BULLETIN OF THE' FLORIDA STATE MUSEUM BIOLOGICAL SCIENCES Vol.ume 12 Number 5 PLIO-PLEISTOCENE MEGALONYCHID SLOTHS OF NORTH AMERICA Sue E. Hirschfeld and S. David Webb UNIVERSITY OF FLORIDA G,ain es v~i lie 1968 Numbers of the BULISETIN ,Of THE Mid#RIbA STATE MUSEUM'are #fib- lishid, at irregulgtr ird*Fals. Volumes contain about 800 Rages and are not necessarily completed in any one calendar year. WALTER_AUFFENBEEC. Managing'.Editor OLIVER L. AUSTIN, JR., Editor Consultants for this issue: BRYAN PATT-ERSON DONALD E. SAVAc~ Communications concerning purchase or exchange of the publication and :all manuscripts should be addressed to the Managing Editor· of the Bulletin, Florida $tate Museum, Seagle..Building, Gainesville, Flbrida. 32601. Published Dedember 116, f968 Pride· for. this issuB $1.20 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 21S PLIO-PLEISTOCENE MEGALONYCHID SLOTHS OF NORTH AMERICA SuE E. HIRSCHFELD AND S. DAVID WEBB SYNOPSIS: Although ground sloths appear in South America in Mustersan (late Eocene) time, they are not found in North America until the Hemphillian ( mid- die Pliocene) when three genera appear, two of them megalonychids. It now appears that the genus Megalonyx was widely distributed through North Amer- ica by that time. The most primitive is M. mathisi, new species, from the upper Mehrtens formation, Merced Co., California. Additional material of M. cum- dens Matthew from the Upper Snake Creek channels of Sioux Co., Nebraska, differs in its shorter cheek teeth and narrower caniniforms from later species. Likewise a fine collection of late Blancan ( early Pleistocene) M. leptostomus Cope from Cita Canyon, Randall Co., Texas, exhibits small cranial, dental and podial differences from typical Iate Pleistocene species of Megalonyx, M. lep- toni/x C Marsh) is a nomen dubium. The general features of MegatonYx are reviewed and the great variability of its dentition is analyzed. The earliest megalonychid from North America is a new genus and species, Pliometanastes protistus, from early Hemphillian deposits at McCehee Farm, Ala- chua Co., Florida. Relatively complete material of this new genus is described. Another new species from New Mexico, ?P. galushai, and other records from western North America are presented. The genus bears no special relationship to Megalonyx nor to any of the West Indian megalonychids, but stems from some Mio-Pliocene stock that gave rise to all of these. i Sue E. Hirschfeld is Teaching Assistant in the Department of Paleontology at the University of California, Berkeley. She completed most of the work reported in this paper as a Master's candidate at the University of Florida. ' S. David Webb is Assistant Curator of fossil vertebrates at the Florida State Museum and Assistant Professor of Zoology at the University of Florida, Gaines- ville. His principal research is on Pliocene mammals of Florida. Hirschfeld, Sue E., and S. David Webb. 1968. Plio-Pleistocene Megalonychid Sloths of North America. Bull. Florida State Mus., vol. 12, no. 5, pp. 213-296. 214 BULLETIN FLORIDA STATE MUSEUM Vol. 12 TABLE OF CONTENTS INTRODUCTION 214 ACKNOWLEDGEMENTS 215 Megaton!/x Harlan 216 M. leptostomus Cope 219 M. Zeptonyx C Marsh) 281 Other Blancan Megatonyx 281 M. curvidens Matthew 235 M. mathisi new species 289 Other Hemphillian Megatonyx 244 Pliometanastes new genus 246 P. p/otistus new species 246 --- - ?P. gatushai new species 283 Other Hemphillian Pliometanastes 284 Other Hemphillian Megalonychidae 286 RELATIONSHIPS 287 LITERATURE CITED 294 INTRODUCTION Ground sloths appear in South America in the Mustersan (late Eo- cene), but the group is unrecorded in North America until the Hem- phillian ( middle Pliocene). Fossil records of these earlier North Amer- ican forms have hitherto been few and fragmentary. For this reason, the discovery of relatively abundant remains of two genera of Hem- phillian ground sloths in central Florida adds significantly to our knowledge of the phyletic and zoogeographic relationships between North and South American forms. The genera are Thinobadistes Hay, a mylodontid from Mirson's Bone Bed, and a new megalonychid genus from McCehee Farm. This contribution concerns the new megalonychid and a review of other early North Ainerican mega- lonychid sloths. A nearly complete skull from the late Hemphillian of California representing a new structurally primitive species of Megalonux, new material of M. curvidens from latest Hemphillian deposits, as well as an excellent collection of Aft. leptostomus from the late Blancan of Texas are described and problems of defining species of Megalon!/x on the basis of isolated teeth are an·alyzed. When the family Megalonychidae ( as constituted by Simpson, 1945) first appears in the Deseadan ( early Oligocene) of Patagonia, it is represented by fragmentary material. By the Santacrucian ( middle Miocene) stage in Patagonia seven megalonychid genera are distin- 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 215 guished, most of them represented by abundant material. Within th,: most completely known genus, Hapatops, 22 species have been pro- posed. Skippjng to the Pleistocene, Megalonux, the best known Pleisto- cene megalonychid, is well represented by skulls, teeth, and two com- plete skeletons. Although it is not found in South America, it ranges widely through North America, reaching north to Alaska ( Stock, 1942). In the West Indies nine genera of megalonychid sloths have been recognized ( Paulo Couto, 1967), all presumably Pleistocene in age. This substantial knowledge of megalonychid sloths in the Miocene of South America and in the Pleistocene of Nearctica and the West Indies contrasts strikingly with our ignorance of the group in the Plio- cene of the entire hemisphere. The new genus from Florida is now the most completely known Pliocene megalonychid. Its description, to- gether with the review of other early North American megalonychids, is intended to f 11 the Pliocene gap and thereby to provide new insight into the taxonomic and geographic relationships of megalonychid sloths. ACKNOWLEDGMENTS The present contribution grew from the analysis of several unpublished col- lections of early North American megaIonychids. We began with the new Hem- phillian genus from Florida in the collections of the Florida. State Museum, University of Florida ( abbreviated UF). The first specimens of this form were discovered by Florida State Museum parties under Clayton E. Ray during pre- liminary excavation of the MeGehee Farm Site in the summer of 1968. Inten- sive work began in 1964 with support from the Frick Corporation and continued from 1965 through 1967 aided by NSF grant GB 3862. Then Donald E. Sav- age suggested that we study the megalonychid material from the extensive Cita Canyon fauna that he and others had collected for the Panhandle Plains Museum at West Texas State, Canyon, Texas ( hereafter WT ) Thirdly, ( we investigated the Tertiary megalonychids in the incomparable Frick collection of Cenozoic vertebrates now in the American Museum of Natural History iii New York ( FAM ), and the Blancan material from Hagerman, Idaho in the U.S. National Museum collections ( USNM). We studied Pliocene megalony- chids in the Museum of Paleontology of the University of California at Berke- ley ( UCMP) and those in the collections of the University of Oregon ( UO) And finally we compared these earlier materials with the fine collection of late Pleistocenc Megaton!