ( p. 163). In Long. Schultz, G. E. 1969. Geology and Paleontology of a Late Pleistocene basin in southwest Kansas. Geol. Soc. Amer, Spec. Pap. 105: 1-85. Schultz, J. R. 1938. A Late Cenozoic vertebrate fauna from the Coso Mountains, Inyo County, California. Carnegie Inst. Wash. Publ. 487: 76-109. 1976 ROBERTSON: HAILE XV A LATEST PLIOCENE MAMMALS 185 Sellards, E. H. 1915. Chlamt/therium septentrionalis, an edentate from the Pleisto- cene of Florida. Amer. J. Sci. 4(40): 19-31. 1917. On the association of human remains and extinct vertebrates at Vero, Florida. J. Geol. 25 (1): 4-24. Simpson, G. G. 1928. Pleistocene mammals from a cave in Citrus County, Florida. Amer. Mus. Nov. 328: 1-16. 1929. Pleistocene mammalian fauna 6f the Seminole Field, Pinellas County, Florida. Amer. Mus. Nat. Hist. Bull. 56( 8): 561-99. 1930. Holmesina septentrionalis, extinct giant armadillo of Florida. Amer. Mus. Nov. 442: 1-10. 1945. The principles of classification and a classification of mammals. Bull Amer. Mus. Nat. Hist. 85: 1-350. 1947. Holarctic mammalian faunas and continental relationships dur- ing the Cenozoic. Bull. Geol. Soc. Amer. 58: 613-888. 1949. A fossil deposit in the cave in St. Louis. Amer. Mus. Nov. 1408: 1-46. 1969. South American Mammals, In Fittkau, E. J., ed. Biogeo- graphy and Ecology in South America. Mono. Biol. 19( 2): 879-909. Skinner, M. F., and C. W. Hibbard ( eds.). 1972. Early Pleistocene preglacial and glacial rocks and faunas of North Central Nebraska. Bull. Amer. Mus. Nat. Hist. 148: 1-148. Slaughter, B. H. 1961. The significance of Dasupus bellus C Simpson) in Pleistocene local faunas. Texas. J. Sci. 13: 311-315. Stirton, R. A. 1935. A review of the Tertiary beavers. Univ. Calif. Publ., Bull. Dept. Geol. 23( 13): 391-458. Stock, C. 1925. Cenozoic gravigrade edentates of western North America, with special reference to the Pleistocene Megalonychidae and Mylodontidae of Rancho La Brea. Carnegie Inst. Wash. Publ. 331: 1-206. 1928. Tanupolama, a new genus of 11ama from the Pleistocene of California. Carnegie Inst. Wash. Publ. 393: 30-37. Strain, W. S. 1966. Blancan mammalian fauna and Pleistocene formations, Hud- speth County, Texas. Bull. Texas Mem. Mus. 10: 1-55. Sulimski, A. 1964, Pliocene Lagomorpha and Rodentia from Weze I ( Poland). Acta Palaeon. Polonica, Polska Akad. Nauk, Komitet Geol. 9( 2): 149-240. Talmage, R. V., and G. D. Buchanan. 1954. The armadillo ( Dasypus nouem- cinctus ). A review of its natural history, ecology, anatomy, and reproductive physiology. Rice Inst. Pamph. 41 (2): 1-135. Walker, E. P. 1964, Mammals of the World. Johns Hopkins Press, Baltimore, 2: 647-1500. 1975. Mammals of the World, 3rd Ed., 2 vols. Johns Hopkins Press, Baltimore. Weaver, W. G., and J. S. Robertson. 1967. A re-evaluation of fossil turtles of the Chrysemys scripta group. Tulane Stud. Geol. 5( 2): 53-66. Webb, S. D. 1974a. Chronology of Pleistocene land mammals in Florida. In Webb, S. D., ed., Pleistocene Mammals of Florida. Univ. Florida Press, Gaines- ville, 270 p. 1974b. Pleistocene 11amas of Florida with notes on the Lamini. In Webb, S. D., ed., Pleistocene Mammals of Florida. Univ. Florida Press, Gaines- ville, 270 B White, W. A. 1958. Some geomorphic features of central peninsular Florida. Florida Geol. Sury. Bull 41: 1-42. Wiegel, R. 1962. Fossil vertebrates of Vero, Florida. Florida Geol. Surv., Spec. Publ. 10: 1-59. Wilson, R. W. 1938. Pliocene rodents of Western North America, Carnegie Inst. Wash. Publ. 487: 22-73. 186 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 3 Winge, H. 1915. Jordfundne og nulevende Gumlere C Edentata) fra Lagoa Santa, Minas Geraes, Brasilien. E. Museo Lundii 3( 2): 1-319. Woodburne, M. 0. 1966. Equid remains from Sonoma Volcanics, California. Bull. So. Calif. Acad, Sci. 65(3): 185-89. Young, C. C. 1934. On the Insectivora, Chiroptera, Rodentia and Primates other than SinanthropUS from Locality 1 at Choukoutien. Pal. Sinica, Ser. C 8: 1-160. Zakrzewski, R. J. 1969. The rodents from the Hagerman Local Fauna. Upper Pliocene of Idaho. Contr. Mus. Paleon. Univ. Mich, 23: 1-36. Contributions to the BULLETIN OF THE FLORIDA STATE MUSEUM, BI- OLOGICAL SCIENCES SERIES, may be in any field of biology. Manuscripts deal- ing with natural history of systematic problems involving the southeastern United States or the New World tropics are solicited especially. Manuscripts should be of medium length-irca 35 to 200 pages (10,500-16,000 words). Examination for suitability is made by an Editorial Board. The BULLETIN is distributed worldwfde through institutional subscriptions and exchanges. It is considered the responsibility of the author to distribute his paper to all interested individuals. To aid in this the author( s) receives( s) 50 copies free, and he may purchase additional separates at cost if ordered when page proof is re- turned, The author is also responsible for any charges incurred for alterations made by him on galley or page proofs. The Museum will send an invoice to the author for this amount upon completion of publication. PREPARATION OF MANUSCRIPT Contributors should consult recent numbers of the BULLETIN for preferred style and format. Highly recommended as a guide is the CBE Style Manual, 3rd Edition, 1972 ( Washington, D.C., Amer. Inst. Biol. Sci.). MSS must be submitted in duplicate ( please no onionskin) and satisfy the following minimal requirements: They should be typewritten, double-spaced ( especially tables, figure captions, and "Literature Cited"), on one side of numbered sheets of standard ( 8-1/2xll in.) bond paper, with at least one-inch margins all around. Tables ( which should be unruled) and figure legends should be typed on separate sheets. All illustrations are referred to as figures. They must comply with the following standards: Photographs should be sharp, with good contrast, and printed on glossy paper. Drawings should be made with dense black waterproof ink on quality paper or illustration board. Ali illustrations should have a cover sheet, All lettering will be medium weight, sans-serif type ( e.g. Futura Medium, News Gothic) in cutout, dry transfer or lettering guide letters. Make allowance so that after reduction no lowercase letter will be less than 1 mm high ( 2 mm is preferred) nor any capital letter greater than 5 mm high. The maximum size for illustrations is 8-5/8x14 in. ( twice typepage size); illustrations should not be less than typepage width ( 4-5/16 in. ). Designate the top of each illustration and identify on the back with soft pencil by author's name, MS title, and figure number. Manuscripts and alt editorial matters should be addressed to: Managing Editor of the BULLETIN Florida State Museum Museum Road University of Florida Gainesville FL 32611