BLANCAN MAMMALS FROM HAILE XVA, ALACHUA COUNTY, FLORIDA By JESSE STEADMAN ROBERTSON, JR. A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1970 UNIVERSITY OF FLORIDA 3 1262 08552 3081 To Shirley ACKNOWLEDGMENTS Fossil material v/as examined from the 2\merican Museum of Natural History, the University of California Museum of Paleontology, the Florida State Museum, and the University of Kansas. I v/ish to thank t?ie persons in charge of these collections for the opportunity to study the materials dn their ci\re, and to eiclcnow ledge t]ic excellent work of Mr. Russell Parks in preparing the plates. I also wish to thank Dr. S. David V;obk\ Dr. 1. H. Patton, and Dr. 11. K. Brooks for their encouragement and help during the course of this study. .1 Ll TABLE OF CONTENTS ACKNOWLEDGMENTS iii LIST OF TABLES ......,.,. vi LIST OF FIGURES vi.ii ABBREIVIATIONS , X ABSTRACT. ^-^ INTRODUCTION. .......... 1 GEOLOGY ..... ...... . 6 SYSTEMATIC PALEONTOLOGY ....... ... 10 Oi'der Isectivora Cryptotis p.a2rv a ....,,. 10 Scalop-js aquaticus . . 3 0 Order Edentata Glo s so bh er i\-im ch ^tp a dina 1 e j i sis ........ 23 ^Z^^I^A^YA^^^-JL psiranensis . , .?! Review of the chlamyther lijiae 55 Evolution of the clilamytheres 60 Dasypus bellus , . 58 Order Lagon-'orpha Sylvilaqus S]j. 78 Order Rodentia Petauria sp. ............... 79 Castor £?n£iden_si_s 87 .^iSriQpoil iTiediuj 88 Order Carnivora Canidao . 90 Pteronura sp. . , . . , . 90 Smilodon gracilis 95 iV TABLE OF CONTENTS (CONTINUED) Order Prohoscj.dea Gomphotheriidro 97 Order Perissodactyla Nannippu3 ph logon 99 PJesippus simplicidens 99 Equus (Asinus_) sp 109 Tapirus sp Ill Order Artiodactyla Mylohyus f loridanus Ill IlGiniauchenia cf macrocephala 117 Odoco ileus virqinianus 122 AGE AND CORREIATION 127 PALEOECOLOGY 135 ZOOGEOGRAPHY 140 SUMI^iARY 143 LITERATUPJ5 CITED l-iG BIOGRAPIlIC'iL SI-Q^.TCH 155 V LIST OP TABLES Table P^c^e 1 Maminalian Fauna 1 List and Minimum Nuraljer of Individuals 11 2 Measuremants (in rw^) of tho Lower Dont.ition and Mandible of UF 17466 and Other Focsil and Recent Crypt ot is parva ......... 12 3 Measurements (in mia) of the Upx^er and Lov;er Dentitions of Three Species of Glosso- therium 24 4 Limb Measurements (in ima) of Glossotherium chapadmalensis and Glossotherium harlani . . 25 5 Measurements (in iran) of the Bones of the Manus and Pes of G loss other ium ch ap a dm a 1 en s i . s From Haile XVA and Glossotherium harlani . . 28 6 Cranial and Dental Measurements (in r\?r.\) of Kraqlievichia and Ch lamy ther i ura . ...... 39 7 Measurements (in mm) of Limb E^.ements of Kraqlievichia paranensis, l)h' ]0902, Haile XVA '. , 44 8 Measurements (in mm) of Limb Elements of Kraglievichia paranensis From Various Florid.Ji Localities [>3 8 Dental Measurements (in mm) of Dasypus bellus 73 10 Measurements (in mm) of the Femora of Fossil and !iDjas simplicidens Upper Dentition . . . 105 18 MyJohyus f loridanus Mandibular Ramus . . . . 113 19 Mylohyus floridanus Mandibular Symphysis , . 115 20 IlGmiau.chg^'iia cS. macrocephala Upper Dentition 119 XX ABBREVIATIONS Abbreviations of institutions and collections used throughout this work are as follov/s: AMNH: Ai-nerican Museum of Natural History U.C.M.P.: University of California Museum of Paleon- tology UF: University of Florida Collections UK: University of Kansas H.C.T.r University of Houston ColJ.ection X Abstract of Dissertation Presented to (ho Graduate Council of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Pliilosophy BLANCAN MAMMALS PROM liAILE XVA, hhACtlVh COUKTY, FLORIDA By Jesse Steadman Robertson, Jr. December, 1970 Chairman: Dr. S David XVehb Major Department: Zoology The Haile XVA fauna is the first Gu] f Cc£:ct-'.l Plain Blancan fauna to be described. AmoJiy the mananals, whic}i are rejjre- sented by 8 orders, 17 families, and 18 genera, arc two edentates which are new to the Nortli /American fauna. Gli:>s- sotherium chapadmalensis (Kraglievich) , a small South American ground sloth species, is reported from North America for the first time. This species was pre>/ious}y kno\\m from, the Chapadma.lalan (early Pleistocene) fauna of Argentina. It is proi-xised that th_is form may be ancestral to both C-:_^ harlani (Owen) and G^ robust us (Owen) . Kraglie-- vichia paranensis (Kraglievich) , a giant, extinct aimaoillo previously }cno\N^i only from Miocene and Pliocene deposits of South America, is reported for the first time in Ncith America. Ti)o Ilaile XVA specimen is described as a now .'jvb- species. Dasypus be.1 lus (Simpson), a smalJ.er extinct arma- dillo, is reported for tlio first time iii Ihe Blancan, mark- ing its earliest occurrence ciny\\'hore. xi A brief review of the Subfamily Chlamy theriinae is pre- sented, and it is suggested that Plain a (Castellanos) is a synonym of Kraqlieviehia (Castellanos) emd that Plof f stetteria (Castellanos) is a synonym of Ch 1 amy th or i um (Lund) . Chla- mythores are sho'v/n to have been present in North Ainerica continuously since the re-establishment of the late Cenozoic land bridvjc between North and South America, and it is proj30sed theit the Pleistocene evolution of these forms occurred simultaneously on both continents. The Old World Flying squirrel Petauria is reported for the first time in the New World. It v/as previously knov;n only from early Pleistocene deposits of Bavaria. Several of the Haile XVA taxa are closely allied with South American Plio-.Vleistocene forms, v^hich further strengthens the previously established correlation between the Chapadmalalan stage of South America and the Blancan stage of North America. The abundance of eiquatic non-mammalian vertebrates indicates that the environment of deposition was probably an open stream. The mammalian members of the terrestrial comm>unity include the tropical or subtropical indicators Iuna is as yet undescribed excei^t for the description of 'J'itanis, a 1 cfiant, flightless 'bird (Brodkorb, 1963). A faunal list for this locality is provided by Webb (in press a). Other Florida deposits whicli are considered to be of Blancan age, but which have not been studied, are Santa Fe IV and Santa Fe VIII, Gilchrist County. Correlation of Florida Pleistocene deposits is diffi-- cuJ.t at best, because of their mode of deposition. The two most common types of Florida Pleistocene fossil vertebrate deposits are filled sinlcholes in limestone and river bottom deposits. The various fissure fillings have trapped sedi- ments representing many steiges of the Pleistocene, and adjacent sinlcholes may contain fau.nas which vary greatly in age. River bottom deposits are usually heterochronous accumulations of bones which have eroded out of fossilifer- ous strata and settled at low energy areas of the stretim. This lack of orderly stratigraphic sequences for m.ost of tlie Florida fossil vertebi-ate deposits necessitates the use of faunal correlation. Usually, a reasonably accurate age assignment can be made based ujjon the presence or absence of certain genera arid species. Ti'ie presence of Bison, for example, indicates a Rancholabrean aoe (Savage, 1951) . Further refinem.ent of the age could be made by determining the species of Bison present. The: presence of B,_ latifrons v;ould represent early Rancholabrcan, B^^ antiquus late Rancliolab.rcan , and B_,_ bison sub-Recon L or Kecent ( Ro))c;rtson , in press). In addition to iaunal correJ.ation, the effects of sea level changes due to Pleistocene glaciation may be eiajployed as an aid to dating Florida Pleistocene deposits. Glacial eustatic fluctuations of soa leve.1 have caused "terraces" or "abandoned shore lines" which are doiainant geomorphir- features of t}ie Florida landscape (Cooke, 1945, MacMeil, 1949; White, 1958; Alt and Brooks, ]965}. The location of a deposit relative to these structures can be quite helpful in age deterniination. If, for example, a coastal terrestrial deposit is presently lying at or below sea level, then it must represent a time when sea level was as low oj" lower than it is now. This would be an indication that Lhe de- posit v/as laid down during glacial tii.ies, as tliese v/er.e tl'e times of low sea level. Once this is determined, faujial elements will indicate which glacial stage is represented. Inland deposits of higher elevations present a probloTi in that they could have been deposd.ted dui-jng eilljer a glacial or interglacial stage. If, however, it can be sho.vn tlval a particular inland deposit v;as cotiStal at its time of deposition, then it must represent an interglacial stage, v;hen sea lovv.l was high. As a rule it is dilficrJt to as-- certain whether or r.ot a terrestrial fauna repreoonti: inland or near-shore conditions, Auffcnberg (.1953), liowever, has been able to do this for certain depofiits. Working with fossil Ter3:apene Carolina, he has shown that particular intergradient populations of this species reflect near- shore conditions. If members of these intergradient popu- lations are present in an inland deposit of higher ele- vation, the deposit probably represents an i)'iterglacial stage. Evidence for near-shore conditions is not always this conclusive, however, and sometimes one is forced to rely upon more tenuous means. Of particular interest among the Florida fossil verte- brate deposits are those near the settlement of Haile, Alachua County, conveniently located only 15 miles from the University of Florida campus. In a radius of only a mile I're clustered numerous vertebrate localities, traditionali.y designated by roman numerals in the University of Florida vertebrate paleontology collections. Most of the Haile localities have produced fossils of Rancholabrean age. Notable exceptions to this are localities V and VI which are coijsidered to be Pliocene (Auffenberg, 1955) and locality XVA v/nicli is regarded as Blancan. The fossil vertebrate assemblage recovered from the IJaile XVA locality constitutes the subject for this dissertation. The Haile ZVA locality is situated on the property of Parkex- Brothers' Limestone Products, Inc., near Haile, R17E, T95, S25, Alr.chiui County, Florida. It was discovered Jn 1964 by Mr. Piiillip Kinsey of Jacksonville Beach, Florida, who has loncj been a friend of, and a contributor to, tlio Florida State Museum. Mr. Kinsey did extensive collecting at Haile XVA, and, upon recognizing it as an unusual local- ity, brougjit it to the attention of the museum staff. Further excavations were carried out in the same year by Dr. S Ddvid Webb, Mr. Robert Allen, and the present author, with the support of NSF Grant GB 3862. The resulting collections, in addition to the large quantity of bones so generously donal-ed by Mr. Kinsey, are now housed in the University of Florida Collections. In addition to the hiami.iialian fauna described in this report, trie locality has also contributed extensive fisli and reptile faunas. A fair amount of amphibian and avian re- mains is also present. Among the reptiles are the type and abundant me-.terial of Chrysemys platymarginata (Weaver and Robertson, 1967). GEOLOGY The Haile XVA deposit, lying at an elevatjon of about 90 feet above sea level, is a filled sinkliole in the Ocala (Eocene) Limestone. The deposit extends appi-oximately 30 feet along its east- west axis; its north-south dimension is narrower and irregular in width. Underlying the fossil-bearing matrix is a layer of darlv brown, unf ossilj.f erous hardpan which begins at a depth of approximately 20 feet and extends to a depth of at least six feet from its upper m.argin (Figure 1) . The fossil-bearing matrix extends from about five to 20 feet belov/ present groiand level, and consists of tv/o units: an upper layer of heavy, greenish clay; and an underlying layer of coarse, poorly sorted, gravelly sand- stone, containing lenses of clay. Wliere these two units merge, an alternation of clay and sand lenses occurs. The lowermost massive sand is the most produ.ctive part of the fossilif erous sequence. It contains much cailcareous cemicnt and calcareous ceme^ited concretions. Many of the larger bones x-ere en/:"rusted with cemented sand and gravel. The 6 Figure 1 Geologic Section at Haile XVA A. Sandstone alternating v/ith gray clay B. Heavy greenish clay C. Coarse, poorly sortc^d, gravelly sand- stone D. Dark brov'j. hardpan E. Ocala Limestone (Eocme) 8 ^ sandstone and clay v/ero shown to be essentially contempora- neous hy the occurrence of fossils which extended from the sandstone up into the clay. Overlying the fossil if erous sand and clay sequence is a layer of thnn, brown sandstone alternating with greiy clay, This layer is unfossilif erous . Since the oxiginal collections were made, mining ac- tivities have scattered the remaining matrix over a wide area. On a recent trip, for example, a tooth of Glosso- therium belonging to the associated skeleton taken from the original deposit in 1964 was found on one of the ad- jacent piles of matrix. The site is presently covered by several feet of sand washed in from the overhanging lime- stone ledge. There is little, if any, of the matrix re- maining in the original site. SYSTEMATIC PALEONTOLOGY There are 20 taxa representing a niiniiuuia number of 34 indiviciiuils in the Haile XVA fauna (Tcible 1) . These cire considered on the follov/ing pages. Order Insectivora Family Soricidae Crypt Otis parva Say Material: UF 17466, right mandible with M.^-M^. Thj.s specimen j.s morphologically indistinguishable f/cm specimens of Cj_ parva from the Irvingtonian Coleman IIA and from recent specimens. The Haile XVA specimen is Sj.igl.tly larger than the previously mentioned specimens (Table 2) , but not significantly so. The only other Blancan record of Crypt Otis is the Rexroad fauna (Hibbard, ]937, 1941), Meade County, Texas o Family Talpidae Ecalopus c^quaticus Linnaeus 1758 2 Material: UF 17466, 2 left M ; UF 17465, right humerus, A comparison of the three Haile XVA specimens with corresponding modern S^ aquaticus from the southeastern 10 11 Table 1 Mamma] ian Faunal List and Minimum Number of Individuals CryptOtis parva 1 Sea] opus aquaticus 2 GlossotTierium cliapadmalensis 1 Graqlievicl-iia paranensis 3 Dasypus bellus 1 Sylvilaqus sjp. 3 Petauria sp. 1 Castor canadensis . 1 Siqmodon medius 2 Canidae 2 PtGX"onura sp. 1 Smilodon gracilis 1 Gomp]iotbcriidae 1 Nannippus p]ileqon 2 Plesippus simplicidens 2 Equus (Jvslnus) sp. 2 TapiruG sp. 1 MyJ.oliyus f loridanus 1 Hemiauc]ienia cf meicrocepliala 1 Odoco ileus yi rqinianus 5 12 o U o -P o O rj 0 > m -rl o c D- fc > u u u o -p « •H CJ -P u r-i as Q 'd 01 rj :§ o (^ di -P O -H ■t-J a 0) o w CO •H CfJ Vi 0) CO O o Ix: •r1 c; w o Ul o CO O ■ r-- I r-l CO *X) I CO r- 00 I U-) LD CO I CN a (U g nj M 0 m H rt -p c Q) rH CN! E s S 0 0 o ^J +J +' T> 'Ci ■•ci •rl f 1 •H C) o 0 r-l r-H ; — [ >-, >1 l^ t:! TJ •0 C f,' c. o O r-. u U u C^J f^J • f') I LT) • CM (O » I CM CO r-- 1.0 c^. • • t • LT) (N CO o; O CM • vj' O • r- •si' • « UD r- iTi Cv! CO CM 4.