/x in the Academy of Natural Science of Philadelphia (ANSP). We wish to thank the many colleagues who received us graciously and loaned us material froni their institutions. We are particularly grateful to Malcolm C. McKenna, Bryan Patterson, D6nald E. Savage, R. J. G. Savage, Morris Skinner, Ted Galusha, S. J Olsen, J. A. Shotwell, Horace Richards, H, K, Brooks, and R. A. Edwards for their advice and assistance. 216 BULLETIN FLORIDA STATE MUSEUM Vol. 12 AIl measurements in this paper are in millimeters unless otherwise indicated. Statistical methods and abbreviations follow Simpson et al. ( 1960). Additional abbreviations of collections are AM ( American Museum of Natural History), -(DMP ( Oklahoma Museum of Paleontology), and PU ( Princeton University). Our dental nomenclature distinguishes the caniniforms from the molariforms, and .therefore the abbreviation M, indicates the flrst lower molariform tooth, which some systems label M:, Megalonyx Harlan . Megalonyx Harlan, R. 1825. Fauna Americana, Anthony Finley, Philadelphia: 201. ( Type species M. ie#ersoni Wistar, West Virginia). Morotherium Marsh, 0. C. 1874. Notice of new Tertiary mammals III. Amer. Jour. Sci., 1874: 581. (Type species M. gigas Marsh, Central California). DIAGNOSIS. - (Modified from Stock, 1925.) Skull short, broad, and deep, with anterior end truncated and strong sagittal crest. Man- dible deep, without spoutlike predental region. Dentition L ~. First tooth separated by diastema from cheek tooth series; cross-section meniscoid with inner bulge. Last superior tooth subtriangular in cross-section . Appendicular skeleton more robust than in Nothro- therium. Winglike process of- calcaneum broad; astragalus retain- ing essentially the structure seen in Hapalops and distinctly less modified than in Nothroth€rium. Astragalus possessing V-shaped notch for reception of the tibia. Shaft of fibula curved dorso-medial to distal end in region of peroneus brevis attachment. Metatarsal TV-short; metatarsal V with relatively slender lateral process. DENTAL VARIABILrrY It has long been recognized that the dimensions of sloths are among the most variable of any mammals. This is exemplified among fossils by the 22 "species" of Hapatops described from the Santa Cruz de- posits of Patagonia. Similarly in Megalonyx nearly every good speci- men has been described as a different species. Whether this variability is inter- or intraspecific is a crucial, but rarely answered, question. Simpson ( 1959) shoys.that while. considerable individual variation is manifest within a single collection of Megalocnus rodens, this varia- tion is continuous and appears to represent a single population. We present a similar analysis of all the complete adult megalonychid teeth from the.Port Kennedy Bone Cave collection in the ANSP ( Tables 1 and 2).Measurements were taken just below the occlusal surface of each tooth. Cope ( 1871) described four species of Megalonyx from this collection and, for the sake of objectivity, we have followed his assignments for all teeth used in this analysis. Table 1. STATISTICAL ANALYSIS OF THE TEETH OF Megatont/x wheatteyi 1968 H IR S C H FE LD /W E B B : P LIO -P LE IS TO C E N E SLO TH S 217 OR X S V Tooth N ML2 AP ML AP ML AP ML 0 21 85.9-29.5 19.3-16.0 32.6 17.7 1.99 1.18 6.09 6.66 C. 23 36.9-29.1 17.7-18.0 32.5 15.5 2.12 1.29 6.52 8.34 6 15.7-14.7 20.0-17.0 15.8 19.0 .88 1.22 2.16 6.43 Mi 14 15.9-14.4 28.0-20.0 15.4 21.5 .66 .77 4.29 8.57 M' 9 15.9-14.4 23.8-19.4 15.8 22.2 .68 1.67 4.20 7.58 M, 10 16.4-18.7 24.8-20.7 15.3 22.9 .72 1.10 4.72 4.80 M' 7 16.5-14.6 25.9-21.0 15.5 28.5 .73 1.87 4.71 7.96 M, 9 17.9-14.6 23.1-20.8 16.2 21.9 1.09 1.02 6.72 4.66 M' 9 12.8-11.0 19.2-17.0 11.4 18.4 .35 .67 3.06 8.65 1AP = anteroposterior. IML = me'diolateral. Table 2. STATISTICAL ANALYSIS OF THE CANINIFORM TEETH OF Megatonyx wheatleqi, M. tortu/us, AND M, loxodon OR X S V ·Tooth N ML' AP ML AP ML AP ML 0 24 35.9-29.5 19.3-16.0 32.7 17.7 188 1.08 5.75 6.10 24 86.9-29.1 17.7-18.0 82.5 15.4 2.08 1.27 6.40 8.25 1AP anteroposterior. 2ML mediolateral. 218 BULLETIN FLORIDA STATE MUSEUM Vol. 12 Two sets of calculations were made on these teeth: The first uses just those teeth Cope designed as M. wheatle!/i; the second includes all adult sloth teeth from the deposit, among them one lower and two upper caniniforms of M. tortulus and one upper caniniform of M. loxodon ( the specimen of M. sphenodon was omitted because it is a juvenile). The second set of calculations yields coefficients of variation that are not unusually high and graph in a homogeneous cluster of points ( Fig. 1). This strongly suggests that the Port Ken- nedy Bone Cave specimens belong to a single population. 5 10 15 20 2540 35 =11 % - 35:0: .X< 30 ]--' :. - 30 25 - - 25 Le ng th 20 - - 20 0 Megalonyx wheotleyi 15 - o Met;alonyx loxodon - 15 x Megolonyx tortulus 10 - - 10 5- -5 . rf 1 0 5 10 15 20 25 Width FIGURE 1. Size distribution of upper caniniform teeth of adult Megalon!/x wheatlegi, M. Ioxodon, and M. tortulus from Port Kennedy Bone Cave, Pa. Measurements taken just below occlusal surface. 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 219 In study of megalonychid teeth too little attention has been paid to ontogenetic variation. Juvenile specimens of Meg(don!/x and of Recent tree sloths have consistently failed to yield deciduous teeth. Owen ( 1840-45) describes a newly erupted ( permanent) tooth of Brad!/pus as conical and completely covered by cement. As the tooth erupts, the cement wears away and exposes the dentinal layers. We have observed in a series of X-ray plates of a young Choloepus ho#- nmni that in younger stages the teeth taper sharply from root to crown, whereas in more mature stages the sides of the teeth become nearly parallel. The same variations occur in Megalon!/x teeth. The length of the inner dentinal layer may also serve as an age indicator. For example, in a tapered lower caniniform of a young Megalonyx from Devil's Den, Florida, the inner dentine layer occu- pies only the lower 10 per cent of the tooth length, but in a parallel- sided mature specimen from Santa Fe I, Florida, this layer occupies 72 per cent of the total tooth length. We have observed considerable change in the shape of the occlusal surface accompanying these other dimensional changes in living Choloepus. Such changes evidently occurred in Megalohyx as well. A comprehensive investigation of Megalonyx probably will show that several of the proposed species are actually ontogenetic stages of a single natural species. Megalonyx leptostomus Cope Megatonyx Zeptostomus Cope, 1898. A preliminary report on the Vertebrate Pa- leontology of the Llano Estacado. 4th Ann. Rept. Geol. Surv. Texas: 49. TypE. - The type of M. leptostomus consists of some skull frag- ments and an upper molariform tooth from the Blanco Beds near the eastern edge of the Llano Estacado in Crosby Co., Texas. Meade ( 1945) referred several additional teeth and jaw fragments from the same area to this species. The new materials described below were collected in the lowest stratigraphic unit in the Blancan section in Cita Canyon, 31/2 miles south and 13 east of Canyon, Randall Co., Texas ( Johnston and Savage, 1955). Thus all of the material known is derived from Blancan age deposits in the Panhandle of Texas. These new materials are the most complete known. DIAGNOSIS. - Skull smaller than Megatonyx brachycephalus ( Sto- vall and McAnulty, 1950), narrow at postorbital protuberances; height and breadth of occiput considerably less than in other spe- cies, jugals flaring anteriorly, mastoid process rounded and not 220 BULLETIN FLORIDA STATE MUSEUM Vol. 12 29 20 Figure 2. Skull of Megatonyx teptostomus, WT 1956, dorsal, lateral, and ventral views. 