- P4 1 1 ! I 1 1 flj O-i r-- U) NJ* ro rH > • • • • < 4 I M O in ijD in CM f^ C^j fO o^ H 01 r-- ^ '^' C\' • IW t • ■ • • • U VD VD in CN OT CM CO X~- ro CO o m • t t • • ♦ KD KD in CM •s}' Ci m CO U P-. •?: CO 1 i<') t 1 r-J ^^ 4-1 U I ■p m rtf CO +J 0 CO fO GJ •J: 13 United States reveals no s.lgnificaiit rlif f ereiicos in size o r mo r ph o 1 c gy . Scale pus is also present in the Blancan Rexroad fciuna of southwest Kansas (nil)bard, 1941) . The only othc^r mole from the Rexroad fauna is Jlesperoscalops . wliicli is Ijased upon a partial lower dentition. A direct compcirison could not be made with this genus as only upper teeth are known from Haile XVA. As Hibbard (1941) points out, however, mole humeri of varioiis genera can be distinguished from one another, so it would be illogical to assume that a Scalopus- like humerus would belong to another genus. Because of this, and since Scalopus is knovm from other Blancan and Irving- tonian deposits, it seems reasonable to ascribe this ma- terial to the living species. Order Edentata Family Mylodontidae Glossotherium chapadmalensis Kraglievich 1925 Material: UP 10922, partial skeleton There is no doubt that the Haile XVA si^ecimen repre- sents one individual as there is no duplication of elements, and the ]eft and right elements agree very clo^c]y in all measurements. Parts of this specimen were collected by several people r>ver a numbe.v of years, but the bulk of the 14 S2:)Ccinion was taken from the sand layer of the fossilif erous sequence and was semi-articulated. In the following discussion, terminology regarding sloth dentitions v;ill be based upon that used by Hirschfeld and Webb (1968): the first tooth (upper and lower) will be referred to as the caniniform; all the rest of the teeth will be referred to as molariform. No homologies are im- plied. Upper dentition: The upper dentition of the Haile XVA specimen (Figure 2) is complete except for the left upper caniniform. Tlie caniniform is triangular in cross-section and we] 1 developed. As in the Argentine specimen of G. chapadiiialensis , this stoutness of the caniniform is re- flected by the expansion of the anterior portion of the maxilla. The anterior portion of the m.axiJ.la is very much reduced in G_^ harlani, which tends to have a reduced or lost caniniform. Tlia first upper molariform tooth is large ajid oval in cross-section. As in G^ harlani and G_^ robust us it is the longest anteroposterior ly of all the upper teeth, but is relatively narrow in transveise diameter. Due to the very well-developed tower caniniform v/hich in part occludes wJ th this tooth, the anterior oblique v/ear surface is much, more accentuated, than, in G^ harlani. In several specimens of Figure 2 Glossotheriuin chapaJinalensis Upper Dentition X 1.13 (linear) 16 17 G. robustus observed, this tooth wc"is worn off smoothly just above the alveolus. A very short diastema separates the caniiiiform from the first molariform tooth in the Ilailc >;v^A specimen . The second u]3per moDariform tooch of UP' 10922 is tri- angular in cross -section, with t]io base of the triangle placed lingually. The anteroposterior distance is less thyn the transverse diameter in this species, while the opposite is true in G_j_ harlani . Of the tv/o specinens of G_^ ^Qbl.isj;u_s for wliich measurements are provided, one agrees with G^^ chapadma 1 en sis v/hile the other is ne?.rly equal in both diameters. The third upper molariform tooth of UF 10922 js similar in shai)c to the second and agrees morphologically v/ith the corresponding teeth in botli G_^ harlani and G^. robustus . The fourth molariform tooth is relatively small compai-ed to the Argentine specimen of G. chapadmalensis al- though the shape is the same. The ]-)alate is very much conrtricted in the area of the Icist upper teeth and very much expanded at the anterior end of the maxilla. Both of these characters are more extreme in UF 10922 than in th.e Chapadmalalan specimen, or in speci- mens of G_^ harlani and G_^ robustus. Lower dentition: The lov/or caniniform of UF 10922 (Figures 2, 4) is:- very well developed and directed labially. Figure 3 Glossotherium chapadtn.alensis Right Mandible, Lateral View X 0.82 (linear) 19 Figure 4 G 1 o s sother ium chapadmalensis Right Mandible, Occlusal View X 0,82 (linear) 21 22 Tills tootii is relatively slightly largei: than the correspond- ing tooth in the South American specimen of G_. chapadmalen- sis . The tooth is pointed when viewed laterally due to the dual occlusio)! v/itli the upper caniniform cind the first molariform tooth. The first lower molariform tooth is also ro]atively robust. It is somewhat triangular in cross-section and pro- trudes fairly high above the alveo.lus. ':Lh.i s tooth is also turned slightly to the outside and closely resembles tlie lower caniniform. The second molariform tooth is rectangular and is turned lingually, so that its long i\xis does not parallel the tooth row. Stock (1925), in discussing this tooth, lised the long axis as the anteroposterior length while Kraglievich (192-.') used the shorter axis of the tooth for this measurement. For comparison; Kraglievich ' s method of measuring is used here for the Haile XVA speciiaen and Stock ' s method for measuring G^ robustus . It v/as necessary to reverse the measurements of G^ robustus provided by Kraglievich (1928) so they would be consistent with the other material con- sidered. Thus, the seemingly great difference in the meas- urements of the second molariform teetli is actually an ex- pression of the fact that the orientci:ion of the teeth is different in the species and does not indicate gross morpljolcgical differences in the teeth. 23 The third lower niolariform tooLli is of a slightly dif- ferent shape in G_^ chapadmalensis . Tlie two main coluinns are separated by a thin bridge, whale in G^ harlani and G. robustus this bridge is thicker. Mandible; Unfortunately, the mandibular symphysis is not preserved in the Haile XVA specimen. The portion cf the mandible present agrees in morpliology with the other two species. Measurements of the upper and lower dentitions are presented in Table 3. Cranial material; The remainder of the Haile XV7i cranial material consists of part of the occijpital bone and portions of the left and right periotic bones. Tlie occipital fragment contains the left condyle and a portion of the otic region. The hypoglossal canal, jugular forawon, and auditory meatus shov; no observable differences from those of G. harlani and G_^ robustus . A portion of the left squamosal is attached to the occipital. Part of the right squamosal is present including the zygomatic process. Vertebrae: The atlas of UF 10'.322 is well preserved. According to Stock (1925) this element differs in Gj^ harlani and Gj_ robustus . The posterior portion of the lateral process does not extend as far forward in G_^ harlani. The location of the dorsal foramina also varies: In G_^ harlani they are quite far apart while in _G_^ robustus they arc 24 o 4-1 C. E 0) M m •H 4J (1) w U rt: C) ■H U q; D, CM UJ •rl fd G •rl (L) fO C a) r-j M M J^ 0 cp; ro M-l ^ rC •f-I W-i t (0 o O u w c o •H 1 M 4-' n r "5 •H fVi CO q ■P P- -H ■H f.' 0.1 (Q 4J Q) rc: (i; s:; Pi u (1) rH (i) b-1 U f; 0 fU' U ^ O g < > ■r! O U ^:i (i) 'd ■IJ 1 C 0 'd fs: (« 'O m W 10 p •r~\ P-i VI 0 fd 10 0 rH rt"! S-1 0) Ph o u 0) r-l a, r-l •H ::3 • fd fd o e M (D '^ 4' m o (T\ C-i (■') f^ CO O 0") -si' 0-) CM r^ CD f\l 1.0 CNi CO O CN fO O^ rH rH CM t-4 CNJ CN! CM CN ON H CTi O ro ^ H O >X) CO OT ^ r^ kD n r-^ r-- o") rs) oT '^ 0> rHrHCOrHCNjOgCNJCMCNJrH OLT, OOlOOOOOO ^ . . . . HiHCNrHcMCMrHrMfNlH ';^oc^jCNlnf^;r-Or^^o rHtHCMHrHCMiHC^OOH CM n ^ rC ■c; .a rC 4J 4J 4-" 4J O o O o o rH 0 CNi 0 ro o sf 4J 4-> 4-> 4J r; ,c rC' -C rC o V 4-) - 4-> .. 4-' - 4-1 -H 1 — I o ,£: O rC O rC O 4S 4J o 4-1 O 4-! C 4-) o -H tr. 4J b-i 4-" bi -P tn 4-) 4J C t^, r; G C m >. 0 •. a; V 01 . Q) rH rC rH r^; rH rC rH ^ P 4J 4-» 4) 4J M tj. M tn 5-i dl i^-l bi Sh O ri o c o (^ o t; (U ■H o ■H 0) •H W -H (U 0-, M c-l Vi t-l M r-l ^-1 r-l o< 0> Q) 0 Q) p 4-i 0) •P 0) 4J (D P ^"i W m 10 m m CO CO C/i O u 0 ',^ o V! o Jl P^ Q) a, Q) p., a.) a; 0) O > o > > o > ^■1 If) !i '.0 C CO u to c; C. O C b r; 0 C 4J fd -'.J (d 4' iTi ■\-> fO a S^l C J; r"* M s. !4 <; Eh <: Th '5 L-! Fh in 4-1 O O LD 4-' - 4-1 rC O 4-) O tP 4-) (D V rH ^ 4-' O C •H Q) U vH 0) .p Q) CO CO O Vl P^ 0 o r> M C/) 4-1 fd « ^ «;1* O CO CO C3 O CNj I'M o in fo CM Lo 0"^ o o CM t-l CM CM CN rH in CM ^ ci CO If) CO CM a> CO r-l IX) CO CvJ CD CXI rH ^) CNJ r-l CM Cvl (M rH U1 CM LC) m o o o o o o ^t O CO CM "vf in, 01 CO rH r-l H CM rH CM ^' H in ^t '■~- in m in c) co OCMOO-N^-xfOOl-^ CM rH rH rH r-l CM "^- rH o 4-J o o 4-1 4-1 o c ■H 0) •H ^ O 4J 4-; f3 •H tri 4-1 4J G O H ^^ P 4-1 U 54 (D > U . , O 0) h^ 4J CJ 0} CO O 5; D^ (U O > 5-1 m (U fi; -P fd C S4 CM rC 4-1 O +1 - 4-» ,C O 4-1 O tn 4J G 0-) rC •p O O 01 ^0- 4-1 o O Kf r^ 4-1 O O ^ (1) 4J 5-1 In O C' 0) Q) -P en tP 4-1 a rH ,C i-i tn O i:^ H (!) -H C) - 4-) rC O 4J O tn -P (1) -. 4J Si Ct' O i:; O 54 54 -p d) m C.0 O ^ (U Ph (1) p^ C) o > 0 > 54 CO 54 CO (L' ^1 (U C -1-J Id -P fd ^ 54 t-l S 5-1 E-i CO m O !-) P^ (U O 54 0) 4-1 54 E^ IT, ^c, iri .— , '; c^j CM 01 sj' r-4 (X> •r^ > -rH rH ,y fd o 0; 5h -p :< f< CO o 54 54 54 0) (L) 0! 4J .|J .|J iM m U-: fd ftj fd CO CO CO -p 4-1 4 J a rl r; CD (U QJ f= !^ P ;-i (-J f-i Q) S (U 54 U 54 r^ r^ P3 CO CO CO fd fd fd D 0) 0) ERE .H Cx! CO 25 clorje together. \JF 10922 definitely agrees with G. harlani in the expression of these tv.'o characters. The remainder of the vertebrae preserved are, for the most part, in poor condition, and as a jresult their posi- tions in the vertebral column could not be determined accurately. Front limits : Neither scapula of UF 10922 is present. The right humerus is present. Neither radius is proser-'^cd , although both ulnae are present. The left uJ.na is so poorly preserved that it could not be removed from the plaster jacket in which it was co.llected.. Several bones of the manus are present in the Haile XVA material. The cuneiform appears to be somev.'liat different ^-^ <^ii. chapadmalensis than in G^ harlani . This element in UF 10922, when viewed from the palmer aspect, is square in appearance as opposed to being rectangular in the figure of G- harlani provided by Stock (1925, Figure 72d) . These differences are reflected by the mefisurements of this ele- ment in Table 4, In UF 10922, the proximal-distal distance is greater than the distance across the ulnar articular surface while the opposite is true for 39 specimens measured by Stock. The c\ineiform appeax's to be a deeper element in G_. chapadma 1 en s i s than in _G_^ bar 3 ani . The remainder of the elements of the manus preserved in the Haile XVA specimen 26 Table 4 LinilD Measurements (in mm) of Glossotherium chapadmalensis and Glossotherium harlani Humerus Greatest anteroposterior distance of head Greatest widtli of shaft at deltoid ridge Width of distal articular surface Ulna Greatest length Width of distal articular surface Femur Total length Transverse diameter of head Least width of shaft Greatest width across diste\l tuberosities Width of distal • condyles V/idth of intercondyloid space- Width of inner condyle VerticQ] extent of inner condyle Tibia Total length Greatest width, proximal end Greatest width, distal end Antero]:)osterior distance, dii-'.tal end Fibula Tota] length Width, proximal end Width , distal end G. chapadmalensis G. harlani UF 1 0 9 2 2 , H aile XVA Ran cli o La Br e a left right 70.0 70.0 71-3 86.0 285 34 355 79.3 99.5 140 111 27.1 47.2 47.2 67 69.8 181 176 114 90 88 64.8 66.6 183 50.5 39.5 123, .8 122, .6 133, .2 395, .9 55, .0 545, .4 127, .8 164, .6 234, .8 188, .5 4 7 88. ,7 120 24 7. ,3 185. 2 142. 5 101. 9 263 103. 6 73. 5 After Stock (1925). The measurements represent the mean of large sampler,. Th::; number of sjDecimeny varied from element to elt;m£:nt and in seme cc'ses was not aiven. 27 arc the left scaphoid, right pisifoxm, Ic^ft metacarpal 111, and ]oft phalanx 11, digit III. Hind Liniljs : The pelvis is missing in UP 10022. The left femur is complete while the right is badly fragmented. Both tibiae, and the left fibula are well preserved. Several elements of the pes are present. These are the right calcancum, right astragalus, left and right meta- tarsal III, left and right phalanx II, digit III, emd an ungual phalanx, digit III. A comparison of the 3 imb elements of the Haile XVA specimen with those of Gj_ harlani frora various Rancholabrean deposits of Florida and with the figures of G_^ harlani pro- vided by Stock (1925) and figures of C_^ robustus (Owen.- 1842) shews, for the most part, no great morphological differences. All three of these species appear to be very similar postcranially . Comparative measurements of UF 10922 and Stock's (1925) Rancholabrean material show the Hai.1e XVh materia] to be significantly smaller (Q^ables 4, 5), Glossotherium robustus is larger, about the size of G. harlani . Discussion: There has been much confusion concerning the taxonomy of the genera MyJodon and Glossotherium. Krag- lievich (192C) has presented a thorough review of the literature. ''"te concludes that My] odon darwinii is tlie geno- 28 Table 5 Measurements (in mm) of the Bones of the Manus and Pes of Glossotherium chapadmalensis From Ilaile XVA and Glossotherium harlani G_^ chapadmalensis G. harlani^" left right scaphoid Greatest distance across articular surface 46.2 75.3 (39)** cuneiform Greatest distance across dorsal surface from inner side to outer Greatest proximal-distal distance Distance across articular surface for ulna pisiform Greatest length Greatest depth Greatest v/idLh Metacarpal III Greatest length Width, proximal end Width, distal end Depth, proximal end Plialanx II, digit III Length Depth, inner condyle Width, proximal end Calcaneum Greatest width, anterior end 57.4 91.8 (21) Greatest depth 83.4 125.4 Astragalus Anteroposterior diameter 76.1 140.2 (41) Metatarsal IV Length Widt]i , proximal end Depth, proximal end Width, distal end Depth, dista] end 42.1 68.5 (42) 34.3 50.5 31.2 G1.3 25.3 5 5.6 (22) 18.9 36.4 20.9 34.7 71.7 103.1 (30) 51.9 75.2 30.5 50.9 42.0 66.0 40.5 48.9 (42) 23.5 39.9 20. 2 47.9 84 . 3 119.3 (29) 41.2 41.5 60.7 34.3 37.1 45.5 38.5 50.1 35.3 43.5 29 Table 5 (continued) Phalanx II, digit III Length Depth, inner condyle Width, proximal end Ungual phalanx, digit III Length 104.4 174.1 Proxinial-distal distance of ungual base 41.0 73.5 V7idth, proximal end 32.4 55.0 G. chapadmalensis G. harlani* left right 30.2 31.0 35.4 (32) 20.4 21.5 21.8 22.6 22.5 35.1 ^measurements after Stock (J.92 5) **nuu>ber in parentheses refers to the sample size from. Vv'hich the mean (the numbers in the right-hand column) was calculated 30 type of Mylodon, and that what Owen (1840) described as i!?iLlPil211 robustiis more correctly belongs to a distinct genus, Glossotherium. However, Kraglievich retains Paramylodon (Brown, 1903) as a valid genus oven though Stock (192 5) shov;ed that the type of Paramylodon merely represented a vari- ation of what was then called Mylodon harlemi , Simpson (1941J) argues that if the North and South American forms (late Pleistocene) are not generical]y distinct, then they all belong to the genus Glossotherium. This view is follov/cd here. Hoffstetter (1952) retains Paramylodon as a subgenus of Glossotheriun, recognized by its tendency to lose tlie vipper caniniform tooth, and because of their narrov; muzzles and long sl^ulls. The partial, associated skeleton from Haile XVA is clearly a member of the Subfamily Mylodontinae as indicated by the single astragalar facet for articulation with the ■ calcaneum. Following the above system of classification, it is a member of the genus Glossotheriumc This si:>eciiAen so closely resembles Eumylodon (<;lossotherium) chapadmalen- sis that it is here ascribed to that species. Glossotherium ch a p a dm a 1 en s i s is an early Pleistocene ground sloth from the Chapadmaj.al fauna of Argentina and is here reported from North America for the first time. It is a small species which appearn to be very closely related to both G. harlani 31 cind G_j_ robusLas , differing froia these two forms in the nature of the anterior dentition and in overall size. It agrees v.'ith these two species in the morphology of most of the postcranial elements with the possible exception of some of the bones of the manus and pes. As this species is T<:nown from the early Pleistocene of Argentina and Floridti it is thus temporally, geographically, and morphologically, a plausible ancestor of both G_^ harlani and G^ robustus . Fami ly Dasypodidae Kraqlj.cvichia pa^ranensis new subspecies Holotype: UF 10902, partial skeleton collected by P. E. Kinsoy, S. D. Webb, R. R. Allen, and J. S. Robart5ion in 196-0 . Type Locality aiid Horizon: Ha lie XVA , R17E, T95, S25, Alachua County, Florida, Blancan. Sub;:pecif ic Diagnosis : A member of the species If.th and ninth teeth are present. The fragmentary nature of the two jaws makes them difficult to compare in detail. The major differences between the mandibles of Kraglievichia and Chlamy ther ium, according to previous v.'ork.ers, are in the shape and orientation of tlie anterior teeth. Unfortunately, this portion is lacking in both mandibles of UF 10902. There appear to be no differ- ences in the po?. Lerior portxon of the mandibles except fo-^ 39 a o (I) r-- lo vD 0> •a • • t • • 03 •.-1 0> C5^ ^.O O f>) CO o fO fM iH fNi CM Cm r-l IJ t-M •H M (L' >X) rTJ CO -IJ vo ra ;> (N 'd fc; ■H 1(5 i S^ rC r-t u < Cm r; <; O I t^ IX; IX) CO I ro CM r- CM I CN ON IX) (N "^ ^i) r- o LO r^ c\! CM in (XJ r-l r-\ rH rH ^' ON. > o X O 0) CTi O n -sf %j' 00 in in (n; in o KD H r-: r^J r-H -H r-l 1 -1 •H Cm tC ^ K ^ IT) r^ rC ,c ,c; .-) Hi -P ■jj •iJ o M O o O 0 CD o o 0 o > •H >1 u -IJ ,j,i -IJ -H M VI 0) ni (U lO +J •■ V •h ■(J rH t:; rC .c ,£1 rfn c r-l a; ■IJ ■i-) -IJ -jJ fd •H u bi c en •rl -IJ rH u H 0 r— I (-1 S iH > fd r-i •m w ^ o M-l H fO >X) r- rO 0 rC i-i rC f) ,c ,C f^ CO O rH a\ ai G -P rC •P ,C -IJ •IJ r*-t rc; rC r^ 0 o •P 0 •';' 0 o •P -p Si -IJ -IJ •iJ c O o o o 0 0 o 0 O o -p -IJ o •p o -!J -p o o -H O 0 f) -l-J •!> +> 4J M A-> -p rQ - *» ^ K (1) rC » rC ^ ,c .c •t « -p •. » 0 4J .c: ■P r«"; -JJ 4J ^ ^ c rf; .t"; ■l-J tr> •p tn .l.l 0> D-« -IJ 4J rd -IJ -•J t; 'd t,' 'd t; C, Xi ncs 'U nj (U •rl 0) •1 I 0) Q) •H •H •rl rl .^ ^ ^-^] jj: (^ h:) ;^ ^ r^ IM a\ 1 — 1 •w to o CI !TJ rH H 0) -P U) fO u fH 0) 4J m Pi w ■p c 0) •rf B (U Q) •(J >-l rd 3 P. '/I •i-i rO 4J H (N 40 size. Meas\irements of the lowej- dentition aj-e provided in Table 6. Vertebrae: Tlis vertebral material from the Haile XVA ske]eton consists of seven thoracic, five ]uirba3", and four cavidal vertebrae. Five of the thoracic vertebrae are cemented together: in the proper sequence by a coarse, large grained sandstone. Comparisons of these specimens with the vertebrae of an excellent Ch 1 amy ther ium skeleton from Branford lA, Suwannee County (Rancholabrean) , show no sig- nificant m.crphological differences except for size. Front -limbs : In discussion of the major limb element", a comparison is a]. so made with Dasypus . Kraglievichia is approximately the same size as Rancholabrean sxjecimens of Dasypus bellus, a large extinct Pleistocene armadillo. Conceivably, limb bones of the two could be confused. The right humerus of UF 10902 is well preserved except for the idsta]. end (Figure 6a, b) . Tlio humerus of Krag- lievichia is greatly expanded laterally (as seen in the specimens from Saiita Fe I) , The supracondylar foramen is relatively larger than in Dasypus . The articular surface for the radius is concave in Kraglievichia, vv'hereas it has a sj.ight convexity in Dasypus . The supinator ridge is rela- tively narrov/er in Kraglievichia, and the deltoid ridge is wider. The proximal end of the humerus is similar i.n tlie Figure 6 Kraqlievichia paranensis Front Limb Elements A. Right humerus, medial vievv? B. Right humerus, cinbcrior view C. Left ulna, medial view D. Left ulna, anterior view E. Loft radius, posterior view F- Left radius, anterior vievv' X O.GO (linear) 42 B D 43 two genera . Comparison between Kraglievichia and ChJamy- therinm shows no distinct differences in this element ex- cept for size. Tliere are three ulnae from t]ie Haile XVA site. Two of these specimens appear to be from the same individua] as tliey are similarly preserved, are of opposite sides, and agree very closely in measurements (Table 7) . The ulna of Kracfljevichia (Figure 6c, d) is laterally flattened and possesses a long olecranon process. The articular facets for the radius and medial condyle of the humeius are a single structure in this form, while in Dasypus they are part-ially divided. Anoth.er striking difference is a lateral gioove which runs the entire length of the ulna in Dasypu;; , but which terminates at the upper boi'der of tlie semilunar notch in Kraglievichia. Size appears to be the only dif- ference between t.he ulnae of Kraglievichia and Chlamytherium. As in Dasypus, the distal end of the radius is massive compared to the proximal end although it is less flattened in Kraglievichia (Figure 6e, f.) . The proximal portion of the shaft is relatively thic};er and less curved in Kragli- evichia . Comparison witli Ch ].amytherium shows only a size difference. The loft metacarpal II is preserved w.ith the Haile XVA skeletoij. in fjoijei al 3Tia]:^e and proportion it agrees with 44 Table 7 Measureraents (in mm) of I.inib Elementt"; of Kr ag 1 i e v i ch i a paranensxs, UF 10902, HaiJe XVA 29.0 17.2 16.1 25.5 25.7 29.4 30.5 Left Riyht riuiiierus Lateral width, proximal end 36. 7 Anteroposterior width, proximal end 36.8 Greatest anteroposterior diameter of shaft 25.5 Lateral width of shaft at sane location iR.2 Distance from proximal end to top of entepicondylar foram.en 100,5 Ulna Total length Lateral width, proximal end Lateral width, distal end Laterfil v/idth at semilunar notch Anteroposterior width at semilunar notch Radius Total length Lateral width, proxim.al end Anteroposterior width, proximal end Lateral width, distal end Anteroposterior width, distal end Metacarpal II Total length 32.8 Width, proximal end 10.9 Depth, proxj.mal end 14.5 Width, distal end 10.5 Depth, distal end 11.2 Metacarpal III Total length Width, proximal end Depth, proximal end Width, distal end Depth, distal end Ungual Phalanx, Digit V Total length 29.1 VJidth, proximal end 11.4 Depth, pioximal end 10, 89.2 87.2 22-6 22, 3 11.7 11.4 23.0 22.3 17.4 16.5 34.4 36.3 13.8 14.5 12.8 13.0 12.6 1 3 . .1 10.5 11.0 o Tt'iblc 7 (continued) Left Right Femur Total length Lateral v/idth, proximal end Anteroposterior thickness of greater trochanter Lateral width of shaft at third trochanter Anteroposterior thickness of shaft at same point Greatest lateral width of articular facets Greatest anteroposterior widtli, distal end Tibia and Fibula Total length Lateral width, distal end Anteroposterior width, distal end Calcaneum Total length Width of articulexr facets for astragalus Detpth of facet for cuboid Width of facet for cuboid Navicular Greatest lateral width Greatest anteroposterior distance Greatest depth Metatarsal II Total length Width, proximal end Depth, proximal end V7idth, distal end (articular surface) Depth, distal end (articular surface) Metatarsal III Total length Width, proximal end Dep'h , proximal end Width, distal end (articular surface) Depth, distal end (articular surface) Ungvial Phalanx, Digit V Total length 24.6 Width, iDroxinial e!id 17.6 Depth, proxiraal end iO,7 193.5 60.9 4 ]. . 7 41.3 23.7 48.4 51.5 -._. 121.0 46.7 — 25.1 63.3 28.0 12.2 3 .1. . 5 ...... 39.5 23.3 22.5 33.2 32.8 12.2 12.5 13.8 14 , 3 13.2 13.7 11.3 10.2 36.5 34.5 15.0 14.9 14-7 14.4 13.2 12.3 46 the same element of Ch 1 amy th er ium . The facet for articula- tion witli the trar^ezoid, however, is qnite different. Wheii viewed laterally, this a'acet is a smooth curve in Ch la my- th er ium, while in Kraglievichia it forms a sharp, V-shaped indention. The facet for articulation with the magnum is ob- long in Ch lamy ther ium , while it is round in Kraglievichia . The facet for articulation with the trapezoid is wider dor- sally in Ch lamy ther ium . This extra v/idth gives the facet for metacarpal III a different shape. In Ch lamy ther ium it bulges out proximal ly to form a quarter diameter for the facet v/r-iic'ii articulates with the trapezoid. It appears that less of the trapezoid articulates with metaci\rpal III in Kraglievichia , Left and right metacarpals III are preserved in UF 10902 >. In this element the facet for articulation wj th m.etacarpal II and the trapezoid shows a relatively greater association with the trapezoid than with the adjacent metacarpal. The facet for the magnum in both Ch lamy ther ium and Kr ag 1 i e v i ch ,1 a is convex dorsal].y and concave ventrally. In Kraglievichia the greater portion of the facet is convex, while the opposite is true in Chi amy ther ium. The ventral portion of this facet, when viewed from the ventral aspect, is parallel in a line at right angles to the long axis of tT:o bone in Chi amy ther ium v^'hilo in I\raglievichia it is oblique. HJnd lililbs: The femur (Figure 7a, b) and the tibia and Figure 7 Kraqliovichia paranensis Hind Linil:) Elements A. Right femur, anterior view B. Riyht femur, lateral view C. Right tibia and fibula, anterior view D. Right tibia and fibuJa, lateral viev/ X O.GO (linear) 48 49 fibula (Figure Ic, d) show no morphological differences from tliose of CI 1 1 amyther ium except for size. The calcanenm of Kraglievichia is relatively less ex- panded at the distal end, and is a relatively less robust element than that of C h 1 a myth e r i u m (Figure 8) . The facets for articulation with the astragalus are quite different. In Kraglievichia the facets are connected, forming a double facet, while in Ch lamy tlierium they are separated by a central valley. The relative proportions of the two facets also differ. In Chlamyther ium, the lateral facet is much larger than the medial facet, while in Kraglievj.cliia they are more nearly the same size. The navicular- of Kraglievichia shows no appreciable differences from that of Chlamytherium> A comparison of tiic three right naviculars from Haile XVA with one another s]:ows only a sliglit variation in the relative shapes and sizes of the facets for articulation with the cuneiforms. Regajiding the metatarsal IX, th^e facet for articulation v/ith the mesocuneiform is relatively narrower ventrally in Kraglievichia than in Chi amyther ium. In Chlamythcrium the medial border of metatarsal II forms an unbroken line, v/hile in K r a g 1 i e V i ch i. a thor-e is an indentation at the rsroximal end to accept the p:.oximal end of metatarsal I. There must have leer, a corresponding bulge in the proximal end oi' Figure 8 Kr a q 1 i e V i ch i £. and Ch 1 amy i :h e r i lam Calcanea ?: . Ki' £1 q 1 j. e V i cli i a pa ran en sis right calcaneum, Haile XV ^'- ?Kr ag 1 i evi ch i a sp. ?right calcaneum, Inglls I A C. Cli 1 amy tlT. e r ium septen-trionalis right calcaneum. Bran ford I A All approx. natural sir^e 51 52 metatarsal I of Kraglievichia; no sucli featiure occurs in notatarsal I of Ch lamyther iuia . Wlien viewed laterally, the proximal articular surface of metatarsal III appears rounded dn Kr ag 1 i e v i ch i a v;hi3e in Chlamytherium it forms a straight line, perpendicular to the long axj.s of the element. Due to the rounding of the proxi.m.al end of this element in ICr sig 1 i e v i ch i a , the articiilar surface can be seen Vv'hen the element is viewed from above; this is not the case for Ch lamyther ium. . This would appear to permit more dorsoventral movement of the toes of Krag- lievichi a . Regarding metatarsal IV, the facets for articulation with raetatarsal III and the cuboid are a single struct/are in Kracflievichia, while they are two separate structures in Ch lamy ther ium . Two ungual phalanges of UF 10902, both from digit V, liave been recovered fi-om Haile XVA. One is rather iiarrow aiid pointed, and the other is blunt and broad. Apx^arent.ly the ungual phalanges of the front feet were pointed, while these of the! hind feet wei-e b]unt. The same is true in Chi amy t h e r i um . Measarements of limb elements of other- examples of Kr a g 1 i e v i ch i a from various Florida localities are presented in Table 8. 53 Teible 8 Mearjurements (in nur.) of Jjimb Elements of Kr ag 3. i e v i c"h i a paranensis From Various Florida Localities UF 104 3 2 (righ^ UF 93 54 (right) Santa Fo I Santa Fe I Humerus Greatest anteroposterior diameter of shaft Lateral width of shaft at same point Latex-al width, distal end Greatest width of distal articular surface 28.8 19.5 60.2 37.8 54 . 2 36. Ulna Total length Lateral v;idth, proximal end Lateral v/ldth, distal end Lateral width at semilunar notch Anteroposterior width at semi].unar notch UF 10449 (left) UF 17474 (right) Haile XVA Sante Fe 116, .5 17, .6 10, .4 20.5 23.4 25.1 31.8 Radius Total length Transverse width, proximrul end Anteroposterior widtl-i, proximal end Transverse widtli, distal end Anteroposterior v^ddth , distal end Femur Lateral v/idth, proximal end Anteroposterior width, proximal end Lateral width of shaft at third trochanter Anteroposterior widtli of shaft at same point Calcaneum Total length VJidth of articular facets for astragalus Depth of facet for cuboid Width cf facet for cuboid UF 10830 (rivht) Sante Fe I 105.2 24.5 13.3 23.1 20.6 UF 17476 (left) Haile XII B 55.3 34.2 38.8 16, UF 17473 (left) Santa Fe II 64.] 24 . 5 14.9 10.6 QMblc 8 (continued) 54 Navicular Greatest lateral width Greatest anteroposterio]: distance Greatest depth UF 17568 (right) UF 17 569 (right) Haile XVA Haile XV A 37.3 33.6 22.6 30.6 Metatarsal II Total length Width, proximal end Depth, proximal end Width, distal end (articular surface) Depth, distal end (articular surface) 18.3 26.6 UF Haile XV 30.8 11.3 13.G 12.3 (left) Metatarsal III Total length Width, proximal end Depth, proximal end Width, distal end (articular surface) Depth, distal end (articular surface) UF 17472 Santa Fe 34.2 16.5 15.2 13.5 11,4 (right) II Metatarsal IV Total length Width, proximal end Depth, proximal end Width, dista]. end (articular surface) Depth, distal end (articular surface) UF 17475 Santa Fe 29.5 11 12 10 2 2 1 5 (righL) II 55 Review of tlic clilamyt?ieriinae Since the genus Kraglievichia is reported here for the first time in North America, this seems to be an appropriate plcico for a review of this group on both continents. The noincnclatura.l liistory of the various chl amy there genera v/il] first ])e discussed in the order of their introduction into the literature. This will be followed by a brief review of each genus, in geological order, beginning with the oldest form. The first remains of a chlamythere were discovered in a Brazilian cave deposit in 1836 by Peter Wilhelm Lund and described by him as Chlamytherium humbo.ldtii (Lund, 1838) . l"n his early works Lund referred to this genus as Chlamy-- theriuin, but later (beginning in about 1840) lie began to c£ill the genus Chlamydotherium v/ithout giving any reasoj; (Castellanos, 1927). Later authors vised the name Chlamy- dotherium, apparently not realizing that this name, in the meantime, had been given to a ganus of glyptodonts by 3rcnn (1838) . In 1875 Ameghino proposed tlie name Pampatherium for this genus, recognizing that the name Ch 1 amy d o th c r i um v/as occupied by the glyptodont genus. lie later abandoned this name v.hen he realized that Lund had previously named the genus Clilamy Lherium . Paula Couto (1956) has argued t liat Pa^Tipatheriu)!! is a valid name, as Lund meant to call it Chlamvdotheri um . 56 The first North American record of Ch laniy th er i urn was reported by Leidy (1839a) when he desigiiatcd the species Glyptodon septentrionale . In the same year Leidy, in another work (1889b) , referred these specimens to the Sonth American species C_^ hvimboldtii . Sellards (1915) believed that the North and South American forms represented differ- ent species and revived Leidy 's original trivial name sep- tentrionale. In 1902 Ameghino described Machlydotherium from the Eocene of Patagonia. Castellanos (1927) named two new genera of chlamytheres : Vassallia, based on c'.n edentulous mandible and several de]:inal plates; and Kraglievichia, based upon two skulls, a mandible, and a small amount of postcranial material. The genotypic species of Vassallia was originally Ch lamy th er ium mi nut urn. (Moreno and Mercerat, 1891). Kraglievichia was erected to include C_^ paranensis , C . intej-media (Ameghino, 1887), and .Q.-- pubini'-er'medius (Rovereto, 1914). The trivial name paranensis was retained. Simpson (1930) established the genus Holmes ina while sLudviiig excellent material of septentrional is from the Seminole Field in v;estern pen;i nr.u] ar Florida. Svibsequent authxrs have tended to ref'.^r to tVte North Jiraerican forms as Holmes:! na ana t.o the South Z^ir.erican forms as Chlamvtherium. 57 Castellanos later (1937) named P] aina, based upon the typo of C_^ in t ox-med i VI G wh. i ch he had eac-lier p].acod in Krag- lievichia. The reason for the estahllsluTient of this new genvis was his interpretation of the lineage of the chlamy- theres. He believed that there should be a form i}Ttermediate in size betv;een KraglievicTiia and Ch lamy ther ium . Since C . intormcdius is larger than the other material referred to Kragl i evi chia , he saw it as representing this intermediate form. The genus Ilof f stetteria of Castellanos (1957) was based upon a skull collected in Ecuador. This skull had previously been described as C_^ occidentalis by Hoffstetter (1952). Each of the chlamythere genera v.'ill nov/ he briefly re- viev;ed in geologic order beginning with the oldest, Machly- dotherium. This form is known from the Eocene of Patagonia and its relationship to the ]ater chlamytheres is not knov.n (Simpson, 1945) . Chlamytljerec are not knov.'n from the Oligo- cene. Vassc.3 lia precedes Kraglievichia teiaporally as it is known from tho ba Vcnta f.iuna of Miocene age in Colombia. l/he author has examined a sku] 1 from this locality (U.C.M.P. 40401) which definitely represents Vassallia. This identi- fication is ba:-:od upon its small 3;I?.e, and the nature of the anterior der; !..jticji, in wlTJ.ch the first five teeth are peg- 58 like and rotated lingually. Vassillia has also been reported from Auracanean dei:>osits (Castellanos , 1946) so it appears that this genus did range into the Pliocene. Porta (1902) has reported Kraglievichia from the La Venta fauna in Coloribia, but this record was based only upon dermal plates which could just as well represent Vassallia. Other Miocene records of Kraglievichia have been reported as some prevJ.ous works considered this to be the age of the Aaracanean deposits; however, more recent in- terpretations of South American stratigraphy (Patterson and Pasqual, 1968) demonstrate the Auracanean to be Plio- cene rather vlian Miocene. Therefore, it appears that Kr£g-- lievichia was restricted to the Pliocene in South Aiv.arica.. Castellanos (1927) has repor"ted morphological difft.\r-- ences between Vassallia and Kraglievichia . In Vassallia the first five teeth tend to be peg- like and rotated lin- qually while on.ly the first four teeth of Kraglievichia show this condition. Just how variable this charactex~istic is cannot be ascertained presently because of the lack of comparative material knovm for these genera. Vassallia j.s also significantly smaller than Kraglievichia . It is sus- pected here (as Castellanos has suggested) thi?t Vassallia is the ancestor of K)- a g ;i j. e v i c't i a . Regarding Plfiir^a, vliich is based on three isolated 59 dermal plates, Castellcinoy (1927) v/cis probably correct in }iis first judgment when he placed this material in the genus Kraqlievichia. lie considor.