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 221 hooklike, foramen ovale surrounded on the posterior side by a round protuberance, wider posterior divarication of palatines, adult cani- niforms with prominent median bulge, and M2 and Ms with rela- tively greater mediolateral diameter than in other species, phalanges 1 and 2 and digit III of pes not co-ossified. SKULL. - The skull, WT 1956 ( Fig. 2) is well preserved but lacks _ the pterygoids, palatines, nasal region, and most of the zygomatic arch. All teeth are missing. Sutures in the occipital region are not well fused, although those between the frontals and parietals are almost obliterated and the sagittal crest is closed. The sagittal crest is well developed as in all species of Megalon!/.r and about 112 mm long, as compared with 78.9 in M. brachycephalus and 127 in M. le#ersoni. The jugal bones Hare anteriorly and laterally, whereas in M. le#er- soni they project more ventrally and posteriorly. The postorbital pro- tuberance is poorly defined and extremely narrow. The distance across the postorbital protuberance is 92 mm, compared to 120 in M. brachycephalus ( Stovall and McAnulty, 1950) and 142 in M. hogani ( Stovall, 1940). The Cita Canyon skull presents a flat dorsal profile similar to that in Owen's specimen of M. je#ersoni ( Leidy, 1885) and in M. milleri ( Lyon, 1938), unlike the profile in Dickeson's specimen of M. ie#er- Table 8. MEASUREMENTS OF Megalon!/X teptostomus SKULL, ( WT 1956) IN MM Length of skull from posterior end of occipital condyles to posterior border of caniniform alveolus 270.0 Width of maxillary at posterior internal angle of caniniform alveolus 65.0 Width of palate anterior to first cheek teeth 42.5 Least width of palate 22.5 Width across anterior end of glenoid fossa 166.0 Minimum orbitotemporal width 72.8 Minimum postzygomatic width 117.0 Transverse diameter of occipital condyles 70.2 Width of basisphenoid anterior to basilar tubercles 48.5 Mediolateral width of process lateral to foramen ovale 16.8 Median length of palate from point of bifurcation to posterior margin 106.8 Table 4. COMPARATIVE MEASUREMENTS OF KNOWN SKULLS OF Mega~On!/X SPECIES 222 B U LLE T IN F LO R ID A S TATE M U S E U M Vol. 12 Measurement M, brachy- M. ietler- M. le#ersoni M. teidgf M. millerf M. hogane M. Zepto- M. mathid cephalut soni (Owen)' ( Dickeson)1 stomus Length of skull from occipital condyles to anterior margin of first cheek teeth 256 856 884 848 291 245 230 223 Breadth of face at post- orbital protuberance 120 127 138 128 129 142 92 Breadth of cranium at narrowest part of temporal region 86 90 102 91 75 98 78 Length of sagittal crest 78 127 127 144 95 95 112 Height of occiput from dorsal margin of fora- men magnum 70 76 70 46 51 Height. of occiput from basioccipitaI 110 110 107 72 75 Breadth of occiput at mastoid process 141 159 165 165 153 175 126 123 Length of diastema 41 50 49 40 50 Maximum width first cheek tooth alveolus 24 21 Maximum width second , cheek tooth alveolus 24 27 28 28 25 1From Stovall and McAnulty (1950). 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 228 soni ( Leidy, 1855) or in M. hogani, and M. brachycephalus, in which the anterior portion of the frontal bones arch above the nasal region. These differences in the dorsal profile may reflect age differences. Dickeson's specimen of M. ie#ersoni appears to be the oldest, as most of the sutures are obliterated and the sagittal crest is closed and fused. In Owen's specimen the sagittal crest is an open fissure and most of the sutures are fused but still distinguishable. In the Big Bone Cave specimen sutures are fused and the sagittal crest is just forming, thus indicating a juvenile. Comparison of the Big Bone Cave juvenile and Dickeson's adult suggests that, as the individual becomes older, the vaulting of the frontal region becomes more pronounced and the frontal ridges run- ning from the sagittal crest to the postorbital protuberances diverge more. Thus the flat dorsal profile of the City Canyon skull probably came from a relatively young specimen. The vertical distance from the dorsal border of the foramen mag- num to the supraoccipital crest is extremely short, measuring 45.5 mm, compared to 70.0 for M. brach!/cephalus. Similarly the height from the ventral border of the foramen magnum measures 71.7 mm compared with 107.0 in M. leid!/i and 110.0 in M. ie#ersoni. The Cita Canyon specimen is only slightly higher in this area than in the new genus from MeGehee where it is 67.8 mm. The breadth of the occiput meas- ured across the mastoid processes is considerably less than in the other species of Megalonyx, 126 mm in this specimen, 141 in M. brachy- cephalus, and 175 in M. hogani. The median occipital ridge does not extend to the foramen mag- num as in Owen's specimen of M. ie#ersoni but is lost in a depression about halfway down the occiput. As a whole the occiput is consider- ably less rugose and lacks the large tuberosities on either side of the median ridge in Owen's M. ie#ersoni. The juvenile specimen from Big Bone Cave also lacks these prominent tuberosities; the base of its mastoid process is not depressed be16w the supraoccipital surface nor is it concave posterolaterally as in M. ie#ersoni. Also, the process is smaller and more rounded than in that species. The occipital condyles of the Cita Canyon skull have shorter mediolateral diameters than the condyle Cope ( 1893) figured for M. leptostomus from the Blanco Fauna. In the posterior portion of the skull the excavation between the basilar tubercles of the basioccipital is not so deep as in M. fe#ersoni. The projection of bone medial to the gIenoid fossa and surrounding 224 BULLETIN FLORIDA STATE MUSEUM Vol. 12 - the posterior border of the foramen ovale is large and rounded, whereas in M. Je#ersoni it is short and narrow. The basilar tubercles in the Cita Canyon skull are situated more ventrally in relation to the condyles than in M. te#ersoni. The median lacerate foramen is 10- cated farther posteriorly in relation to the basilar tubercles than in that species, in which it opens adjacent to the anterior border of the tubercles, and the foramen ovale and foramen rotundum are farther apart. The pterygoids and palatines project at a more obtuse angle from the basisphenoid and the parasphenoid does not project as far ventrally as in M. ie#ersoni. The epitympanic ring is complete on the left side of the skull and forms a circular loop of bone which is firmly fused with the mastoid process of the squamosal posteriorly, the entotympanic medially, and the squamosal anteriorly. The external auditory meatus thus formed is directed laterally. The entotympanic is fused with the stylohyal process anteriorly and the epitympanic laterally. The eustachian tube apparently occupied a groove running obliquely anteroventrally from the anterolateral margin of the entotympanic toward the midline of the skull. The stylohyal process appears to be abnormal, either it is exostosed or contains a fused fragment of the stylohyal. At the base of the stylohyal process, posterior to the epitympanic, is the opening of the stylomastoid foramen. The foramen immediately posterior to the stylomastoid ( second stylomastoid, Leidy, 1855: 10) opens into a ca- nal that leads into the