-:d these plates to bo inter- mediate in size and sculpturing between Kraglievichia and Chlamytherium. They seem, however, to fall within the size range expected for the older armadillo gen\-'S. Furthermore, it seem.s quite difficult to interpret slight variation in plate rugosity as being generically significant. It is therefore proposed here that Plaina is a synonym of JCrag- lievichia . T]ie genus Ilof f stettcria (Castellanos , 1957) is based upon supposed differences in the shape and m.easurements of the teeth. The deviations, however, are clearly attributable to individual variation. Hoffst otter ia is here considered to be a synonym, of Ch lamy ther ium . In Simpson's (1930) description of HoLnesina, he listed a numljer of characters in which this new genus differed from .ChljLni^'tj2.erium. However, James (1957) has shown that the characters pointed out by Simpson are not sufficient to separate the North and South American forms generically, and that Holmes ina is a synonym of Chlamytheriiim. This report strengthens Jcimes ' views on this issue. Simpson (1930) has also argued thav if Holniesina is not valid, then all the South American forms should be p.laced in the genus Chlamy- 60 theriuiii as they are no more different frojn each other than Cn larny th or iuin is from Holmesina. There may be some justifi- cation for doing just thi.o. In this report, however, new characters of Kraglievichia, different from Chlamyth ez'ium, have been pointed out v/hicli, together with what previous workers have considered generic characters, justify retain- ing Kraglievichia as a valid genus. The genera of chlamytheres, then, recognized in this report are Machlydotherium (Eocene) , Vassal lia_ (Miocene and Pliocene, South America) , Kraglievichia (Pliocene, South America and early Pleistocene, North AiTierica), and Chlamy- theriuni (Pleistocene, North and South America) . Evolution of the chlamytheres The principle morphological trend in Chlamythere evolution since the Miocene was an increase in size. Vassallia was Hie smallest form in the known chlamythere lineage, follov/ed ■^Y Kraglievichia which was slightly larger. An examination of the tvv'o femora in Figure 9 shews the great size difference between Kraglievichia and Ch 1 am.y th er iu.m , The intermediate steps in this size increase are illustrated in Figure 10, \/hich shows dermal plates from deposits representing various stages of the Pleistocene. Tlie Blanca;i Haile XVA plates are the smallest, followed by those from the Irvingtonian Jngli.'j TA site. Slightly larger still are the plates from Figure 9 Kraqlievichia and Chlamytliorium Femora A. C5i]^mi'_yieiiium septcnLricnal i^i right femur Bran ford I A B. Kraqllovichia paranensj.s right femur, Haile XVA X 0.60 (linear) 62 Figure 10 Kraglievichia and Ch lamy th er ium Dermal Plates A. Kraglievichia paranesis, Haile XVA, Blancan B. PRtraqlievichia sp., Inglis ?.A, Early Irving- tonian ^' Ch lamy th er ium septentrionalis , Coleman IIA, late Irvingtonian D, Cli 1 ?tmy ther ium soptentrrj cnalis, Bran ford lA, Ranchol abrean X 0.40 (linear) 64 65 i.liG later Irvingtonian Coloni?n IIA site, nnd the largest plates of all are those from the Kancholabrean ]iranford lA locality. A trend from peg-like to bilobatc teeth, and from >:)blique to jparalle] tooth rov/ axes was previously noted. Both of these trends seem to have extended from tlie rear of the dentition toward the front, as the number of bilobate and parallel teeth increases from Vassallia through Chlamy- therium. Concerning the geography of chlamYthere evolution, it is apparent thivt these forms origincited in South Tuner ica as all pre-Pleistocene records of this group are restricted to that continent. The earliest known Chlamythere is the Eocene Machlydo- therium, and no clilaraytheres are known from tlie Oligoccne. Vassallia is present in Miocene deposits of South Americ-a, and both Vassallia and Kr ag 1 j. e v i ch i a v/ere present in the South American Pliocene. Kraglicvicriia is knov/n from North American early rieistocene deposits, but its presence in South America during the Pleistocer.e is questionable. Chla- :tiytherium is known froiri the Pleistocene of North and South America. It has generally been believed tl;at the cvoJiitioi) of the chlamytherc'j hys hc-cn j;trict]y a Scutli AmericPn phenom- G6 cnon, v.'ith only the end product, Ch 1 a my t ]i e r i um , inigrating to North TVmerica during Rancholabrean time. The evidence sup- porting this hypothesis is, briefly, that all pre-Rancho] a- brcan records of clilamytheres have been from South America, a}id all genera of chlamytheres except Chl_ajij;^rtherjjara seemed to have been restricted to South America. The study of the Haile XVA fauna., and other pre-Rancholabrean sites of Florida, has yielded conclusive evidence warranting modifi- cation of this theory. Apparently Krag lievichia migrated to North America shortly after the establishment of the j.ate Cenozoic ] and bridge between the two continents. This is siibstantiated by its presence in the Haile XVA :-i.nd other Elancan faunas of Florida. Three hypotheses can be offered regarding the zooge- ography of the evolutionary transition from Kraglievj c?iia to Ch lamyther i.uig . The first hypotliesis, a Ixeady discussed, is that the evolution of Ch 1 a my t li e r i xu a tool" place in South America and this genus then moved into Nortli America in Rancholabrean times. This intepretatn on lias already been shown to be in erroj. . The second hypothesis suggests that Kraglievichia moved into North America auring the latest Pliocene and became extinct in South xAmerica, This would make tlia evolution of 07 Chlamytliorium strictly n North American event. According to this ]:ypothesis, Chlamytheriu:Ti, a North American form, then rf;invaded Soutli Amer.ica in tlie later PleistoconG, Tliis would account for the ap^oarent lack of chlainytheres in tlie early Pleistocene of South Anierica. The apparent lack of chlamy cheres from tlie Argentinian Chapadmalalan fa\}na should be briefly discussed here. A dermal plate has been questionably attributed to th.-i.s fauna (Kraglievich. , 1934) ; however, the specimen was from a local- ity near the major deposit and its horizon is actually un- known. Because of their unusua] .ly high number of bony jjlates available for preservation, if t'lio chlamytheres were present at all xn the fauna, tlicy should be v/ell rcjpreseni-ed. It is possible that their absence from the fau.ua js ecolog- ically determined. The early Pleistocene Chapadmalal fauna represents a temperate part of the continent, whereas the chlamytheres may have been restricted to tlie more tropical parts of tlie continent. Tlio lack of early Pleistocene de- posits in tropical South Anierica hampers a definite estab- lishment of the presence or absence of these forms c>n that continent at that time. The third hypothesis i:o be considered is that the dis- tribution of chlamytheres has been continuous on both con- tinents throughout the Pleistocene, and the^volution of 68 Ch 1 amyth e r ium from Kraglievichia occurred simultaneously in North and South America. That is the hypothesis favoi'ed here. Dasypus bellus Simpson 1929 Material : UF 16598, left nasal, right maxilla and man- dible, right astragalus, 51 isolated dermal plates. The preceding material pro?3ably represents one individu- al as the maxillary and mandible cirticulate nicely and a] 1 the material (with the exception of some of the plates) was found closely associated. Comparisons are made here v/ith other specimens of D. bellus and v/ith its closest living relative, D^ novemcinctus . The only other known D_^ bellus material contair.ing teeth is from Crankshaft Pit, Missouri (UK 15544, Oesch, 1967). Postcranial comparisons are made with inaterial from a number of sites of various ages in Florida . The nasal bone is represented by only the anterior portion and is Darger than, but morphologically similar to, tlKit of D^ novemcinctus . In discussing the dentition, teeth wi'll be designated 1 2 as follows: T will refer to the first upper tooth, T tiie second upper tooth, and so forth. The same procedure will be used in referring to the lower teeth. The maxillary is fragmented and contains only T^ -• T^ . Except for size, the lateral portion of rhis specimen com- 69 pares favoraloly with the corresponding portion of the Cranlc- 55haft Pit specimen. The first three teeth of UF 16698 v/ere probably transversely flattened as this is the nature of T . Tlie remaining upper teeth were probably round and pcg-liTce 4 G as this IS the condition in T - T . T?ie Crankshaft Pit skul] is from a juvenile individual, and T' he^s not completely erupted. In this specimen the first tooth is transversely flattened, but all the rest are round and peg-lilce. In D. novemcinctus the first, three teeth tend to be somewhat ] aterally flcittened, but not as much as in the Haile XVA specimen of D_^ be 3. 3 us . That tooth number and structure is variable in Dasypus has been jjointed out by Talmage and Buch.anan (1954), as well as by others. In four IX_ novem- cinctus skulls exeimined, the total number of teeth varied from seven to nine. The mandible of UF 16698 (Figvire 31) is lacking tlic proxi;nal end. Tlie number, shape, and location of the laental foramina are high3y variable in tliis genus. In both kncvm D. bellus mandibles there are two foramina, one large and one sma3.1. In UF 16698, the larger of the two foramina occurs between T^ and T^. The smaller of the two foramina is located 38 mm anterior to T-, . In UK 15544 , the larger foramen is anterior to the f^maller and is located directly beneat}) T2 • The smaller is 3ocated beneath T-^ . In D^^ Figuie 11 Dasypus bell us Mandible UF 1669S, Ha.iJc XVA X 2.26 (linear) 71 n CQ 72 novemcinctus , the number of mentnl foramina varies from one to four. The symjphysis of the Haile XVA specimen is v/eak, which is characteristic of the genus. The mandible of UF 16698 lacks T/ - T,. . T., - T-. are laterally flattened but T, find T-^ are round. This differs from tlie lower dentition of UK 15544 in which all the teeth round. Except for smaller size and a tendency toward flattening of the anterior teeth, the mandibles and dentition of the Haile XVA specimen differ little from other Dasypus material. Dental measurements are provided in Table 9. Differences in the astragalus of the Haile XVA D_j_ bel-\u_s and other specimens of Dj_ bel.lus from various FJ.orida Pleistocene sites and Recent specimens of D^^^ novemc inc tus are insignificant except for size. A comparison of the astragaJ.i of Dasypus from various stages of the Pleistocene reveals a size trend similar to that of the chlamytheres , with one exception (Figure 12). The size of Dasypus in- creased from Blancan through Rancholabrean ti.me, then de- creased in Recent times when it js represented by D_^ novem- cinctus . The dermal plates of DF 16698 do not differ from any other specimens of Dasypus ej:capt for size. Martin (in press) has me-r.sured samples of plates from several Florida 73 Tabic 9 Dental Mearuremonts (in inm) of Dasypus bellus UF 16698 UK 15544 Haile XVA Crankshaft Pit Florida Missouri left right Maxillary- Total distance, teeth 3-6 19.9 23.8-'^ 23.7 Anteroposterior distance, tootli 3 3.5 4.6 4.5 Transverse distance, tooth 3 2.1 4.6 4.5 Anteroposterior distance, tooth 4 4.4 Transverse distance, tootli 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 5.2 4.5 Transverse distance, tooth 6 4.1 4.8 Mandible Depth of ramus at tooth 1 6.8 12.4 VJidth of ramus at tooth 1 2.8 3.7 Deptli 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 Anteropostei-ior distance, tooth 1 3.0 2.8:; Transverse distance, tooth 1 1.3 1.8 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, tootli 4 4.8 4.7 4.6 Transverse distance, tooth 4 2.8 4.5 4.2 Anteroposterior distance, tooth 5 4.0l 4.1 4.4 Transverse distance, tooth 5 3.3 4.5 4.5 Anteroposterior distance, tooth 6 4.5 4.3 4.5 Transverse distance, tootli 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 TT alveolar .1 voolar ot completely erupted '•alveolar, estimated j> Figure 12 pasypus beJ-lus Astragali A. UF 16698, H?iile XVA, Blancan B. UF 13187, Coleman IIA, Irvingtoi^.iar C. UF 24 78, Mefforcl Cave, Rancholabrei X 2.2 3 (linear) 75 76 sites O-iid has shown little overlap in plate size between U^. fesll^us^ and D_^ novemcinctiis . His measurements indicate that the size of the plates do not fo^lov/ the trends through- out the Pleistocene which are seen for other parts of the skeleton. The Co-leman JIA pJ.ates are f-mall, and overlap slightly with those of D_._ noventcinctus . However, this was probably due to the small simple of plates used. There are approximately 2,500 plates in a single armadillo carapace and little or no sa^tisfactory method for determining the area of the shell from which most of them come. Therefore, size trends based on plates should involve large samples to insure accuracy. Inasmuch as the morpliological differences between the Haile XVA material and other D^_ bellus material are so slight, and those which were observed probably represent individual variation, this mciterial is referred to Dasypus bellus . Dasypus bellus has typically been considered a southern North American species of Rancholabrean age. iMartin ' s ( ir\ press) descrijjtion of the Coleman IIA fauna, however, ex- tended the temiDoral range of the r,pecies back to Irvingtcnian time. This report extends this species still further back, into the Blancan, and is thus the earliest known record. There is a question as to the relationship between Dj^ 77 hollu?^ and D_.^ novemcinctus . These animals are clearly more related to eacli otlicr tlian to any of the other armadillos (Auffenberg, 2957). The question arises as to whether D^, novemcinctus was derived from D,^ bellus or existed allo- patrically with IX_ bellus during the Pleistocene, replacing it during the ]ast few thousand years. Considering the former possibility first, it v.'ould have tal;en quite rapid selection for smaller size to develop D^ novemcinctus from p_-^ bellus. In Miller's Cave (Patten, 1963), a date of ca 8,000 BP is given for the Travertine stratum bearing D^ bellus remains, and a date of ca 3,000 BP for the Brown Clay deposits containing D^ novemcinctus . According to Patton (personal comjnunication) , even if the D. novemcinctus is a Recent intrusive (which he doesn't believe to be the case), the difference between 8,000 EP and 3,000 BP, or 8,000 BP and 1,000 EP is not that great — it is still rapid evolution, but possib.le, particularly if the only change is size. Concerning Um second hypothesis, if D^ novemcinctus existed somev/herc else during the Pleistocene and replaced IL^ ^G.-l-l'^^s very late, it should be known from some Central or South American Pleistocene deposit. To date, however, P.- novemcinctus has no fossil record except for Miller's Cave, hv.'c I ho fo.T.'.-il record in tropical America is extremely poor . 78 That p_^ fcellus io not known f.:om pre-Rancholabrean de- posit?; ovitside Florida is puzzJ.ing. Slaughter (3.961) lists two conditions which he feels were necessary for the occur- rence of this species: winters could have been Jio more severe than those of North Central Texas today; and rainfall probably had to be more than 20 inches per year. That temperature and rainfall could have restricted D_._ bellus to Florida?, during the entire Pleistocene is possible, but it seems more likely that the known distribution of this species is a reflection of a lack, outside of Florida, of Gulf Coastal Plains sites of the proper age. Order Lagoraorpha Family Leporidae Sylvilaguj sp. Gray 1967 3 Material: UF 17561, 2 M , UF .17562, M^ ; UF 17563, lumJ^ar vertebra; UF 17564, 3 innominates; UF 17565, 2 femora; UF 17566, 1 tibia; UF 17567, 2 metatarsals. jHypolagus and Sylvilagus are tv7o smaller lagomorplis common in North American Blancan deposits. Unfor tuneitely , the generic diagnosis provided for Hypolagus by D3v/son (1958) does not include discussion of either M"' of M„ , which are the only teeth preserved in the Hai'le XVA fauna. A compari- son of the innominates from Ilaile XvA with the descriptions and figures ot these elements of Hypolagus in Dawson (1958) 79 shov;s tlio I.iaile specimens possess a less elevated illial crest, a condition considered to be ipore advanced. A] 1 the rest of the Haile XVA material also agrees in size and morphology with Sylvilagas . Unfortunately, a lack of diagnostic elements prevents a species deterni.-i nation. Two living species are present throughout FJorida: S_^ palustris and S_^ f loridanus. The present range of the genus is throughout North Ainerica, with two species being present in South America: S_. floridanus and S_. brasielensis (Hall and Kelson, 1959). Previous Blancan records of Sylvilagus include the Curtis Ranch fauna (Gazin, 1942) and the Broadwater fauna (Barbour and Schultz, 3.93 7) . Numerous Irvingtonian and Rancholabrean occurrences of this gen vis are recorded through- out Nortli America. Order Rodentia Family Sciuridae Petauria sp. Dehm 1962 Material: UF 123 53, partial right mandible wj th M . By Miocene time flying squirrels had diversified into two groups (James, 1963), One of these groups ha3 compli- cated clieek tooth J.oph patterns. T'lis is due to the presence of "pc-otolopliules and metalophules, and csijccially meta- lophujcs that extend from ihe metaloph posterior.! y to the so posterior cingulum, dividing the posterior valley into two, and sometimes three or m.ore smaller valleys," (Oames, 1963) niis coin|Dlicated toothed group contiiins several living genera including Petaurista, Belomys, Trogopterus, and Sciuropterus, as well as the fossil genera Pliopetaurista (Kretzoi, 1959), Petauria (Dehra, 1962), and Pliosciuropterus (Suliir.ski, 1964), The second group has simxoler loph patterns in its cheek teeth. This group is represented by the living genera Hylopetes , Aeromys , Eoglaucomys , Glaucomys , and others, and by the extinct genus Pliopetes (Sulimslci, 1964) and several extinct species referred to Sciuropterus . Formerly, fossil species referred to Sciiiropterus in- cluded both simple and complex toothed types. Hugueney and Mein (1965), however, have placed the complex toothed fossil species of Sciuropterus in the genus Pliopetaurista based on DexDcret's (1897) S. pliocenica. The result is that all fossil species remaining in the genres Sciuropterus in both the New and the Old World have simx:)le teeth, while the living species which are restricted to the Old V/orld have com.pli- cated teeth. This curious arrangement leaves some doubt as to the evolvitionary unity of the fossil and recent species referred to Sciuropterus . The Haile XVA specimen (Figures 13 and 14) marks the first occurrence in the New World of a inember of che group Figure 13 Petauria s^^. RiyhL Mandible UP 123 53, Haile XV/x X 7 (linear) 82 Figure 14 Petaur.-ja sp. Right M^ UF 12353, Ilaile XVA 84 4 mm 85 witli complicated teeth. It is clearly more closely related to Petaurista than any of the other Jiving genera. The Haile XVA specimen (UF 12353) differs from Plio- petaurista in tliat its lA is oval in shape as opposed to being pointed at the posterior end. The metastylid in UF 12353 is only moderately developed as opposed to being well developed in PI j.opetaurista, and the postflexid ir. not as well developed in the Haile XVA speciinen. UF J2353 differs from the Mo of PliosciuropLerus in its shape, whicli is narrow and tapers toward the posterior end, as opposed to bejng oval in the Hai.le XVA specimen. "ITie mesoconid iti only moderately developed inthe Haile XVA specimen v;hile in Pliosciruopterus it is so v/ell dove]cn-cd that it forms a labial spur. The postflexid is also well developed iii Pliosciuropterus . UF 123 53 is clearly more allied with Petauria tlian any of tjic fossil genera. On the other hand, one striking dif- ference separates these forms: the mesoconid is completely absent in Petauria, but it is moderately well developed in the Haile XVA form (Figure 13 and lA) . The Haile XVh specimen sliaros one i.mporLant characteris- tic Vv'ilh Petaurista that does not occur in Petauria, nam.ely the piesenco of a mescjconi c! , On thC' otl'Cj- hand it shares the fol lowJ.nf.; charactorist Ics with Putarria vliich do iiot 86 occAu- .in Petaurista: the grinding surface of the talon.id is less complex; and tlie metastylid is moderately developed as opposed to the nietaconid-metastylid chaos (McKenna, 1962), Finally, it shares the following characteristics with both Petaurista and Petauria: the grinding surface of tlie trigo- ne d is complex, with extra lophids and fossettids; the post- flexid is poorly developed; the M-, is oval in shape; and the entoconid is connected to the hypoconid through the hypolophid. Since the Haile XVA specimen shares charact.e3"s with both Petaurista and Petauria, assignment to a genus is dif- ficult. Wl'iether the lack of a mesoconid is more important than the presence of extra fossettids and lophids on the talonid would probably be a disputed point among students of rodent evolution. However, Petauria is known from only one dentition, and tl:e variations which may have existed are not known. Petaurista, being extant, has been studied in greater detail and its variations have been taken into accouiit. Thus characters known to be peculiar to Petau- rists are more reliable than those in the diagnosis of Petauria. The simple grinding surface of the talonid (not observed in a very large sample of Petaurista) becomes a more imiiortant character with regard to the Haile XVA specim^.tn viian th.e presence of tlie mesoconid, \vhich may 8 7 have occurred in Pctauria. Following tliis line of reasoning, it seems itiore appropriate to refer tlie llailc XVA specimen to F-ctauria. The onJy specimen of Petauria previously known is from Bavaria, in the Region of Eichstatt. The deposit was a fissure filling of red loamy sand in the SoJenhofn Lime- stone. Tlie filling has been ascribed to the "older Pleis- tocene" by Dehm (1962). It is almost certain that the populations of flying squirrels of Florida and Bavaria during the early Pleisto- cene were reproductively isolated; thus the Haile XVA specimen is not assigned to the type species, _P_^ helleri . It also seems pointless at this time to describe a new species based upon tliis single specimen, as even the generic characters are difficult to ascertain with this ] imited amount of material. Until more material is available, it see;m.s more prudent to refer to the Haile XVA specimen as Petauria sp. Family Castoridae Castor canadensis Linnaeus 1758 Material: UF 17489, left femur. This element, lackijrj the proximal end, represents a young incli.vidua.l . It shoves no morphological differences from tlie .1 i\'Ing species. A comparison of measurements 88 (TablG 10) with examples of young. Recent C^ canadensis shows no significant size differences. Other Blancan re- cords of Castor are the San Joaquin locality (Kellogg, 1911; Stirton, 193 5) and the Hagerman fauna. Family Cricetidae Sigmodon medius Gidley 1922 Material: UF 12335, left M-*- ; UF 12341, right M"'-; 2 3 UF 12337, left M ; UF 12339, UF 12340, right M" , UF 12334, UF 12338, left Mjl ; UF 12336, left M^l UF 12342, right M" and M-5 (both unworn) . This species characteristically possesses only two or three roots on the M, , If accessory roots are present, they are centrally located and are very small, peg-li!ke struc- tures. The labial root is always better developed than the lingual. The two specimens from Haile XVA, UF 12338, agree with the above features. Sigmodon medius is nearly identical morphologically with a closely allied species, Sigmodon minor. According to Martin (in press), the only difference betv/een the t\\/o species, other than size, is that the re-entrant folds are deeper and narrower in S_^ minor than in S_^ medius. This cliaracter is more obvious in teeth with little wear and since the essential Haile XVA specimens shov/ significant v/ear, sise alone must be relied upon for the separation of 89 Table 10 Measurements (in mm) of the Femora of Fossil and Recent Castor canadensis C. canadensis C. canadensis Width, distal end Anteiopostcrior depth, distal end Width, external condyle Width, internal condyle V^Jidth of shaft at third trochanter 23.2 6 29.1 2 7.0--30.5 Anteroposterior diameter of shaft at third tro- chanter 13.9 6 11.3 lC.2-13.5 UF 17489 Re cent Haile XVA N X O.R. 38.7 6 36.6 34.1-38.4 29.2 6 29. 7 28.5-32.5 13.5 5 14 . 0 12.2-15.5 12.5 5 12.1 11.6-12.5 90 these hv/o specieG . Sigmodon medius is significantly larger th^^n S_, nii.nor . The larger measurements ot the llaile XVA material (Table 11) indicate that it belongs to S_^ medius . IVo subspecies of S_^ medius (_S_j_ nu_ m.edius and S. m. hibbardi) have been described,, baaed entirely upon size difference. The sraal]^ seimple of material from Hciile XVA, however, prevents the statistical treatmient necessary to determine the subspecies designation. Sigmodon medius is known only from Blancan deposits of North .?\merica. These include the Benson fauna, the Valle- cito-Fish Creek sequence (Downs and Wliite, 1968) , the Red Light Local Fauna (Akerston, 1970) , the Iludspetli fauna (Strain, 3.966) and the Broadwater fauna, Oirder Carnivora Family Canidae Gray 1821 Material: UF 17492 and UF 17493, 2 tibiae. The presence in the fauna of tv/o poorly preserved tibiae probably represents the Family Canidae. They are about the size of a fox, but positive identification is not possible as only tlie badly v;orn proximal ends are pre- servea, Family Mustelidae of Pteronura Gray 183 7 il2Lt:erJjil : UP 17487, humerus; UF 17491, right meta- 91 i-H * u o c •H B C o Q ■H w 13 c ni p •H 0) e c o 'd o e tr -rl W o •P C e u to m ■P (U P CO r\i c:> H ■■J' vo in t • • • CM 1 1 1 tN 1 1 o r^ iH fxj (N CO IT) n CO o 0^ en (NI in 1 1 01 CM i 1 • ' 1 1 1 ■-I 1 1 1 1 1 CO 1 1 ! 01 1 1 r^ CN 1 1 t • 1 1 (N r^ 1 ! lO ^' C5^ CO CM CM ^ tn • • • • IN H r-1 nH o CM rg r J t-l rj CN F: ;>; .H s (M O SJ ^ M rC rC ■-W 4J rH JL) r-"; tjl 4.' D~ aj ■■d a '•d a 'd (L) 0) •'1 ^i* •H -)J 1^ t . t"- M ^ PJ o I (J I (11 92 tarsal II; UF 17490, right metatarsal III; UF 17494, n.edial phalanx. The humerus (Figure 15) is lacking the proximal end and the distal end has a pathologic abnormality. The e] cmcnt is very porous in the region of the olecranon fossa and on the Jateral and medial edges of the distal end. The ab- normal growth of bono on the medial side of the distal end has res\ilted in a downward extension of this portion of the. humerus to a greater degree than is usual. This grov^th (and possibly a similar abnormality in the ulna) has caused a shift in the articular contact between the humerus and the ulna. The result of tliis shift in articular bearing pres- sure was a "planing off" of a poj'tion of the distal ar- ticuJ.ar surface. Disregarding the abnormalities of the element, it cor- responds very clo.'^ely morphologically with both Lutra and Pteronura . A possible difference between these two genera, other tTjaa size, may exist in the shape of the entepicondy] ar foramen. In Pteronura the foramen is nearly round, v;hile in Lutra it is elongate. The Haile XVA specimen more closely agrees V7ith Pteronura. A ]ength/width ratio of the measure- ments of the foramen shows the following results: Haile XVA specim.eri 1.3; Pteronura 1.4; and Lutra 2.2. It should be pointed out, however, that this cliaracter is likelv to Figure 15 Ptcronura hunerus UF 17487, lUiile XVA X ] .85 (linear) 94 95 be hicjlj]y variable, as one of the specimens of Lutra studied completely lacked this foramen. In size tlie Flaile XVA specimen also agrees more closely with Pteronura. Tlie majo.vity of the measurements of this specimen agree with the measurements of three Pteronura specipiens and were exclusive of the remge of measurements of seven Lutra specimens (Table 12) . Nonetheless, the rela- tively small samples gives cause for some reservations, and thus the Haile XVA specimen is only tentatively referred to Pteronura. Family Felidae Smilodon gracilis Cope 1880 Material: UF 17496, right tibia (distal end) and right astragalus; UF 17498, right metacarpal IT. This materi£il represents a medium-sized cat and is probably from one individual since the tibia and astragalus articulate v/ell, all the material is similarly preserved, and all the elements were found together. The material is referred to the genus Smilodon on the basis of tv;o characters presented by Merriam and Stock (.1932). First is the presence of the astragalar foramen. Second, the medial facet fo3- the calcaneum and the facc:t for the navicular arc merged. The IJaile XVA sabercat was very small, being similar 96 .-I CD +> i o:) M ffl 'd 0 M 'd G 5h m O c 0 O UH (^' CJ TT u M fri U r-1 H CJ rH a fO fj 4-1 >1 r ) -p -P d ^C5 >1 CO CO OJ C ^ •rl •r< o f-; ^ '0 c; u o C •H u ^ o M o. -H rC rC IH CD P-: •)J JJ -(J 0 ^a 'Ci Ti t; JJ •rl •H Q) G! C ;? ^ •-1-: 0) •p -p CO o '[-! en CO a o Q) (D 0) rC"! +> •I-' r' -U rc; (0 fd • — - in +J c 0 C. T5 V; 5^1 Qj CJ) CJ n P CN « O <^> 1.0 '^^ CO rH ^ p:: M' CxJ r 1 rl rH to t 1 1 ! 1 1 1 1 •rl o CO LT) ^' o 0- L,-) r- cc < » t • • • • C m r-l iri ^^ ^ C5^ o (U CO OvJ rH rH -H r-l •H lO 00 C^ oo H CN O V) • « • • • 1 • Co 4J iXi 00 in Kf \D O '^ U n IX C^) CM r->. rH r^ A s ' c; • 0. P^ r:5 CO CO CO CO (^ 0") CO fO ^ CO a si' r; o o m rH "^ Osl oo f^ o Q) • • • « • > • rH Sh rH CO r-l in -^d- r- O 1 — i a) •H 01 CM H r-l H fe -p m P P: m C o 1 •H 1 :^ +J r^ u rj U •H u •rl r^ o .C •p rH -!-' (U u 0 U r-l o E O >i ■H rj -H "J M CO !h C Q) •P -!-! u CO fO CO o O H 0^ .tj c; Di m o '■in '-P 0 -IJ u rii Hi U 0) Q) rC •r^ Q) ■H -P CO > 4-> -d n C (0 4-1 0 111 IM o ','"> O -!-> u jJ 10 U) Qt crj CO d) CO > G) CO P C3 CO 4-> a al c C fij c (U m Q.I H ^ ! . O Lh 6 97 in size to Smilodon gracilis. A comparison of the measure- ments of the astragalus from Haile XVA with those of the astragalus of _S^ gracilis from Port Kennedy Bone Bed (Cope, 1899) shows a close correlation in size (Table 13). Inas- mvi.ch as S_^ gracilis is the only small species of Smj lodon in Noitli America, the Haile XVA specimen probably repre- sents this species. Two sxDCcies of small sabercats are known from South America. Smilodon cruciens is knowii only from, a single mandible (Ameghino, 1904). Smi lodon todon riggi (Kraglie- vich, 1948) is present in the Chapadmalal Fauna and is known only from postcranial material. Tliis form is slightly larger than the Haile XVA form. The astragaluo is preserved in this specimen but it is in poor condition, and the critical chairac-ters of the facets cannot be verified from the figures. VJliile other species are shared by the Chapadmalal fauna and the Haile XVA faunei, it does not follow that the two cats are necessarily allied. The Haile XVA astragalvis so closely resembles Smilodon that if it is the same es the Chapadmalalan form then the Chapad- malalan form must )je synonymized with Smilodon. Order Proboscides Family Gomphotheriidae Cabrera 1929 Material : UF 17464, a small i:)C)rtion of a molar and two smal]. pieces of ivi:>ry. 98 Table 13 Measurements (in mm) of the Astragali of Smilodoii gracilis From Haile XVA and Port Kennedy*' Hailc XVA Port Kennedy UF 17496 Length 41.6 48 Width 45.2 38 VJidth 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 *meeisuroments after Cope (1899) 99 The complicated nature of the tooth clear ]y indicates that this specimen belonged to a gomphotheriid. Further identification is not possibJe because of the sme^ll amount of material availcable. Order Perissodactyla Family Ecjru.dae Nannippus phlegon Cope 1892 Material: UF 17484, and UF 1748 5. both upper cheek teeth; UF 17547, metapodial; UF 17548, proximal phalanx; UF 17549, four ungual phalanges. The relatively unworn cheek tooth indicat-i^s this form was strongly hypsodont. The other tooth (Figure 16) and the po.stcranial material agree in all respects with the typical Blancan species, N. phlegon. Measurements are pro- vided .in Tiible 14. Plesippus simplicidens Material: UF 10909, partial sKu.U; UF 3 7556, partial maxilla v-ith deciduous molariforiii tooth; U]' 10894, 15 cervical vertebrae; UF 1087 7, 3 7 thor^^cic vertebrae; UF 10895, 13 lumbar vertebrae; UF 10896, 2 sacral vertebrae; UF 10898, 3 humeri (2 left, 1 right); UF 30919, 1 right radius; UF 10910, 2 left pelves: UF 10913, 4 femora (2 left, 2 right); UF 10915, ? right metatarsals; UF 10921, 9 splints; UF 30916, 4 cuboids (2 left, 2 right); UK 10917, 3 calcanea; Figure 16 Hannippus phlegon Upper Molar UF 17484 Haile XVA X 4,4 (linear) 101 102 Tcible 14 Measurements (in mm) of Nannippus phlegon Crand.al and Post- cranial Materia] From Haj.le XVA Upper cheek tooth Anteroposterior distance Transverse distai'ice Crown lieiglit UF 17484 17.5 20.4 UF 17485 19.6 51.0 Metapodial Transverse widtli, distal end 25.7 Anteroposterior width, distal end 24.0 Proximal P;ialanx TA/idth, proximal end Width, distal end 24.4 20.6 Ungual Phalanx Total length Width, distal articular surface UF 17549a UF 17549b UF 17 549c 43.4 43.6 43.3 22.7 23.9 22.8 103 UF 2 0920, 2 astragali; UF 10917, 4 proximal phalanges; ur 10913, 4 distal sesamoids; UF 10914, 7 ungual phalanges. The Haile XVA skull (UF 10909) is in poor condition, and not enough of tlie fragments a.re present to allow complete re-construction. Both left and right dentition, liov^'ever , are complete (Figure 17) and a portion of the premaxillary has been rebuilt. A sm.all portion of the occipital region is preserved but the rest of the skull consists of isolated fragments . The skull is referred to tiiC genus Flesippus on the basis of the v/ell develoised P , the poorly developed para- styles and metastyles, the di]^^tion of the fossettes, and the moie rounded nature of the protocones. In dental morphology, UF 10909 resembles both Plesj ppus shoshonensis and Plesippus simplicidens . Gazin (1936) states that the ranges of measurements of these two species overlap though P_,_ simplicidens tends to be smaller. The Haile XVA matoricij. (Table 15) is smaller than £iny of the individuals measured by Gazin (1936). It is, however, about the size of the specimen of P^._ simplicidens described by Hibba.rd (1941) froiit the Blancan Rexroad fauna of Kansas. The postcranial material is also small (Table 16), and falls within the range of nioasurements for P._ sj.mplicri dens provided by Gazin (1936) . Figure 17 Plesippus simplicidens Upper Dentition Ur 10909, Haile XVA (slightly less than natural size) 105 *^^i 106 Table 15 Measurements (in imii) of the Upper Dentition of Plesippus simplicide2"is From Haile XVA Length of tooth roW'^ Length, P-i Widtli, Pi LeiKjth, T?2 Width, P2 Length, P, Width, P3 Length .- P^ Width , P4 Length , M-, Width, Mj_ Length, M„ Width, M2 Length, M-^ Width, ^^3 Left Right 158 155 12.2 5.4 32.9 32.8 24.7 24.6 24.1 24.0 26.2 25,2 23.3 21 .9 24.8 24 . 3 22.3 23,8 23.8 25, 5 22.0 22.4 22.6 23.9 28.8 33.2 23.0 35.3 All measurements were taken at the grinding surface; width measureiaents do not include cement. *not including P-, 10' Tabic 16 Measursiaeiii-s (xn mm) of Flcsippus siniplicidcnr; Porstcrania] Material Frc>m Hailc XVA Humerus Transverse wic'th, distal condyle Anteroposterior width, distal end N Y. 3 70.1 3 80.7 O.R. 69.7-70.5 77.0-84.0 Radius Total length Anteroposterior width, proximal end Transverse width, proximal end Anteroposterioi' v.'idth, distal condyle Transverse width, distal condyle 316. 0 1 77.6 1 36.3 1 37.4 1 57.9 Fenura Total Length Transverse width, proximal end Anteroposterior v/idth, proximal end Transverse width, distal condyle Anteroposterior width, distal end 1 368.5 1 1J4.0 1 C8.5 4 87.2 4 116.8 84.6-89-7 114.0-118.0 Metatarsal Antex'Oposterior \\'idth , proximal end Transverse width, proximal end 2 41.2 39.6-42.8 2 48.5 48.2-4 8.8 Calcaneum Total ]ongth Lateral v.'idth 3 107.7 103.0-110.0 3 42.0 43.2-46.5 Astragalus Total length Laterial v;idtli Proxima]. Phalanx Total length Widtli, proximal articular surface ^\' id th , d •; .