mastoid process of the squamosal. Posterior to this foramen a groove runs along the posterior side of the mastoid process, fading out just ventral to the supraoccipital crest. In other specimens this groove is roofed over and opens in the mastoid process as the mastoid foramen. Apparently the latter two fbramina are for venous drainage from the cranium and exhibit considerable individual variation. On the right side of the skull the stylohyal process, entotympanic. and ectotympanic are missing, exposing the petrosal quite clearly. The opening of the fenestra ovalis in the dorsolateral portion of the petro- sal is directed anterolaterally The facial foramen is located anterior to the fenestra ovalis. Posteroventral'to· the-fenestra ovalis is the fen- estra rotunda, the opening-of which is directed posteroventrally. Lat- eral to the petrosal, the epitympanic recess opens dorsally into the epitympanic sinus, which extends anteriorly within the squamosal to the posterior margin of the. glenoid fossa. In its anterior region the palate resembles that of M. le#ersoni, but 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 225 more posteriorly the angle of bifurcation between the palatines be- comes greater than in M. ie#ersoni, thus resembling the type specimen of M. leptostomus (Cope, 1893 ). The posterior alveolar border for the caniniform does not display any of the distinct grooves or concavities in the M. ie#ersoni teeth. DENm'ION. - Isolated megalonychid teeth available from Cita Can- yon include four upper caniniforms and one lower canihiform, four Mg's or M3's, one upper or lower Mi, one M2, and two M3's. In ad- dition a left mandible, WT 580, with a fragmentary caniniform, Mi, and M2, and a right mandible with a caniniform, M2, and M,· A fragmentary palate includes the left Mi . M#, right MS - M< and a partial alveoli of Mi - M2 ( Table 5). This collection is of interest because it comprises a series of speci- mens ranging from very young to presumably fully adult. The varia- tions in the teeth are presumed to represent stages in ontogenetic age rather than specific differences. The left mandible, WT 580, is from an Table 5. MEASUREMENTS OF TEETH oF Megalon!/x Zeptoston:115 IN MM Catalog Estimated Measurennent Tooth Number Age Ap MLZ Ci WT 579 juvenile 26.9 18.4 C, WT 2548 juvenile 22.8 12.0 Mi WT 580 very young 7.9 10.6 M~ WT 580 very young 8.0 10.0 M. WT 2821 adult 15.5 24.0 M2 WT 579 juvenile 12.6 18.5 Ms WT 579 juvenile 13.7» 18.4 8 WT 2376 juvenile 12.9 18.0 M, WT 2876 juvenile 18.5 18.0 0 WT 1746 adult 30.1 19.6 0 WT 1746 adult 30.0 19.4 C' WT 1746 juvenile 25.8 15.0 0 WT 2876 juvenile 26.9 16.4 Mi WT 580 juvenile 10.8' 13.5' WT 580 juvenile 11.9 16.23 WT 580 juvenile 11.7 15.8 M* WT 580 juvenile 7.9' 12.9 M' or M' WT 1746 adult 15.5 24.6 MI or M' WT 1746 adult 15.0 25.3 M' or M' WT 2548 juvenile 14.2 19.6 M'·or Ma WT 1789 juvenile 18.0 19.9 M' or Mi WT 2548 juvenile 18.4 17.0 ZAP - anteroposterior 2 ML = mediolateral 8 Estimated 226 BULLETIN FLORIDA STATE MUSEUM Vol. 12 extremely young iddividual, for it is very small, short, and rather rounded anteroposteriorly in comparison with the longer, slimmer adult jaw. The bone of the mandible is rather porous in texture. The distance from the anterior border of Mi to the posterior border of M2 is 17 mm. the teeth are embedded in the jaw in such a way that their taper cannot be measured, nor can the amount of internal dentine be determined. Observed differences between the presumed juvenile and adult teeth are larger size and greater relative width in the latter. Only the lateral wall of the caniniform is preserved. It is long-ovate in shape and measures 16.6 mm in length. The caniniform associated with the adult mandible WT 579 differs from all the other caniniforms from Cita Canyon, as well as from the caniniforms ef later species of Megalonyx, in lacking the prominent median bulge on the linbual side and in being more convex on the lateral side. In this respect it closely resembles the canine of M. cur- videns. M2 and Ma show no significant differences from the other specimens and M, closely resembles one from the Snake Creek beds that Matthew ( 1924) referred to M. curvidens. Two isolated upper caniniforms, a right and a left, WT 1746, repre- sent adult individuals. The extreme development of the median bulge on the lingual side makes both teeth almost triangular in cross section. This bulge is better developed in these specimens than in the other Cita Canyon caniniforms, as well as in any other species of Mega- loni/x. The anteroposterior dimension is shorter than in most specimens of M. wheatle!/i and the mediolateral dimension is greater. The oc- clusal surface is concave, and the median bulge projects above the rest of the occlusal surface. A third upper caniniform, also numbered WT 1746, is smaller than the two just described and represents a younger individual, as the sides are not quite parallel and the internal dentinal layer is not so deep as in the preceding specimen. Otherwise this tooth has the same characters as the adult specimens. Another upper caniniform, WT 2376, appears to be an adult but does not have so large a median bulge as the WT 1746 specimens. The occlusal surface is worn evenly and thus lacks the central excava- tion and projecting lingual bulge characteristic of later Megalon!/x caniniform teeth. The lower caniniform, TNT 2548, is similar to WT 2376 in lacking a prominent median bulge and having an almost Rat occlusal surface. These two teeth do not differ in age from those with 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 227 the prominent median bulge; the differences may represent extreme individual variation or sexual dimorphism. Two molariform teeth, a left and a right M2 or Mi both numbered WT 1746, presumably represent fully adult individuals, while two other right upper M2 or Ms, TNT 2548 and WT 1789 represent younger individuals. The former are considerably longer mediolaterally than in either WT 2548 or WT 1789. The mediolateral widths are corn- parable to those of M. le#ersoni, but the anteroposterior dimensions are considerably less than for that species. The mediolateral widths fall within the upper limits of the observed range of MS and near Table 6. MEASUREMENTS OF Megalonyx leptostomus POSTCRANIAL MATERIAL IN MM Tibia ( WT 1716 ) Length at anteroexternal surface 251 Max. breadth of head 182, Max. breadth of tarsal end 112 Max. anteroposterior diameter of tarsal end 63 Calcan€um ( WT 2599) Width of neck 25.8 Length of neck 44.9 Distance from inner border of external astragalar facet to inferoexternal prominence of articulating end 72.0 Distance from dorsal border of external astragalar facet to ventral border of cuboid facet 78.0 Metatarsal II (WT Uncatalogued ) Max. length 54 Breadth of shaft 22 Depth of proximal end 88. Depth of distal articulation ( obliquely) 871 Metatarsal III (WT 1836) Anteroposterior diameter through middle 89.9 ' Depth of median vertical convexity at distal end 55.0. Depth of proximal articular face 88.5 Max. width 51.5 Phalanx I of Digit III of Pes (WT 1714) Max. anteroposterior length 34.5 Max. width 46.8 Depth of proximal end 55.6 Height of distal condyles 41.7 Phalanx ZI of Digit III of Pes ( WT 1714) Max: anteroposterior length 58.0 Width of proximal end 89.5 Depth of proximal end 51.4 Width of distal contvles at middle 27.4 Depth of distal condyles 86.6 1 Estimated 228 BULLETIN FLORIDA STATE MUSEUM Vol. 12 the upper end of the range for M2 from the Blanco beds. The tooth of M. /eptostomus that Cope ( 1893) thought was a last superior molar is probably an M2 or M3. Its dimensions correspond to two M.g or M3 ( WT 1746) from Cita Canyon. Cope's tooth measures 14 mm anteroposteriorly and 25 mediolaterally, whereas the Cita speci- mens measure 15.0 anteroposteriorly and 25.3 mediolaterally for one, and 15.5 anteroposteriorly and 24.6 mediolaterally for the other. The shapes of these teeth agree with those Cope ( 1893) de- scribed for the Blanco specimens of M. leptostoinus. In cross section they are subangular trapezoids with one side of the occlusal surface bevelled outward. The posterior side is slightly concave, the anterior side slightly convex. TIBIA. - The left tibia ( Fig. 46, H), TNT 1716, is smaller than that of AI. ie#ersoni described by Leidy ( 1855) by about one-fifth, but is larger than that described by Stirton ( 1939) from the Mulholland Fauna in California. The element is essentially complete but lacks the external femoral facet and posterior border of the internal fe- moral facet. In overall outline, the Cita Canyon tibia does not differ significantly from that of iefiersoni figured by Leidy ( 1855 ) -./.. ·r~- .' S'.2.2 , k../ .4--* :i.>»- ° .9.. I. I . I A 8 1 413 FIGURE 3. Left caliancum of Megahmyx leptostomus, WT 2599. A.-inner view, B.-outer view. x 14. 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 229 It differs from the megalonychid from Mulholland in the greater ventral projection of the internal malleolus and the smaller and less conspicuous astragalar spine. The facet for the astragalus is not as wide as in that Specimen; in this respect it resembles M. fe#ersoni, CALCANEUM. - The left calcaneum ( Fig. 3), WT 2599, from Cita Canyon lacks only the extreme posterior border of the tuber calcis and the medial edge of the internal astragalar facet. It is slightly smaller and less robust than the calcaneum from Hagerman, Idaho, USNM 23209. The tuber calcis of M. leptostomus is similar in shape to that of the new Hemphillian genus ( below) in its greater pos- terior expansion. The proximal articular surfaces lack the distinct grooves between the facet seen in M. iegersoni californicus Stock ( 1925), and the euboid facet lacks the inner lip flattened at right angles to the rest of the surface observed in that species. The dis- tance between the internal astragalar facet and the euboid facet is greater in M. leptostomus than in the specimen from Idaho. The area for attachment of the peroneus muscle is not so prominent and does not project laterally as a narrow ridge of bone as in that specimen. METATARSALS. - A poorly preserved left metatarsal II is avail- able from the Cita Canyon collections. It is smaller than the same element in the Hagerman, Idaho sample, but in overall appearance the two specimens agree. A third metatarsal in the Cita Canyon collections lacks only the ventral extremity. It has about half the bulk of metatarsal III of M. ie#ersoni, and differs from that of Hatmlops and M. le#ersoni in having the medial and lateral distal articular facets for phalanx I raised above the surrounding surface and very well defined. How- ever, the carina is less robust than in M. ie#ersoni and the facet for articulation with metatarsal IV projects farther ventrally. The facet for the ectocuneiform is similar in shape to that of M. fe#ersoni, except that it is slightly convex at the center of the facet rather than uniformly concave. PHALANGES. - The Cita Canyon ~ material includes phalanges I and II of the third pedal digit ( Fig. 4). The remarkable feature of these specimens is that they are unfused even though they appear to represent a mature individual Or individuals. The same is ob- served in specimens from the Santa Fe I site in FJorida and from Hagerman, Idaho, both of Blancan age. This is in sharp contrast to all observed later Pleistocene Megalonyx specimens, including 230 BULLETIN FLORIDA STATE MUSEUM VoI. 12 i t, f. 4 ' ~- -~ 3 4 -4 4 ABC . / P r'\ f . C 4 DEF . . ''.M:'t 1 1 fs, vt 4 4.1/,·R Di.1 I % ./ . I G H 1 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 231 those of M. ie#ersoni described by Leidy ( 1855) and Stock ( 1925), in which phalanges I and II are completely co-ossified. Phalanges I and II from Cita Canyon are less robust than those from Idaho, but do not differ significantly in shape. The anterior facets on phalanx I are much more flattened in these early speeies of Megalonyx, than in later Pleistocene specimens. The posterior surface of the phalanx II is similar to that of Hapalops. A facet is located on the dorsal and another on the ven- tral end of the median ridge between the articular surfaces for phalanx I. These facets contact the dorsal and ventral margins of phalanx I and prevent much movement between the two phalanges. Considerably less movement is possible between these phalanges than in Hapalops. Megalonyx let*onyx ( Marsh) Morotherium leptonyx Marsh 1874, Notice of new Tertiary mammals III, Amer. Jour. Sci., Vol. 7, p. 532. TypE. - Marsh ( 1874) described a small ungual phalanx of the third digit of the manus from the Snake River Basin, Idaho, as the type of Morotherium leptonyx. Evidently it represents a small meg- alonychid, probably Megalonyx, but more refined determination is impossible. The only locality data Marsh provided in the original description is "Pliocene beds of Idaho." According to Hay ( 1927) this claw is the same one that Leidy ( 1871) mentioned as having come from Castle Creek, Owyhee County. Gazin ( 1935) tentatively referred two mandibles, some teeth, and a few footbones from the Hagerman beds ( early Blancan) to Megalon!/x leptonuxP ( Marsh) on the basis of their small size and the lack of determinable ma- terial of other ground sloths from the area. The type of Morotherium /eptonyx has been misplaced. As the description of type is inade- quate for specific diagnosis and the age and locality are uncertain, we consider Megalonux lepton!/x a nomen dubium. Other Blancan Megalonyx HAGERMAN, IDAHO. - We doubt Gazin's ( 1935) tentative ref- erence of the Megalonyx sample from Hagerman, Idaho to M. FIGURE 4. Postcranial elements of Megatonyx Zeptostomus. A. Phalanx I of digit III of pes, WT 1714, lateral view, *1/5. B and C, proximal and distal views of same, *36. D. Phalanx II of digit III of pes, lateral view, X M. E and F, proximal and distal views of same, X 1/5. G. Left tibia, WT 1714, at17- terior view, x 0.38. H. Inner view of same, XO.38. 