-. t a 1 end 2 58.0 57.0-58.9 2 58.8 57.7-59.8 3 73.5 76.5-81.2 2 43.7 42.8-44.5 4 36. 1 33.7-38.5 Table 16 (continued) ■Medii'.j Phalanx Total length Width, proximal articular surface Width distal end Ungual Phalanx Total length Width, proximal articular surface 108 N X 0 . R , 45.5 44.1-47.4 6 40.7 6 40.9 3 9.1-43.2 38.8-42.8 >4.4 50.6-58.8 38.5 34.1-43 109 Equus (Asinus) sp . Materla.1 . UF J.7750, upper cheek tooth; UF 17483, lower molar; UF 17482, M3 ; UF 17551, left humerus; UF 17552, astra- galus; UF 17553, proximal plialanx; UF 17554, medial plialanx; UF 17555, ungual phalanx. The three teeth listed above do not agree morphologically with those of P] esippus . Regarding the upper tooth, the fossettes arc mucli more complicated than in PI esippus . The lower teeth are also quite distinct from those of Pi esippus figured by Gazin (1936), Hibbard (1941), andMcGrew (1944). In all the lov/er teeth figu:i:ed by the above authors, the me- dian valley enters the commissure. This is not the case with the Haile XVA specimens. In these, the median valley approaches the commissure but does not enter it. In tlii s respect, the Haile XVA specimens resemble Equus fraternus, a Pleistocene ass. Several postcranial elements in the Haile XVA faunei also suggest the presence of an ass. These are much smaller than specimens attributed to £^ simplicidens . Hibbard (1956) has reported the presence of an ass in the Meade County fauna, in Meade Co\inty, Kansas. Tlie referred speci- men was a medial phalanx. The Haile XVA medial phalanx agrees very closely with regard to size (Table 17). 110 Table 17 MeeiGurements (in mm) of Jlc[uu3_ (Asinus ) sp. Cranial and Post- cranial Material From Haile XVA Upper Cheek Tooth Antei'02::>osterior distance Transverse distance 25,4 23.4 M3 Anteroposterior distance Transverse distance UF 17482a 11.2 UF 174C2h 28.7 10.7 Humerus Total length Transverse width, anterior end Lateral Vv-idth, distal condyles Anteroposterior width, distal condyles 255.0 88--'- 68.1 46.9 Astragalus Total length Greatest width 50.4 50.0 Proximal Phalanx Total length Width, proximal end Width, distal end 60.5 34.8 30.5 Medial Phalanx Total length Width, proximal articular surface Width, d.Tstal end 40.0 38.1 34 . 8 Ungual Phalanx Total length Width, proximal articular surface 45.7 33.1 ^estimated 11^ Family Tapiridoe .Tapjrus^ sp. Brisson 1762 M^-te^VXal: UP 17468, partial cxown of a cheek tooth. Since only one genus of Tapirs is known from post- Jlempliillian deposits of North America, it seems reasonable to ascribe this specimen to Tapirus. The material only serves to record the presence of this genus in the fauna. The only other 13lancan record of this genus is the Santa Fe I fauna (Webb, in press a) . Order Artiodactyla Family Tayassuidae Mylohyus floridanus Kinsey (in press) Material: UF 18002, left and riglit mandibular rami and upper left caniiie (holotype) . This species has recently been described by Kinsey (in press) based upon tlie material from Hailo XVA. It represents one of the few Blancan records for this genus. The material is figured (Flgur-s 18 and 19) and the diagnosis is included liore in order to complete the study of the fauna. Measure- ments of the holotype are presented in Table 18. Close to ^l^^■lohyus nasutus in size and elongation of tlie snout. Larger and more robust than Mylohyus fossilis and lacking extreme constriction of the symphysis diag- nostic of that species (Lundelius, i960). Pre- and Post-canine diastema longer than in other species of Mj^lohyus. Symphysis much more slender than in Pros- thcnops. Coronoid process triangular with straight, not conve::, anterior margin. Figure 18 Mylohyus f loridanus Mandibular Rainu; UF 18002, Haile XV'A X .89 (linear) 113 W Figure 19 Mylohyus f loridanus Mandibular Symphysis UF 18002, Haile XVA X .89 (linear) 115 116 Table 18 Mcasu3:ements (in mm) of the Dentition of Mylohyus f lorideinus- Length, upper canine Width, upper canine I^ength, lov/er canine VJidth, lower canine Length, P2 Anterior v/idth, P2 Posterior Width, P2 Length.. P3 Anterior v.'idth, P3 Posterior v/idth, P3 Length , P4 Anterior width, P^ Posterior width, P4 Length , Mj Anterior width, Mj^ Posterior width, M-. Length, M2 Anterior v/idth, M2 Posterior width, M2 Length, M3 Anterior widtlj, M-^ Posterior v/idth, M-, Length, premolars Length, molars Length, raolar-premolar series Post-canine diastema Pre-canino diastema Depth of jaw at Mj^ Thickness of jaw at M.i Width, jaw at canines Width betv/een canine alveoli Least width, symphysis Length, articular condyles to anterior end of symphysis Left 15.1 11.1 14. 10, 8. Right 4 5 12 9 9 13 10 12 13 12.6 12,8 15.2 14.0 13.9 21.1 13.1 11.8 33a 49.9 83.0 80 14 0 41, 21, 41, 17, 26, 2 2 2 6 4. 13, 12, 12, lb. 14, 14, 21, 13, 11.9 50, 13.0 41.0 21.1 2 73.0 *ineasureiuent3 aftej- Kinsey (in press) 117 Total length of tlio P2~''^'^3 scries loss tlian in other Mylohyus, both absolutely and relative to the overall length of the jaw. M^ and M2 relatively shorter tlian .in other Mylohyus except M_^ exoticus (USNM 8876) . Anteroposterior diameter of V greater than in al] other Mylohyus, both absolutely and relative to the cheek teeth series. Width of V greater relative to width of I>2 31'^cl M^ than in other Mylohyus. (Kinsey, in press) It is j.nteresting that Mylohyus does not occur in the other Blancan faunas of FJorida. Such a large fauna as Santa Fe I, for example, should contain some remains of this genus if it were present in the area in any great nuraJ^ers. That it was scarce in Florida in pre-Rancholabrean time is suggested by the Coleman IIA fauna in which Platygonus out- numbers Mylohyus eleven to one. Platygonus is piesont in the Santa Fe I fauna. The occurrence of different peccary genera j.n the Haile XVA and Santa Fe I faunas may suggest an ecological difference between the two. This possibijiLy v^/ill be discussed in the paDeoecology section. Family Camelidae Hemiauchenia cf macroccphala Cope 1893 Material: UF 10900, partial skull; UF 10894, 2 cervical vertebrae; UF 10899, phalanges (3 media], 1 ungua.1 ) . T]\e Ilailc XVA skull (UF 10900) consists of tlic pa.late with upper dentitjon (Figure 20) arid j:iarc of the cranium.. The cjrania]. fragment incTudos parts of the. parietal, left squam.osal, left exoccipital, and supraof:cipital. This por- Figure 20 Hemi au ch en i a cf macrocephala Upper Dentition UF 10900, Haile XVA X .93 (linear) 119 f' 120 tion of the pkull, particularly the external auditory meatus and the subsquamosal foramina ,ac;rees very closely v/ith the type material for Tanupolama figured by Stocl: (1928) , The Plaile XVA specimen represents a juvenile individual. The left DP' is present as are the left and right DP^' and DP . Left and right P and P- are present but unerujjted. The complete right m.olar series is present v^'hile the left is ■J represented by only M". Tlie Haile XVA specimen is a small llama referrable to what most authors have called Tanupolam.a which is now partly included in Hemi^iuchonia in a recent revision (Webb, in press b). In this revision (V.ebb, in press b) three North American species are recognized: H_. b] ancoensis, H. s.S''P'iilrlr en sis , and I;I_^ macrocephala . Flemi au ch en i a bJancoensis is large and has a relatively narrow P . Hemiauchenia se/^Triour-_ en s i s is about as large as H_^ blancoensis and differs fro:r; it in the nature of the P4 (Hibbard and Dalquest, 1962). Ilemi- auchenia macrocephala is the smallest of the North Am.erican species and has a relatively v/ider P^ . The Haile XVA speci- men lacks the P., but the specimen is quite small (Table 19) cind tentative species allocation is based on size. The occi^.rrence of H_._ macrocephala in the Haile XVA faunei marhs its earliest record anyv/liere. Thcit this species has a long continuous history in Florida is indicated by its 121 Tabic 19 Mcasurenientfi (in inm) of the Upper Dentition of Hemiauchenia cf macroccphala From Ilailc XVA Length , DP^ Width, DP^ Length, DP^ Anterior V'/idth, DP-^ Posterior width, DP-^ Length , DP^ Anterior width, DP^ Posterior width, DP'^ Length, P-^ Width, P^ Length, P'^ Width, P"^ Length , M Anterior width, M-'- Posterior v/idth, M Length, M Anterior v/idth, M^ Posterior width, Vr Length, M-^ Anterior width, M^ Posterior width, M-^ Length of deciduous premolar series 50.1 Length of molar series Ib.O Length of tooth row functional at time of demise 125 Length of permanent tooth row 110-'- Palatal width at DP-^ 34.5 Palatal width at DP'^ 37.5 Palatal v;idth at M^ 47.5 Left Right 10.8 5.8 19.7 19.2 11. s-" 12.4 17.3 19.6 18.8 17.7 17.8 19.2 19.1 16.5 9.9 18. 8^ 15.2 26.2 25.7 20.2 21.? 19.8 20.4 22.7 21 .2 19.7 26.6 14.2 19,1 1 estimated "^unerupted 122 presence in the Irvingtoni" an IngJis lA fauna (Webb, in press a) . Hem i au ch en i a blancoensis occurs in the Santa Fe I fauna (Webb, in press b) possibly indicating ecological differences. This genus was widespread in the Blancan, being present in raost North American faunas of this age. It also occurred in the Argentinian Chapadnialalan Fauna (Kraglievichia, 1946) . Family Cervidae Odo CO ileus virginianvis Zimmerman Material: UF 10885, 6 partial upper dentitions; UF 17481, right mandible with P^, DP4, M2 , M-^ ; UF 17478, left P^^ , UF 17477, right M3 ; UF 10883, 19 cervical vertebrae; UF 10889, 25 thoracic vertebrae; UF 10884, 3 5 lumbar vertebrae; UF 10891, 3 left humeri; UF 10893, right humerus; UF 17479, 3 metacarpals (2 left, 1 right); UP 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 bean possible. The definitions of closely related species (and geneia) of deer are very similar, and the taxonomic problem is cora- poundcd by the large amount of dndividual variation in these 3.23 forms. Tho nature of the P. is considered important- in deer taxonomy at the generic or subgenerj.c .level; unfortvinateJy, this tooth is also the most variable. In Simpson's (1928) discussion of Blastoccrus extraneous, he pointed out t?iat spccim.ens of Odocoileus studied approached Blastocerus in character, and some Recent Dlastocerus material diverged from his specimen towa.rd Odocoileus . The subgenus Procoileus (Frick, 193 7) is based primarily on the characters of the P4, An examination of a large sami^le of Odocoileus -l.eeth from the Santa Fe J. fauna showed individual variatioii ap- proaching the characters of Blastocerus, Procoileus, and eveii Craniocerus , A similar, ttiough not as drastic, example of individual variation observed in a sample of Odocoileus teeth from several later Pleistocene and Pvccent sites in Florida. It appears as though P, morphology is not reliable for taxonomic stv;dies. One P. is preserved in tlie Haj.le XVA material. Only the labiciJ. portion of the tooth is preserved, £ind it shows little wear. Th.e portion present, however, compares favor- ably v/ith most specimens of 0_^ virginianus observed. The shape and relative size of antlers are other characters used in deer taxonomy (Frick, 1937), Unfortun- ately, no ant.ler material is present in Ilail.c XVA. Size also has been a criterion for species determination 124 of fossil deer. The Haile XVA deer are relatively small. This is not considered taxonomically signif iceint , however, as a size comparison of certain postcranial elements from vari- ous stages of the Pleistocene (Table 20) shows random si'?;e fluctuations. Evidently, Odoco ileus exhibits no such size trend throughout the Pleistocene as was shov/n fo.r the edentates. The size variations seen are probably ei reflec- tion of ecological conditions. Harlow and Jones (1965) shov/ that the size of Recent deer fluctuectes significantly from liabitat to habitat within the state of Florida. Finally, an element by element comparison of the Haile XVA deer material with specimens of 0^^ virqinianus fro;r- other Florida^ Pleistocene sites, and with Recent specim.ens, shows no conspicuous differences. It is considered then, th^^t the Haile XVA cervid represents Odoco ileus virqinianus . 12 5 o CM 0) r i rj E-i •H D1 5h ■H > y 0) -.H o u o o (0 0) •H -!-' •H r-l a u o o c (D OT U 4J O o Q) w E4 O U C ^1 ■(J C 0) £:■ 0) 0) H P4 o ^. !;><; ^. O '^A CO CM VD o • o; cr; o • (NJ o • »i) (N -vf OJ "J- I O o fO m (N CN CO ro CNJ CC CM <-{ r-1 iH CM in • • CO o CM CN 1 t 1 < • kf) o (N CM ro CNJ 1 1 1 1 » * CM CN rH ID r^ • • rH fO 00 CM CM CM CM o cn r>j CM iH r-l C-) \£) • t C^l UD CM 1 Cs) 1 1 1 • • rH Cn] CM CJ CO CO r~ o CM ^' CM CM r^j (M rsi CM CO CM Ul Q) 1 1 iH •H •rH >i '0 X ^ X 11 (^ 0 i-l 0 t! Q) J-l 0) Sh 0 a D: U (T3 ^ H K ^ e n} •r^ -IJ •rH -P •P X 'O X '0 w nH 0 ■rH iH 0 •H •H m u ^ rO u :s 73 a, a w 0< W M u 5-1 M ?3 - rt> V 0 (U ■. 0 M ,.a u rC; •-■1 +J rc; •H a) -p (« +J u fd •p M e 'o ^j ■d Qi 'C! +> •a 0) ^ P -H Q) •H 4; C. : •j: 0 (U r-l r-l f/jrH r-< Pi r-l fO fO 0 m ,1 o \x < H HI E (U H O u i I I I I I I I I I I I I I I I I I I I I I ^-; 1 1 CTi .-M • « CD 10 • •^t H IX, 1 1 • ^J- «JD 0 • CTi .-1 r-) rH in ro • • ^ r-l o o CM ^ rH ro rH vi' r-l D^ U. • H ^ r- r- rj rH <£) 1 1 1 KD 1 • • CO r- ro CM VD 01 CM • CO CO I CM • o CO • "si' fO I cn • rH C-) 0"i CM CO o rsi CM 1.0 Oi CM 10 i 'O X •« c 0 c c 5-1 0 p. u rH (d - X T-i E rC c fd •H p fd ^CJ 4J X Ti :-! c W rH 0 •H fd dJ •H fd ^ 5 ^ -d 0^ 0, Pj rH to u 'H rj ri - fd - 0 P £■ !-l ^ u r, -H w •H (!) ■p ai -I-) M ^j X r; 'Ci -ij TJ CO V) •t-'i 0 ■J ■H C) •H p t (^ K r-l ^ rH D-, a 1- r^ % Id fd 0 fC jj ^' M M ^-! K t)-"- P 0) (U (U C ^C5 •vJ p -p X) 1 1 ] • • ! r o '^' II ^' H 126 0^ m r-- n r-\ CM s)< (N (N 111 CI (N CO lO OJ O! r~ r- LO '^ • • lO i-~ r-j 1 1 1 o • • '^' 1J3 CN CN lO ro • • ro ro r^ r~- • • <^ LO •H "Ci X C 0 (U M a< rH frt ■. e ^ X •H .P c X -^d (n Xi ■H O -H t^ c; fti M 15 (0 CD to Dh rC !-! W f^l rH nj V o m -:-i ,-d -.H -P J.; fU i) 5h V) -H -P Tj 0) -d u X (L) -H +J c* ■•-[ 0 ^ •; oT 'ji 1h u o a, <-i P-, rH fij 0 (J ,c •. M J-i g -p ^ GJ a) tJ" 4J +J -p t; 'Ci ro r; d; ■H t^ < h^. *^^ (D > •H (U 0) c ,5^ u ;3 -p ai U -P H n 0) Q w iH •.-I > Q O N IX o I CD rs H fO l.O • ro f^, 1 • LO o t OT ro (X> CM CM r-l LO (Ti rH CM I CNJ 1 ] 1 r— I • • ro r- CM <-i (X) rH i^! o H ro 1 CM t 1 O 1 in • e r- 0:1 rsi r-l ro 1^ 10 CM ^ C^ IT) V& kO cri CM 1 1 1 0 1 c 4 rH CM 00 CN IT) CM CO 0 r- CM 1 CM 1 1 ro 1 • • vt in CM CN 01 CO CO •■Ti ro 1 H 1 1 1 !0 > • U5 •H CO rH LO r^ • • r I CM S!' rH rH rH I I I I I ! I I I I I I I I I 1 I I I I LO CO 0 LO IX) 1 1 ro CN CM IN CM CM I-l 1 1 1 1 CO rH •-i rH •-{ I i CO 0) ! 1 r-l .,-• •H N ^ X ^ X ■0 c 0 a 0 C Q) u rti M J: ^ 0 '0 Ch a. to Dh fO 10 i: !! JH !h Clj rH ri » td *. 0 (0 V 0 fd u ^ u ri-H •H +> .c •H 4J C 0) +) a V M fd -P Sh CO ■H G ^ -p T.1 0 •v JJ ^ 0 ^ M X P -rl Si •H JJ a (U •H 4-1 c: •H 0 m '$ s^ /■. CO 0 Q) !-> w 0 0 rn J.i r-l r-l .0, i-l r-l 01 r 1 m rj 0 fTl fij 0 rj r*-i V J^l ^ U E M ^-1 l~; 4-' rCl G) 0 (L) O) CO tr^ .IJ Jj 4-' •!-• -P -P c; 'Ci fi< ai C (C C 0 •H ^^ lT <; 1:1 <: !-1 r^: AGE AND COHRi;j^\TIOi>J Workex's such an Schiill.z (193&), \\.i.l.so3.i (.193B) , Mc&iev; (1944), and others liax'-a long recognized tho. Ii the Blancv'.n faunas of North T^jiisrica Cc'in "ii^ idontjfied. by tlio prcse.nce and absence of certain marriraalj aji genera. McC4rev; (1944) suminarize;d these as folJo'.vS: (]. ) the absence of typically Fliocone genexa; (2) the p3:ese]:ice of characteristically middle and late Pleistocene genera; (3) the presence of certain genera that survived from the llemphi llicin but which did I'ot outlive tlie Blancan; (4) the presence of certain genera, tliat were limited to the Hlancan; and. (5) the absence of ce.■Lt^^in genera that did not irioke their appearance untiJ. after the close of the J3.1ancan, In the follov.'ing secLion, eacii of the above criter.ia will be examined with regard to the ilaj le >r\'A fauna, (.1) There are no typically (No^hli Pi;ri.>riccin) Pliocene genera rep- resented in Haile XVA. Kr ag 1:' o^/ i ch i a ds of Pliocene age in Soi.(th /jTiorica but is B] anf:an or later in North America. 31 should be po."inted out here, however, that negative evidence, while rarely ever good, is particularly dangerous to weigh lieavily in rolrition to t}io Hailo XVA fauna. The small number of taxa and iiidividuals probal^ly indicates that t})e sXjeciiviv'ns collected are not a tlioxough s-ample of the fauna ei)f: i.Tc^a at that time. 127 128 (2) There are Eorae characteristically middle and late Pleistocene genoira jpresent in the fauna; Odocoilevts , Mylohyn.s, Smilodon , and Sigmodon represent this element. (3) Of the genera which survived the Hemphillian but not the Blancan, only Nannippus is present foi: cert^iin. The presence in the ft^una of a goraphotheriid may fit tliis cate- gory, but tlie lack of a definite generic alloceition raakes its inclusion here tenuous. (4) There is only one rei^resentat ive of the genera which are totally restricted to the Blancan, the Plesippus . The occurrence of Sigmodon laedius and Nannippus ph logon is critical, however, as these species are restricted to the Blancan. (5) No genus appearing after the close of tlie Blancan is present in Maile XVA. This again is negative evidence, and the rem.arks previously made concerning this also apply here. Wien all the mammalian faunal evidence is considered (see Table 21), particularly the presence of Nannip]jus, Plesippus, and Sigmodon medius, a Blancan age seems con- clusive . More refined correlation of the Haile XVA fauna witii other North American Blemcan faunas (outside of Florida) leads to difficulties resulting from gecgi'aphic distaiicc and 129 Tabic 21 Faunal Comparisons of Hailc XVA and Othf:r Blancan Localities IlailG XVA Santa Fe IB Florida B_anco Texas Rsd Light Texas fludspsth Texas 0 'J o K Anita Arizona Benson Arizona Curtis Ranch Arizona Sand Draw Nebraska u ■X.