282 BULLETIN FLORIDA STATE MUSEUM Vol. 12 leptonyx ( Marsh). It seems more probably referable to M. lep- tostomus. In any case we defer specific assignment until the dis. crimination of species within Megalonyx is placed on a more secure and consistent basis. Gazin ( 1935) described the following material from the Hager- man beds: imperfect mandible with complete lower dentition, as- sociated with upper caniniform, upper cheek tooth, claw and patella USNM 13477; mandible edentulous except for left caniniform of immature individual, USNM 12669, both from Plesippus quarry; last upper molariform tooth, USNM 13568, co-ossified trapezium and first metacarpal, USNM 13570; and two fourth metacarpals, USNM 12675; a claw and several toe bones, from "various localities south D,of the horse quarry. Two additional lots of associated specimens Elmer Cook and George Pearce collected about 5 miles south of Table 7. MEASUREMENTS OF ASSOCIATED RIGHT HIND LIMB ELEMENTS OF Megalonyx FROM HAGERMAN, IDAHO, ( USNM 23209 ) IN MM Fibuta Estimated length 280.0 Anteroposterior diameter at middle of shaft 88.5 Transverse diameter at middle of shaft 26.5 Anteroposterior diameter of distal end 63.3 Max. transverse diameter of distal end 68.8 Calcaneum Max. length 178.0 Max. diameter of tuber calcis 171.0 Max. thickness posterior border 30.0 Width of neck 28.8 Thickness of neck 48.0 Distance from inner border of external astragalar facet to .inferoextemal prominence of articulating end 71.3 Distance from dorsal border of external astragalar facet to ventral border of cuboid facet 85.4 Astragalus Max. ariteroposterior diameter parallel to fibular facet 92.0 Max. transverse diameter measured at right angles to fbular facet 66.2 Depth of fibular facet at distal end 44.1 Max. transverse diameter of head 49.9 Anteroposterior length of external calcaneal facet 68.2 Cuboid Max. diameter from astragalar to cuneiform surface 48.9 Max. diameter from calcaneal to dorsal surface S2.4 Max. diameter from medial to lateral surface 43.5 N avicular Max. diameter across cuneiform facets ( width) 60.1 Diameter taken at right angles to greatest ( dorsoplantar) diameter 58.0 Max. dickness through convexity articulating with astragalus 27.6 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 288 Cuneiform Max. dorsoplantar diameter 49.9 Max. medi6lateral diameter 81.8 Max. thickness through navicular to mtt III surface 18.2 Metatarsal II Max. length 85.0 Breadth of ·shaft 37.8 Depth of proximal end 89.9 Depth of distal articulation 29.7 M€tatarsal III Anteroposterior diameter through middle 51.8 Depth of median vertical convexity at distal end 66.6 Depth of proximal articular face 58.8 Mak. width 47.6 Metatarsal IV Anteroposterior diameter throu~h middle 57.9 Depth of median vertical convexity at distal end 50.1 Depth of proximal articular face 51.9 Max. width 41.0 Metatarsal V Anteroposterior diameter 51.1 Depth of major convexity at distal end 24.1 Depth of proximal end 25.8 Max. width 21.1 Phalanx I of Digit III of Pes Max. anteroposterior length 58.8 Max. width 49.8 Depth of proximaI end 54.6 Height of distal condyles 45.8 Phalanx II of Digit III of Pes Max. anteroposterior length 45.9 Width of proximal end 45.5 Depth of proximal end 86.7 Width of distal condyles at middle 81.5 Depth of distal condyles 88.3 Phalanx I of Digit IV of Pes Max. anteroposterior length 28.9 Max. width 88.8 Depth of proximal end 86.7 Height of distal condyles 80.2 Phalanx II of Digit IV of Pes Max. anteroposterior length 62.3 Width of proximal end 84.0 Depth of proximal end 87.1 Width of distal condyles at middle 28.2 Depth of distal condyles 29.1 Phalanges I and II of Digit II of Pes Max. anteroposterior length 87.5 Width of pfoximal end 21.0 Depth of proximal end 28.9 Width of distal condyles 17.9 Depth of distal condyles 17.0 284 BULLETIN FLORIDA STATE MUSEUM Vol. 12 Plesippus Quarry, near Hagerman, Idaho, contain: 10 podi,il frag- ments, USNM 23208 and a right fibula, calcaneum, astragalus, cub- oid, navicular, cuneiform and four metatars.ils, three sesainoids and many of the phalanges of the three lateral digits, USNM 2.3209. The unfused condition of phalanges I and II in the third ped:,1 digit has been noted above. Phalanges I and II of the second ped:21 digit are fused. The measurements are presented in Table 7 and the foot is illustrated in figure 5. PROCTER PITS, TEXAS. - A Alegalont/x palate with the right first molariforin, left caniniform, and alveoli for the remaining upper teeth, FAM 77811, was collected iii Proctor Pit C in the Texas Pan- handle. The widest dimension of the palate is 86.6 mm across the .. 4OMM Ab f FIGuRE 5. Left pes of Megaton!/x from Hagerman, Idaho, USNM 28209, dorsal view. 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 235 caniniforms. The palate is constricted posterior to the caniniforms to a width of 37.2 mm. The width across the posterior molariforms is 58.0 mm; between them it is 16.8. The posterior narial opening lies opposite the last molariforms. The caniniform measures 29.8 min long by 16.6 wide and has the lingual convexity characteristic of advanced Megatonyx. MI measures 17.2 by 14.0 mm. The cheek tooth row is 65.5 mm long. In anterior view the rostrum presents an elliptical narial open- ing with a maximum width of about 38 just dorsal to the palate and narrowing to about 20 between the lateral edges of the nasal bones. The tips of the nasals stand 8 behind the anterior edges of the canines. The anterior edge of the jugal lies immediately pos- terior to the caniniform. The jugal plate slopes ventrad, posteriad, and slightly laterad. The posterior edge lies posterolateral to M2, SANTE FE I, FLORIDA. - A medial phalanx of the third pedal digit, UF 10441, was collected by D. Webb, R. Allen, and J. Rob- ertson in June, 1965 from locality I, Sante Fe River, Gilchrist County, Florida. Its length is 64.9 mm, its width 32.1, and its depth at the proximal end 46.5. The significance of this specimen lies in the fact that it is unfused to the proximal phalanx as in other Blan- can Megalon!/x. JACKASS BUT[E, IDAHO. - An isolated intermediate lower molari- form tooth, UO 2404, from locality 6, Grandview Fauna, Jackass Butte, Idaho measures 17.9 by 12.4 mm at the occlusal surface, Megalonyx curvidens Matthew Megaton!/x curvidens Matthew, W. D. 1924. Third Contribution to the Snake Creek Fauna. American Mus. Nat. Hist. Bull., 50; 149. TypE. - Tooth, probably RM3, AM 17601, from upper level, Quarry 1. Upper Snake Creek Beds, Sioux Co., Nebraska, collected by Albert Thomson in 1918. REFERRED MATERIAL. - Left ramus and symphysis with canini- forms and left cheek teeth, FAM 77800, collected by M. F. Skinner in East Pliohippus Draw about 10 feet above the site from which the type of Pliohippus leidyanus came along with rhinoceros, mas- todon, horse, camel, and peccary material ( Skinner field notes, vol. 2, p. 44); two isolated upper cheek teeth, AM 20493 and 21460; humerus, FAM 77801, from East Pliohippus Draw; navicular, PU 12079 ( Sinclair, 1915)1 two proximal phalanges of the third manual digit, AM 81034 and 85970, the latter collected by the late Harold 236 BULLETIN FLORIDA STATE MUSEUM Vol. 12 Cook 6 September 1949 while walking with Glen Evans and Gray- son Meade, in high exposures about 100 yards east of the Pliohip- pus leid!/(inus site ( Cook's notes ); second phalanx of third manual digit FAM 77802; and an ungual phalanx, AM 14051. DIAGNOSIS. - Caniniforms long ovate with concave lingual sides, symphysis protruding relatively farther and more rounded at an- terior tip than in AI. leptostomus. Alveolar ridge posterior to the caniniform teeth same width as adjacent cheek teeth. Molariforin teeth slightly more curved than in later species. DESCRIPTION. - The niandible, FAM 77800. (Fig. 6) belongs to a young animal as indicated by the tapering teeth. This is most evident in the caniniforms, the occlusal dimensions being 14.8 by 8.0 in contrast to basal dimensions of 21.5 by 11.1. 20mm r 4 k AM 77890 451*79428¥1(49 FIGURE 6. Mandible of Megato,1!/x curvidens, FAM 77800, occlusal and lab eral views. 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 287 The caniniform teeth present a long subcircular occlusal surface that wears nearly flat, much as in typical Megalonyx. The canini- forms differ in having faintly concave lingual sides, whereas canini- forms of M. ie#ersoni show a pronounced lingual bulge. The absence of this bulge also has been noted above in some M. leptostomus specimens. The symphysis protrudes somewhat more than in late Pleistocene Megalon!/x specimens, reaching 14.3 anterior to a line between the anterior ends- of the caniniforms. But as the canines erupt and reach mature proportions the protrusion presumably decreases ac- cordingly. The keel on the anterior slope of the symphysis is weak and confined to the upper third of the slope. The anterior border of the symphysis rises in a nearly straight profile, in contrast to the sigmoid curve, concave at the base, convex at the top, typical of Iater species of Megalonyx. The mental foramen appears 10 mm posterior to the "chin" and 9 mm below the alveolar border, about as in typical Megalonyx. The symphysis attains a depth of about 42 mm below the caniniform teeth. The posterior end of the sym- physis lies between the roots of the caniniforms as in other Mega- Zonyx. The alveolar ridge posterior to the caniniform retains the same width as the adjacent cheek tooth. In this respect M. curvidelis differs frorn typical late Pleistocene Megalon!/x specimens and agrees with Meade's ( 1945) description of a mandible of M. tel*os- tomus from Mt. Blanco, Texas. The cheek teeth closely approximate those of later Megalonyx species in shape and the development of wear facets. The principal wear is on the posterior edge of each tooth. The last molar in the mandible agrees closely with Matthew's type tooth, except that the type is slightly larger and comes from the right side. Measurements of the base of left M3 are 16.9 by 13.4 mm as compared with 17.8 and 15.0 in the type. These teeth are slightly narrower on the labial than on the lingual side. A concavity passes down the posterior side and a faint one down the labial side. The highest corners are the posterolabial and antero- lingual. Each tooth curves gently along its length producing a slight arch posteriorly and a lesser one lingually. This feature of the type tooth inspired the species name and distinguishes it from late Pleistocene Megalonyx. The upper half of the ascending ramus is missing. A large fora- 288 BULLETIN FLORIDA STATE MUSEUM Vol. 12 men through which the mandibular nerve and blood vessels passed, occupies the base of the coronoid process, but somewhat higher than in most Megalonyx specimens. An upper molariform tooth, AM 21460, measures 14.0 mm an- teroposteriorly and 19.8 transversely. It is slightly concave on the posterior and convex on the anterior side. Another such tooth, AM 20493, measures 13.7 by 19.0 mm. The left humerus from the Snake Creek channels is an incom- plete distal half of the shaft. The deltoid and pectoral crests are well developed, being especially well defined at their junction near the midpoint of the shaft. Between this point and the distal condy- lar expansion the narrowest width of the shaft is 44.2. Sinclair ( 1915) described a navicular from the Upper Snake Creek channels, P.U. 12079. Measurements taken from the illus- tration of the Snake Creek specimen show its natural size to be about 52 mm across the cuneiform facets and 46 mm in the dorso- palmar diameter. The Upper Snake Creek navicular is similar to that of Plioinetanastes in that the mediolateral diameter is greater than the dorsopalmar diameter' and unlike Megalonyx ie#ersoni, M sierrensis, and the Idaho specimen where the opposite occurs. The Snake Creek navicular differs from Pliom€tanastes but is sim- ilar to the navicular of Megalon!/x in having the mesocuneiform facet elevated above the ectocuneiform facet. Table 8. MEASUREMENTS OF PHALANGES OF Megaloriyx curuidells IN MM Proximal Phalanges (AM 81034) (85970) Prox. Length 58.3 58.1 Prox. Width 42.9 46.8 Max. Depth 81.8 84.2 Dist. Length 48.1 43.5 Dist, Width 82.1 80.2 Second Phalanx ( FAM 77802) Max. Length 62.2 Median Length 55.3 Prox. Width ca. 89 Dist. Width 82.4 The phalanges from the Upper Snake Creek channels closely correspond to specimens of later Megaton!/x, except for their gen- erally smaller size. Their principal dimensions are presented in Table 8. 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 289 Megatonyx mathisi new species TypE. - Ventral half of skull, UCMP 80416 ( Figs. 7 and 8). TYPE HORIZON AND LOCALITY. - Black Rascal Creek, V-67223, Upper Mehrten Formation, east of Merced, Merced County, Cali- fornia. Hemphillian, probably late Hemphillian, tentatively corre- lated with the Pinole Tuff site, Contra Costa County, California, which has a potassium-argon date of approximately 5.2 million years. ETYMOLOGY.- Patrollymic for Glen E. Mathis whose generous assiltance made collecting the type specimen possible. DIAGNOSIS . - A relatively small animal, slightly smaller than Mega- 20mm -li~bi~45». V.1,779*&1-5«**0 1-6 Plig f -0/ --lbaL'.*I'."- FIGURE 7. Skull of Megatonyx mathist, UC 80416, lateral and palatal views. 240 BULLETIN FLORIDA STATE MUSEUM Vol. 12 lonyx Zeptostomus but larger than Pliometanastes protistus. Caniniform smaller than first molariform, oval in cross section with great- est diameter obliquely anteroexternal to posterointernal, unlike all other species of Megalonyx in which the greatest diameter is ob- liquely anterointernal to posterointernal. Caniniform without me- dian bulge, lingual side of tooth relatively flatter than labial side. Maxillary plate between caniniform teeth wide and very thin. Palate distinctly arched, concave posterior to caniniform teeth, be- fr SS fO est. saf ent \ 1 fl 1 cf ~ Pif - . C stp )\\ mps - /- mf, FIGURE 8. Basicranial region of skull of Megalon~x mathisi, UC 80416. Ab- breviations: cf, condyloid foramen; ent, entotympanic bone; er. depression for epitympanic ring; est, groove for eustachian tube; f, facial foramen; fo, foramen ovale; fr, foramen rotundum; mf, mastoid foramen, mlf, median lacerate foramen, mps, mastoid process of squamosal; 0, fenestra ovalis; p, petrosal; plf, posterior lacerate foramen; 0, fenestra ovalis; B petrosal, plf, posterior lacerate foramen; r fenestra rotunda; saf, canal for inferior ramus of stapedial artery; smf, stylomastoid foramen; stp, stylohyal pit; th, tympanohyal; x, unnamed foramen posterior to stylomastoid foramen. 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 241 coming convex at level of M2. Palate with long narrow canals for nutritive foramina. Nasal region not as highly vaulted as in other species. Sagittal crest present. and supraoccipital crest ~ prominent. Rectus capitus ventralis excavations posteromedial to basilar tuber- cles. Occipital condyles relatively closer together and do not pro- ject so far posteriorly as in M. let*ostomus. Dorsal margin of the condyles nearly confluent with the occiput. SKULL. - The skull, UCMP 80416, lacks the nasals, dorsal part of the frontals and parietals, expanded portion of the jugal, ventral part of the pterygoids, ectotympanic, and all teeth. All sutures are firmly fused indicating a fully adult individual. The skull is slightly smaller than Megalonyx leptostomus but larger than Pliometanastes protistus The most striking differences between M. mathisi and the other Table 9. MEASUREMENTS OF MegaZonyx Mathisi SKULL, ( UCMP 80416) IN MA Max, length 286.0 Length posterior end occipital condyles to posterior border caniniform alveoli 268.0 Width maxillary at level of caniniforms 92.7 Width palate anterior to first m61arifoms 45.8 Min. width of palate between molariforms 21.8 Width across zygomatic arches anterior end glenoid fossa 152.0 Max. width acr6ss mastoid processes 122.8 Min. postzygomatic width across squamosal 104.0 Transverse diameter inside foramen magnum 23.6 Dorsoventral diameter foramen magnum 28.2 Max. width across occipital condyles 66.0 Max. width across one 66ndyle 22.8 Max. depth of one condyle 28.0 Height occiput from dofsal margin foramen magnum 51.01 Height Inaxillary, anterior alveolar border to nasal suture 50.0 Width between premaxillary sutures 56.5 alveoll' left right Max. length caniniform alveolus 18.5 18.5 Max. width caniniform alveolus - 12.0 11.8 Max. anteroposterior alveolar diameter Ml 16.5 16.0 Max. mediolateral aIveolar diameter Mi 21.0 21.5 Max. anteroposterior alveolar diameter M2 17.0 17.5 Max. mediolatefal alveolar diameter M' 24.0 24.5 Max. anteroposterior alveolar diameter Ms 17.0 17.5 Max. mediolateral alveolar diameter M» 24.0 28.0 Max. transverse aIveolar diameter M, 16.5 17.0 i Estimated ? Measuremints of molariforms taken just ,below level of palate where alveolar walls are straight. These measurements do not take into account distortion due to preservation and reconstruction and thus may be. as much as 1 mm more than iruc alveolar diameter. 