- \i, 0 0) r-; o S: r-. (I) fL' ■-J H Overton Nevada Reno Nevada Crypt Otis X V Seal opus X X Glossotheriura X X X pasypus X Kraqlievichia X Sylvi] aqus •? X X 1 1 I 1 1 Petauria Castor X X X Sigmodon ^1. X X Smilodon ... . X X 1 Fteronura Tapirus X Nannippus V X X X X X X P.l esippus y,_ X X X x_ ',> >: ■p X X X X X fivloh^'us X X Odocoileuy x_ •^ X X Ilciniaut.'hf-nia X X X .■■.- 1 .< r X 1 1 1 1 1 Table 21 (continued) 130 Ilaile XVA Wichman Nevada San Joaquin So. Calif. Asphalto So. Calif, CO •H rJ . -p <-H a -H ;3 rH O fO s, u o • CO O o CO u San Timoteo So. Calif. Bautista Creek So. Calif. Vallicito-Fish Creek So. Calif. Tehama No. Calif. m tn -H u .--, ::i (Ci CO +) • ■P 0 ■H iL, Pa Ritchie Creek No. Calif. Lcv/er Lake No. Calif. Cabo Colnett 1 Mexico Chapadmalalan 1 Argentina ' Cryptotis Scalopus Glossothorium X Dasypus Kraqlievichia ■p Sylvilaqus Petauria Castor X Siqmodon X Smilodon 7 Pteronura Tapirus Nannippus Plesippus ^7- X X X X X X X X X ? ? Mylohyus Odoccileus X X X X H em i a v ch e n .i a X X X X 131 ecoloyical differpnces. Most Blancan faunas are located in westeirn North Arrierica, and no Blancan faunas outside of Florida are known from the Gulf Coastal Plain. Consequently, it is easy to see that the lack of a more detailed faunal comparison need not imply a great diffejrence in time. The absence of Kr a g ] J. e v i ch i a and Dasypus from Blancan sites elsewhere in. North America is probably due to the geo- graphic difference and not a temporal one. Conversely, a closer correlation with other Blancan faunas in Florida (for example, Santa Fe I) does not necessarily indicate precise equivalence in time. Although close similarities exist be- tween the Haile XVA and Santa Fe I faunas, enough differ- ences are present to indicate a difference in time. The Santa Fe I cam.el belongs to the typically Blancan species Hemiauchenia blancoensis while the Plaile >CVA form most closely resembles IK_ macrocephala . This possibly indicates a younger age for Haile XVA, as H_^ macrocephala is the typical Irvingtonian and Rnncholabrean form. There are certain forms found in the Santa Fe I fauna \vhich one vs^ould also expect to find at Haile XVA but does not. Notable among these is tlie typically Blancan dog BorophaguK . The absence of this form from the Haile XVA fauna probably mupt be attributed to sampling error. The absence c > C Ca.-jtoroid^-s and Jlydrochoerus from Haile XVA is 132 prohab]y clAie to tho same phenomenon. Other differences be- tween the faunas of Haile XVA and Santa Fe I are most likely ecologically induced. These are discussed in the next section. Araong the best represented animals in the Plaile XVA fauna are the edentates, a group of undoubted South 7\merican origin.. Of the three genera at Haile XVA, only Glossotherium has been previously recognized in Nortli 7\merican Blancan faunas, being present in the Blancan faunas of Texas and Nebraska (Table 21) . Dasypus has not been found so ea^rj.y elsewhere in North America. Kraglievichia has not been recognized at all in North Anierdca, although its probable descendant Ch lamy ther ium (=Kolmesin_a) lias been found in Irvingtonian and Ranc:holabrean deposits. This record of a major influx of edentates from South America is a further indication of the Plio-Plei;vtocone age of Haile XVA. The small chlamythere from Haile XVA shows a remarkable resemblance to the species K2: ag 1 i ev i cli i a p^j^l'Il ne s i s fi'om late Pliocene deposits of South America. Al- though no authentic Chapadmalalan (early Pleistocene) species is known, it probably lived in regions, more tropical than Argentina and probably also resembled the Florida mater-ial. Florida and Argentina ap^oarently jrepresent the northern and southei^vi liir.its of the range for this form, and it seems 133 safe to conclude that the momboi-s occupying the middle por- tion of the range were similar. The close rescirib] ance of the Ilailc X\^A Glosaotherium to that frora tic Chapadmalal also supports the correlation betv/ecn the Haile XVA deposit and the Argentine Chapadmalalan deposits. This Blancan influx of edentates into Florida under- liiies previous correlation betv/een the Blancan stage of North America and the Chapadmaltilan stage of South America. •That correlation previously lias been based upon the appear- ance of numerous Nearctic species in Sou.th America (Patter- son and Pascual, 1968). Those Nearctic forms from flaiJe XVA which closely resemble elements of the Cliapadmalalan fauna are Smilodon, Equus, and Hemiauchenia . The combination of correlative taxa of both Nearctic and Neotropical ot. igir! makes the resemblance between the Blancan fauna of Florida and tlie Chapadmalalsm fauna of Argeritinei extraordinarily strong . The presence in the Ilaile XVA faunft of shark vertebrae that are simiJar in preservation to the other vertebrates (aiid therefore probably not intrusive) indicates that the sea stood nearby during the tjiae of deposition. Even thovtgh shax-ks are partially able to tolerate fresli water, their presence in large enougli numbers to become part of the fos::jJ fauna suggests that the sea was nearby. During the 134 interg.lacial stages of the Pleistocene sea levels were high. Inland localities of higher elevation such as Haile XVA (90 feet above present sea level) which were near the sea must represent interglacial deposits. This evidence, in combination with the Blancan age of the mammalian assemblage, iiidicates that the Haile XVA deposit represents the Aftonian Interglacial . PALEOECOLOGY A complote paleoecological study of an area is iioL possible by examining a small sample of its vertebrates. That tlio Haile XVA fossil mammal assemblage is not n rep- resentcitive sample of the mammalian coimminity which once lived in that region is shown by the relatively small number of individueils present and the lack of a significant variety of small mamnial species. The localjry seems to represoit an aquatic local situ- ation such as a lake or a stream bed. This is indicated by the largo amount of fish, amphibian, and aquatic reptile remains recovered from the site. Among the fish v.re three species of catfish as well as Ami a, Centropomus, Lepisosteus, and a host of other as yet unidentified species. These forms avo presoiitly n.ot found in sinkh.oles containing v/ater, but live in o^^c^n ponds or streams. The various turtles whicli have been collected at Haile XVA, including Chrysemys platy- niarginata, Trionyx, Kinosternon, and Chelydra. also indicate an ciquatic local ciivironment . The box turt-le Terrapene is also represented in the fauna but vvas probably trapped in 135 136 the same inannc: as the terrestrial nianmials . The Amphibia include both anurans and urodeles, and further suggests the aquatic situation as most of these forms rarely travel tar from water. Among the birds present at Haile XVA arc a duck, a heron, and a grebe. A thorough study of the birds is presently being conducted by Mr. Kenneth Campbell, a graduate student in zoology at t'ne University of Florida. Altliougli the non-mammalian fauna has not been studied in detail, a preliminary analysis is enough to indicate sampliiig of an open aquatic environment of dex^osition. The raarrraals divide readily into two ecological components, a smaller group of aquatic types and a larger group of terrestiisl taxa. Castor, as is well known, inhabits medium to large permanent bodies of flowing water. Pteronura inhabits streams from the Guianas to Argentina (Walker, 1960) . Its surrounding liabitat is probably secondai-y to its stream habitat as it lives in suitable rivers which flov/ through a variety of habitat types. These manmals, therefore, are ecologically allied witli tlie turtles, fishes, and water birds, as primary inhabitants of the aquatic site. The otlier manup.als ] ived on land adjacent to the site of deposi- tion . Before considering the terrestrial community, let us further »..onsid?r the aquatic situaticjn. In attem£)ting to 13 7 interpret the aquatic cnvironraant of: deposition at Ilaile X^'A, the following possible situations are considered: (1) a cave containing a pond into vhicli animals wandered; (2) a quicksand or mud mire adjacent to a slower moving por- tion of a strecun which trapped animals as they came to dri)ik; (3) the spring-head of a rivor into which animals fel] . Thcce possibilities arc discussed below: (1) The cave pond hypothesis seems unlikely for tv.'o reasons. First, the very coarse nature of the principle fossiliferous horizons suggests more rapid transportation of sediments than would occur in a cave pond. Second, the abundance of inhabitajfits that prefer open moving v/ater seems to preclude the cave poiid hypothesis. (2) A quicksand or mud trap would tend to select for larger animals as the smaller ones would be less likci^y to become bogged down. However, it would not be expected to a Jarge and var5ed aquatic fauna, nor would it normally contain such coarse sediments as those of the Haile XVA deposit , (3) The spring-liead hypothesis appears quite appealing. It could explain the accumulation of a great numlDer of large terrestrial animals and ihe relative scarcity of smaller terrestrial ones. Tlici presence of a few dense elements of Siqmodc^n, Scalopu^-;, Cryptotis, a}id Petauria would have to 138 be ascribed to fortuitous burial. "Tlie only ether smaller maii-UTials in the fauna are Castor and Pteronura, and because these are normally aquatic animals anx^v/ay, they should occur throughout the stream. All the rest of the mammals in the Haile XVA fauna are large terrestria.1 forms, the types one v/ovtld expect to be trapped by the treachiorous slopes and abrui^t limestone ledges which often occur arouiid Florida springs. Next, we may consider the nature of the terrestrial community adJ£K:ent to the spring-head site of deposition. The terrestrial anim.als sampled consist of a few small manimals and a greater variety of laj-ge mamjnals. Ihe presence of wide ranging forms such as Hemiauchenid , Plesippus, Nannippus, and SmiJ.odon and eurytopic forms such as Odocoileus, Cryptotis, Scalopus , and Sylvilagxis do not reveal much about the paleocology of the area. These forms are present in most of the North American Blancan faunas and probably ranged thi ough many heibitats. Several taxa in the Haile XVA fauna indicate the jpres- erice of a forest community adjacent to the site of dei^osi- tJon. Tlie presence of Mylohyus lends critical svipport to this interpretation. Lundelius (3 960) reviews the evidence thai- strongly implicates Mylohyus as a forest dweller o He also includes Castor , Tapirus, and Glaucomys (among others) 139 Qs forest forms. Both Castor and Tapirus arc present in the Hailo XVA fauna, and the fjying squirrel Petauria (like Glaucomys) certainly indiccites a forest environment. Glosso- therium also presumably favored forested areas. Dasypus and Kragl.i evicliia were probably restricted in North America to the Gulf Coastal Plain by their need for moderate to warm temperatures and high rainfall. Tli.is is indicated by their absence from all other North Jimerican Blancan f avmas . In South America, Kraglievichia seems to have been largely restricted to the northern tropical por- tion of the continent. These edentates suggest a tropical or subtropical climate for central Florida during the 7\ft(jni- an . In sumjnary, the geologic and paleontologic evidence seems to indicate that the Ilailc XVA site lay at the spring- head of a coastcil stream which flowed through a tropical or subtropical forest. ZOOGEOGRAPHY The Haile XVA fauna is tlis first Gulf Coastal Plain Blancan fauna to be described. Apparently its unusual faunal assemblage ODiTipared to other Blancan faunas is a reflection of its location. All other described Noi-th American Blancan faunas probably represent upland areas which v/ere peripheral to the subtropical corridor through v>/hich most of the South American emigrants passed. Among the edentates, only Glossotherium was net re- stricted to the tropical portion of North America. This is indicated by its presence in the Blanco (Texas) and Broad- v/ater (Nebraska) faunas. In South America this genus was pjrobably distril:)Uted throughout the continent as the presence of G_^ chapadmal ens i s in Argentina would indicate. They were no doubt also present in the more equatorial regions of South Mierica during Plio-Pleistocene time al- though fossil deposits of corresponding age in these areas have not yet been found. 11-(roughout the later Pleisto- cene this genus v/as widespread on both continents. In contrast to G 1 o s s o t h er i um , Kraglievichia was 140 141 ai:}parc)itly restricted to tiie tropical or subtropical por- tions of North Aoierica during Blancan time. Their iota] absence from all other NorLh American Elancan faunas out- side Florida bears thj.s out. The presence of these forms in Pliocene deposits of Argentina suggests that this genus was able to move into more temperate climes in that continent, something that it api^arently never accompld.shed in North America. Apparently the evolution from Kraglievichia to Ch 1 amy th er ium involved, among otlier thi)igs, an increased tolerance to colder weather, perhaps reflected by the greater size of Chlamytherium. Ch 1 amy th er iuiin was able, later in the Pleistocene, to disperse northwestward from the Gulf Coeistal Plain and in South America it v/as able to spread southv\?ard throughout Argentina. It never, howevoj:, attained the v/idespread distribution on both continents enjoyed by Glossotherium. The genus Dasypus seems to have been more restavictcd to tropical regions during the Elancan than were Glossotherium and Kraglievichia . It, too, was able to move out of tlie Gulf Coastal Plain later in tlie Pleistocene, reachi.ng as far west as Texas and as far north as Missouri and Oklahoma. Tlie probable moclianism allowing for tolerance to colder weather may a] so have beer> an increase in size as jn Chla- myttierium. Its drastic size decrease in latest Pleistocene 142 time (assuming _D_^ bel.lus is ancestral to D_ novemcinctus) could account for its present southern distiibutio'i . Dasypus bellus is not knowii from South American fossil deposits ailthough there is litt.le doubt that it originated there. This is good evidence that it was totally restricted to tropical areas on that continent, w"nere fossil sites happen to be rare. The unusual distributional pcittern of Petauria (early Pleistocene of Florida and Bavaria) indicates a wide dis- tribution complicated tooth group of flying squirrels. Appa.r-ontly this group was restricted to the eastern portio:! of Noirth A-me3:ica while an alternate form existed as an ecological equivaD.ent in the west. James (1953) has re- ported flying squirrels of the simple tooth group in deposits of western North America. SUMMARY The Haile XVA mainmals represent a unique fauna as it samples one of the few Gulf Coastal Plain Blancan faunas, and is the- only such fauna v/hich has boon extensively studied to date. Among the important new pieces of information brought to light is the presence in North America of Glossotherivim chapadmalensis , which was heretofore known only from Argen- tina. It is possible that this form gave rise to G_. ro - bustus in South America and G^_ harlani in North An^crica. Apparently, Glossotherium was able to apread throughout North America while the other edentates represented in the Haile XVA fauna were restricted to the Gulf Coastal Plain throughout most, of the Pleistocene. The occurrence of Kraglievichia paranensis at Haile XVA and Santa Fc I is the first record of this geniis in North America. The presence of this species and of G. chapadmalensis strengthens the correlation bct\^?een North and South American faunas of Blancan and Chapadmalalan ages. Faix.1y rapidly evolving mammals in the Haj.le XVA 143 144 fauna wh.i.ch are conspecific with those in the Chapadma J.alan fauna and in soiiie of the Auricanean fannas are good evi- dence of closely similar ages. The earliest knovm occurrence of Dasypus bellus is at Haile XVA. This creature undoubtedly originated in South America although its ancestors are r^ot known. Further discovery of northern South American fossil deposits will more tlian likely turn up the ancestral forms as it was probably restricted to the equatorial portions of the con- tinent where fossil deposits are rare. The Old V\forld flying squirrel Petauria gives the feiuna an unexpected link with the European faunas. The presence of this form reflects the Miocene or Pliocene m.igration of European forms to North TVinerica (Simpson, 1947). That more remains of these forms are not known is probably due to the natural scarcity of small arboreal forms in the fossil re- cord, as well as the possible restriction of this lineage of flying squirrels to the forests of eastern North America. The presence of Pteronura in the Haile XVh fauna marks t?ie first record of this large aquatic otter as a fossil. The possibility that the Haile :<^>7A form, was ancestral to both Pteronura and Lutra should be considered. The presence of Mylohyus f loridanus at Haile XVA inarks one of ffie earliest records for this genus, though its ancestor, Prosthenops, is known from the Pliocene. Platy- qonus occurs at Santa Fc I, suggesting a difference in ecology. If the tentative assigniaent of the Hai.le XVA camelid to liemj.auchenia macrocephala is correct, it marks the ea];liest record for this species anywhere. The typical Blancan species IK_ blancoensis occurs at Santa Fe I suggesting a possible age difference between the two localities. The remaining members of the fJai le XVA fauna of fairly typical representatives of the North American Blancan (early Pleistocene) fauna. Such characteristic taxa as Plesippus, Nannippus ph logon , and Siqmodon medius strengthen the assign- ment of a Blancan age to the fauna. The presence of sharks and brackish v/ater fishes in- dicates that the sea stood near Haile during the time that Haile XVA was deposited. Presumably tliis stand of the sea near 90 feet re;presents the Altonian interglacial level. Although the bulk of the fauna at Haile XVA represents an aquatic stream, an adjeicent terrestrial commvniity is also well represented. Among the mammals only Pteronura and Castor represent the aquatic cominunity. T)\e terrestrial comniunity includes several clear indicators of a tropical forest situation, these being Myloliyus, Petauria, and the edentates . LITERATURE CITED Tikersten, W. A. 1970 Interpretat j.on of sediments and verte-- brate fossils in fill of Red Light bolson, southeastern Hudspeth County, Texas. Soc. Econ. Paleo. Miner., Permian Basin Sec., 1970 Field Trip Guidebook, Publ. 70-12: 82-87. Alt, D. and K. K. Brooks. 1965 ZVge of the Florida marine terraces. Journ. Geol., 73(2): 406-411. Ameghino, F. ]875 Notas sobre algunos fosiles nuevos de la Formacion Pampeana. Ameghino Dbras, 11: 13-17. . 1887 Apuntes preliminares sobre algunos mami- feros extinguidos del yacimiento de Montehermoso exis- tentes en el Museo de la Plata. Estracto, Entraga 1, Tomo 1, Boletin Museo La Plata Imprenata, El Censor, 8: 339-353. . 1902 Notices preliminaires sur des mamraiferes noveaux des terrains cretaces de Patagonie. Boletin de la Academia Nacinal de Ciencias de Cordoba, Tomo 17: 5-70. . 1904 Nuevas especies de mammiferos, Cretacios y Terciarios, de la Republica Argentina, Ann. Soc. Cient. Arg., 16: 193-208. Auffenberg, V/. 1955 Glass lizards (Ophisaurus) from the Pleistocene and Pliocene of Florida. Herpetologica , 2: 133-136." . 195 7 A_ no t o on an unusually complete specimen ofPasypus bellus (S-impson) from Florida . Qua r t . Jcurn. Florida Kci\J. Sci., 20(4): 233-23^7. . 1958 Fossil turtles of the genus Terrapene in Florida. Bull. Florida State Mas., 3(2): 53-92 146 147 Bad or, R. S. 19!j7 Two Plcistocono mamir.nVi.an faimas frorn^ Alachua County, Florida. Bull. Florida Staue Mas., 2(5): 53-75. Bcirbour, E. II. and C. B. Schultz. 1937 An early Pleistocone fauna from Nebraska . Amer . Mus. Nov! bates. Mo. 942: 1-10. Br is son, M. 0". 1762 Regnum animale in classes IX distributu.m sive synopsis niethodica. Editio Jilt era. Leiden, Thcodorum Haak: 1-296. Brodkorb, P. 1963 A giant flightless bird front the Pleis- tocene of Florida. Auk . , 80(2): 111-115. Bronn , M. 1838 Lethaea geognostica. .Stuttgart, Vol, 2 ; 545-1350. Brown, B. 1903 A new genus of ground sloth from the Pleis- tocene of Nebraska, Paramylodon nov. gen. Bui 1 . Airier . Mus. Nat. Hist., 19: 569-583. Cabrera, A. 1929 Una revision de los mastodontes Argentinos. Rev. Mus. La Plata, 32: 61-14d. Castellanos, A. 1927 Breves notas sobre los clamiterios ■ Publicacion Centre Estudiantes de Ingenieria de Rosario: 1-8. . 193 7 Anotaciones sobre la linea filogenotica de los clamiterios. Publicacj.ones de la FacuD '.:ad de Ciencias de la Universidad Nacional del Literal. Serie Tecnico-Cientif ica Publicacion 8: 1-13 5. . 1946 Una nu eva especie de clamitorio VassaLTia maxima n. sp. Publicacioncs del Institute de Fiso- grafia y Geologia, Unjversidad Naciona.T de Litor^il, 6(26): 5-47. . 19 57 Nuevos generos de clamiterjo y do gervo y una nueva especie de Plohophoru.s . Amcgh in iana , 1(3): 5-14'. Cookcj, C. W. 1945 Geology of Florida. Florida Geol. Surv. Bull. 29: 1-339. 148 Copo, E. D. 1B80 On the extincb cats of Ainerica. Amer. Nautrali&t, 14: 838-858. . 1892 A contribution to the vertebrate paloon- toloqy of Texas. Proc. Amer. I-hilos. Soc . , 30; 123-131 . 1893 A preliminary report on the vertebrate paleontology of the I.lano Estacado. Geol. Surv. Texas Fourbh Annual Report: 1-136,. . 1899 Vertebrate remains from Port Kennedy bone deposit. Journ. Acad. Nat. Sci. Phila., 11: 193-286. Dawson, M. R. 1958 Later Tertiary Leporidae of North Ameri- ca. Univ. Kan. Paleo. Contr., Vertebrata, 6: 1-75. Dehm, R. 1952 Altpleistocane sauger von Schernfeld bei Eichstatt in Bayern. Munich. Bayer ische Staatssaminlung fur Feilaontologio und Ilistorische Geologie. Mitteilun- gen, 2: 17-61. Deperet, C. 1897 Les animaux Pliocenes du Rou.Gsillon„ Memoires de la Socie'te' Geologique de France, Paris. Palcontologie, 3: 1-164. Downs, T. and J. A. White. 1968 A vertebrate faunal suc- cesion in superposed sediments from late Plio con e__t q middle Pleistocene in California. 23rd Int. Geol. Congress, 10: 41-47. Friclv, C. 1937 tlorned ruminants of North America. Am.er. Mus. Nat. Hist. Bull. 69: 1-669. Gazin, C. L. 1936 A study of the fossil horse remains from the upper Pliocene of Idaho. Proc. U. S. Nat, Mus., 83(2985): 281-320. , . 1942 The late Cenezoic vertebrate faunas from. j"jlg. .S^^-" Pedro Valley, Arizona. Proc. U, S. Nat. Mus., 92(3155) : 475-518. Gidley, J. W. 1922 Preliminary report on fossil vertebrates g-l_the San Pedro Valley, Arizona. U. S. Geol. Surv., Prof. Paper 131-E: 119-131. Gray, J, E. 1821 On _the_j25j^^5jL_?..??^5:29L§i!I?Pj^_Q^ animals. London Med. Reposit., 15(1): 296-310, 149 . 1837 DoGcription of some new or little known Mammalia, principly in the British Museum Collection. Mag. Nat. Kist. New Series, 1: 577-587. 1867 Notes on the skulls of some hares (Lcpori- ^■^ig-)_^Qd picas (Lagomyidae) in the British Museum. Ann. Mag. Nat. Hist., 20: 219-225." lla.ll, E P.. and K. R. Kelson. .1959 The mammals of North America. The Ronald Press, New York, 1-1079. Harlow, R. F. and F. K. Jones. 19G5 TIt c wli i t e - 1 a i ] e d deer in Florida. Florida Game and Freshwater Fisli Commis- sion Tech. Bull. 9: 1-240. Hay, 0. P. 1917 Vertebra ta mostly from stratum No. 3, Vero, Florid a, together with descriptions of n ew species . Florida Geol. Surv . , 9th Annual Rept . , 43-68. . 1928 Again on Pleistocene man at Vero, Florida. Journ. Washington AcL.d. Sci., 18(9): 233-241. Hibbard, C. W. 1937 An upper Pliocene fauna from Meade County, Kansas. Trans. Kan. Acad. Sci., 40: 239-265. . 1941 Manmials of the Rexroad fauna from the upper Pliocene of southwestern Kansas. Trans. Kan. Acad. Sci., 44: 265-312. Hibbard C, W. and V7. W. Dalquest. 1962 Artiodactyls from the Seymour Formation of Knox County, Texas . P ap . Mich. Acad. Sci., 48: 83-99. Hirschfeld, S. E . and S. D. Webb. 1968 Plio-Pleistocene Megalonychid s.loths of North America . Bull. F 1 a . State Mus., 12(5): 213-296. Hoffstetter, R. 195? Lcs mammi feres Pleistocenes de la Repub .1 i qu e d e 1 ' Equ a t eu r . Mem. Soc. Geol. France, 66: 1-391". Hugueney, M. and P. Mcin. 1966 Les rongeurs Pliocenes du Roussillon, dans collections Lyonnaises. Lyons Uni- versite, Faculte dos Sciences, Labcratoire de Geologic, Travaux, Nev/ Series, 13: 243-266. 130 Jaracs, G T. 1957 An edentate from the Pleistocene of Texas. Joxarn. Paleo,, 31(4): 796-808. . 1963 Paleontology and nonmorinc stratigr^iphy of the Cuyama Valley badlands, California. Part. I. Geology, faunal inte rpretations, and systematic de- scriptions of Chiroptara, Insectivora, and Rodent ia . Univ. Calif. Publ. Ceol . Sci., 45: 1-3 54. Kellogg, L. 1911 A fossil beaver from the Kcttleman Hills, California. Univ. Calif. Publ., Bull. Dept. Geol., 6(17) : 401-402. Kinsey, P. E. (in press) A new species of the peccary, Mylohyus, from the early Pleistocene of Florida . Kraglievich, L. 1925 Cuarto nuevos gravigrados de la fauna araucana Chapadmalense . Anales del Museo Nacional de Historia Naturel, 33: 215-235. . 1928 jMylodon darvvinii Owen, es la especie genotipo de Mylodon Owen. Rectif icacion de 1 a nomen- clatura generica de los milodontes. Physis, 9: 169- 18 5, . 1934 La antiquedad Pliocena de las faunas de Monte Hermo so y Chapadmalal, dedicidas desu comrjara t^i o n con las que le precedieron y sucedieron. Mon t i e vi deo , 1-136. . 1946 Sobre camelidos Chapadmalenses . Insti- tute del Museo de la Universidad Nacional de La Plata, Notas del Museo de La Plata, Tomo 11, Paleontol.ogia 93: 317-329 . 1948 Smilodontodorj riggi, n. gen, n. sp., vra nuevo y peaueno esmilodonte en la fauna Pliocena de Chapadir.alal. Revista del Museo Argentina de Cienci.as Naturales, Ciencias zoologicus, 1(1): 1-44. Kretzoi, M. 19 59 Fauna und f aunenhorizont von Csarnota. Jahresbericht der Ungarischen Geologischen Anstalt, 1-385. Leidy, J. 1889a Description of the verj^ebra.te remains from Peace Creek, Florida. Trans. Wagner Free Inst. Sci., 2; J 9-32. .1^)1 .1.8S9b FosBJl vcrtohrates from Florida. Proc. Acad. Nat. Set. Phila., 9G-97. Linnaeus, C. 1758 Systoira Naburae per rogna tria naturae, s_ecundum g_lasPGS, ordines, genera, species cum chf'irac- ■teribus, dif f erentiis, synymis, locis, Editio decinia, Reformata. Stockholm, Laurentii Salvii, 1: 1-824. Lund, P. V;. 1838 Blik paa Brasiliens Dyreverden _f or Sidste Jordoinvaoltninq. Overs. K. Danske Videnskebernes Selskabs., Forh . , 8: 29-144. Lundelius, E. L. 1960 Mylohyus nasutus, long-no.sed peccary of the Texas PJ.eistocene. Bu.ll. Texas Mem. Ilus., 1: 9-40. MacNeil, F. S. 1949 PJ.eistocene shorelj.nes in Florida and Georgia. U. S. Geol. Surv., Prof. Paper 221-F: 95-107. Martin, R. A, (in press) Fossil mammals from the Coleman IIA Local Fauna, Citrus County, Florida . McGrew, P. 0. 1944 An early Pleistocene (Blancan) fauna froi.i Nebraska . Field Mus. Nat. Hist., Geol. Se?:., 9(2): 33-66. McKenna, M. C. 1962 Eupetaurus and the Mvjj^g_p^etauri_sy^ sciurids . Amer. Mus. Novitates, no. 2104: .1-38. Merriam, J. C. and C. Stock. 1932 The Felidae of Rancho La Brea. Carnegie Inst. Washington Publ. 422: 3-231. Moreno, F. P. and A. Mercerat. 1891 Exploracion arqueo- logica de la Provincia de Catamarca. Paleontologica , en dichma misma Revista, del Museo de La Plata, 1: 222-234. Ocsch, P.. D. 1967 A preliminary investigation of a Pleisto- cene vertebrate fauna from Crankshaft Pit, Jefferson County, Missouri. National Speliological Soc. Bull., 29(4): 163-185. Owen R. 1840 The zoology of the voyage of H. M. S. Beagle Part 1 : Fossil Man Jiu' Jia . London, 13-111. . 1842 Description of the skelet-on of an extinct 9i£iP}'^''^^-c sloth, Mylodo;i robustus 0\-/en._. . London . 1-376. 152 Patterson, B. and R. PaBqual . ] 968 EV-OluJl-lpn of mamma ls_on_ southorn continents. V. T'ne fossil manuiial fauna of South America. The Quart. Rev. Biol., 43(4): 409-451. Fatten, T. H. 1963 P'ossil vertebrates from Miller's Cave, Llano County, Texas. Bull. Texas Mem. Mo.s . , 7: 1-41. Paula Couto, C. de 1956 Resume de memorias de Lund sobre as cavernas de Lagoa Santa e seu conteudo animal . Universidade do Brasil Museu Nacional, Publicacoes avulsas, 16: 1-14. Porta J. de 1962 Edentata del Miocene de La Venta (Colom- bia) . I Dasypodoidea y Glyptodontoidea . Boletin de Geologia, Dniversidad Industrial de Santander, 10: 5-23. Ray, C, E. 19 57 A list, bibliography, and index of the fos- sil vertebrates of Florida. Florida Geol. Surv., Spec. Publ. no. 3: 1-175.. Robertson, J. S. (in press) The fossil Bison of Florida, Rovereto, C. 1914 Los estratos Araucanos _y_ s^i3s__fj^s j^lej^.- Anales del Museo Nacional de Historia Naturel de Buenos Aires, 25: 1-236. Savage, D. E. 1951 Late Cenozic vertebrates in the San JLJ^g-ng isco Bay Region . Univ. Calif. Publ., Bull. Bept . Geol. Sci., 28(10): 215-314. Schultz, J. R. 1938 A late Cenezoic vertebrate fauna from the Coso Mountains, Inyo County, California . Ca rn egi e Inst. Washington Publ., no, 487: 76-109, Sellards, E. H. 1915 Chlamytherium septentrionalis , an edentate from the Pleistocene of Florida. Amer . Oourn. Sci., 4(40): 19-31. . 1917 On the association of human remains and extinct vertebrates at Vero, Florida. Journ. Geol., 25(1) : 4-24. Simpson, G. G. 1928 Pleistocene manmials from a cave i n Citrus County, Florida. Am.or. Mus . Novitates no. 328 1-16. 153 1929 Pleistocene mammalian fauna of the Seminole Field, Pinellas County, Florida. Amor. Mus. Nat. Hist. Bull., 56(8): 56]-599. 1930 Holmcsina septentrionalis, e.>:tinct giant armadillo of Florida, i-'micr. Mus. Ncvitates no. 442: 1-lb. . ].945 The principles of classification and a classification of manunals. Bull. Amer . Mus. Nat. Hist., 85: 1-350. . 1947 Holarctic mammalian faujias and continentail relationships dui-ing the Cenozoic. Bull. Geol. Soc. Araer., 58: 613-888. Slaughter, B. II. 1961 The signific^mce of Dasypus bellus (Simj;json) in Pleistocene loca] faunas. Texas Journ. Sci., 13: 311-315. Stirlo)!, R. A. 1935 A review of the Tertiary l>«avers^. Uni . Calif. Publ., Bull, Dept . Geol"., 23 (13) :~ 391-458 ." Stocjc, C, 1925 Cenox.oJc gravigrade edentates of western Nortii America, v/ith special reference to the Pleisto- ■ggj}fi i^'^g.^-^.'^^y'^hidae and Mylodontidae oJ_Jj_ancJ20_La Brea,, Carnegie Inst. Washington Publ. no. 33]: 1-206. . 1928 Tanupolama, a nev; genus of llama from the ilJ-.gAg,lL9iri:P:!:-_ ,9^. Cal if ornia . Carnegie Inst. Washington Pu}5l. no. '393: 30-3 7. Strain, V*'. S. ] 966 Blancan mammalian fauna and Pleistocene ±.QIiTL^L4QJ[L-'-/.-.ll^'^^^'Peth County, Texas. Bull. Texas Mem. M'ln., 10: .1-55. Su 1 im.sh i , A . 1964 Pliocene Lagomorpha and Rodentia from Weze I (Poland) . Acta Palaeontologica Polonica, Polska Akademia Navih , Komitet Geologiczny, 9(2): 149-240. Ta linage, R. V. and G. D. Buchanan. 1954 Tlie armadillo (Dasypus novcmcinctus) . A revio\7 of its natural his^ tory , ^co^j2Siy.j-J^}}^J^EZj^.3^Q^L^EPr}i9S^^^^''^^^^ P^'^Y^ iology . The Rice Inst.; l:utc Pamphlet, 4 3(2): 3-135. V^alker, j; . P. 19C0 Maniruals of the World. Tlie Jo'ms Ilopjiins Press, Baltimore, vol. 2: 647-1500. 154 VJoaver, VJ. G. and J. S- Robertson. 1957 A re-evaluation of fossil tu rt.les of the Chrysemys scripta group . Tu 1 ane Studies in Geology, 5(2): 53-66. Webb, S. D. (in press a) Chronology of Pleistocene land maiamals in Florida , . (in press b) Pleistocene llcunas of Florida with notes on the Laimini . Wiiite, VI. A. 1958 Some geomorphic featijres of central penin- sular Florida. Florida Geol. Surv., Geol . Bull. 41: 1-42. Wiegel, P. 1952 Fossil vertebrates of Vero, Florida . Florida Geol. Surv., Spec. Publ . no. 10: 1-59. W i 1 s on , R . W . ] 9 3 8 Pliocene rodents of western North Ameri- ca . Carnegie Inst. Washington Publ., no, 487: 22-73. Winge, H. 1915 Jordfundne og nulevende Gumlere (Edentata) fra Lagoa Santa, Minas Geraes, Bras i lien. E. Museo Lundii, 3(2): 1-319. VJoodbiirne, M. 0. 1955 Equid remains fx'om the Sonoma Vol- canics, California. Bull. So. Calif. Acad. Sci,, , 65(3): 185-189. BIOGlVvPHICAL SKETCH JcGso Stoadman Robertson, Jr., was born in St. Augus- tine-, FJ.orida, on Doceniber 11, 1934. He attended public schools of Nev/ Smyrna Beach and St. Lucie County High School. He entered the United States Navy in 19132 and served for four yeais. In 1956 he entered Jacksonville University from v;hich institution he was awarded a Bachelor of Science degree in biology in 1960. He taughit in the public high scliools for tv7o yeaxs in Jacksonvil.1 e and entered Graduate School at the University of Florida in 1962. He received a Master of Science devgrce in zoology from this institution in 1955, After another year in public high school teaching, lie joir.cd the faculty of Jacksonville University in 1965. In 19bfi he again enrolled in the Graduate School at the Uni- versity of J'^lorida where he has been pursuing work toward a JDoctor of i'liilosophy degree. lie is married to the former Kiss Shirley Joyce Harth of Now Tov/j-i , West Virginia, and has one son, Danny. 155 156 He is a member of The Society of Vertebrate Paleon- tologists, American Society of Mammalogists , Tri-Peta Bio-- logical Society, and The American Association of University Professors. I cert.iry that I have read this study and that in my opinion it conforms to acceptable stJ^ndards of scholarlir presentation and is fu].ly adeciuate, in scope and as a dissertation for tlie df^^nX'^aXof Doctor, As.'aOciaLC' Professor of ZooJofiy 'jji' I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is ful.ly adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.. '(p4^(U^ Thomas H, Patton Associate Professor of Zoology I certify that I have read this study and th^lL in my opinion jt conforms to acceptable standards of scholarly- presentation and is fully adequate, in scope and quality, 9s a dissertation for the degree of Doctor of Philosophy. H. K. Brooks /associate Professor of GeoJogy This dissertation was submitted to the Dean of the College of Arts i\''\6. .c;ciences and to th.c Graduate Council, and was accepted as partial ful f il])-aent of the requirements for the degree of Doctc-r of Philosophy. Dccem};e3:, 1370 Dean, College of Arts and 'Sciences Dean, Graduate School H' ^ff MJ- f Ja.y-1j^ /<^,^.6. iSS \ 0^ ,e.