242 BULLETIN FLORIDA STATE MUSEUM , Vol. 12 species of Megalonyx are found in the anterior region of the skull, The nasal bones are lacking, but from the curvature of the maxil- laries apparently the nasal region was not so highly vaulted as in the other species. The maxillary between the caniniform teeth is a thin broad plate in M. mathisi, whereas in M. leptostomus it is consid- erably thicker, attaining its greatest thickness in M, leffersoni. This difference in thickness of the maxillary is greater than would be ex- pected if it were due solely to allometric increase in skull size. Medial to the caninform alveoli, on the antero-internal side, is a pair of depressions for the premaxillary. The distances between these articular surfaces is relatively farther apart in M. mathisi than in other species. The size, shape, and orientation of the caniniform teeth distin- guishes M. mathisi from all later species of the genus. The canini- form alveoli are smaller than the alveoli of the Mi and are distinctly oval, the internal margin being straighter and less convex than the external t6 posterointernal, in marked contrast to other species where the greatest diameter is anterointernal to posteroexternal. The posterior margin of the tooth projects medially in M. mathisi, whereas in other species it projects laterally. The size and cross sec- tion of the caniniform aIveolus are structurally intermediate between the small triangular tooth of Pliometanastes and the typical long ovate tooth of Megalonyx Zeptostomus. The molariform alveoli are subtriangular and do not differ in shape from the later species of the genus. The palate is distinctly arched, concave immediately posterior to the caniniform teeth, becoming convex at the level of the M2. M. mathisi resembles M. ie#ersoni in this character and differs from M. leptostomus in which the palate is straight along its length. The palate contains long narrow canals for nutritive foramina. These run the length of the palate as in Pliometanastes and they are especiallv prominent in the anterior part of the palate, just posterior to the caniniform teeth. There is some distortion and displacement in the dorsal part of the occiput, but not enough to obscure the major characters. The supraoccipital crest is strongly developed, as is the vertical occipital crest. The occiput rises nearly perpendicular to the basi-cranial axis. The occipital condyles do not project so far from the occiput as in other species and are nearly confluent with the occiput along 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 243 their dorsal margin. The occiput crest projects outward posteriorly above the condyles. Compared with Megalon!/x leptostomus: The mastoid process is not quite so prominent and is somewhat pointed; the area of attach- ment of the digastric muscles dorsal to the mastoid process is more rugose and more deeply excavated; the occipital condyles are closer together and the foramen magnum more rounded; the bone of the dorsal margin of the foramen magnum is considerably thicker and has a prominent pair of processes projecting posteriorly on either side of the midline and articulating with the atlas; the occipital con- dyles are more elongate medio-laterally; and the triangular notch so prominent in M. leptostomus is barely evident in M. mathisi. The basicranial region is remarkably well preserved; only the ectotympanic ring is lacking. In ventral view the occipital condyles are positioned closer to the skull and are relatively closer together in M. mathisi than in M. Zeptostomus. The condyloid foramen opens anterior to the condyles and postermedially to the posterior lacerate foramen, the largest of the basicranial foramina. The stylohal process borders the lateral edge of the posterior lacerate foramen and ap- pears to be formed by the paroccipital process, tympanohyal, and a process of the mastoid region, which together form an articulating surface for the stylohyal. Anterolateral to the tympanohyal and post- erlateral to the epitympanic recess is an oval depression for the epi- tympanic ring. The entotympanic forms the anterior margin of the posterior lacerate foramen, sutured to the paroccipital process at the anterior margin of the stylohyal process on the lateral side and the basioccipital on the medial side, lying against the lateral margin of the basilar tubercles. It is widest posteriorly, narrowing anteriorly where it meets the pterygoid and lateral margin of the median lacer- ate foramen. Dorsal to the entotympanic is the petrosal which lies in the epitympanic recess. The fenestra ovalis opens from the dorsal portion of the petrosal, with the opening directed laterally. Postero- ventral to the fenestra ovalis is the fenestra rotunda which opens posterolaterally. Some confusion exists in the literature concerning the designa- tion of the stylomastoid-mastoid foramina. The skull of Megalonyx mgthisi has two foramina at the lateral base of the stylohyal process and a dorsal foramen in the mastoid process on the posterior side. The anterior foramen lies immediately posterior to the oval depres- sion for the epitympanic ring, lateral to the tympanohyal, and at the 244 BULLETIN FLORIDA STATE MUSEUM Vol. 12 base of the stylohyal process. This foramen, the stylomastoid fora- men, communicates via the sulcus facialis with the facial foramen, which is located antero-dorsal to the fenestra ovalis. The canal of the foramen immediately posterior to the stylomastoid bends posterodor- sally into the mastoid region. This latter foramen and the mastoid foramen on the occiput probably function as openings for venous drainage leaving the skull. The latter foramina are highly variable, as is demonstrated by the skull of M. mathisi in which the mastoid foramen is lacking on the right side of the skull. This variability is further indicative of venous circulation. In the anterior part of the ear region, the eustachian tube lies in a very shallow and poorly defined groove that begins at anterolateral margin of the entotympanic and continues forward obliquely across the pterygoid towards the midline of the skull. Lateral to this groove are two structures of uncertain function. Anteroventral to the epi- tympanic recess is a dorsoventrally aligned furrow that may have served to house the anterior portion of the epitympanic ring. Medial to this furrow a canal runs within the squamosal but close to the pterygoid from the anterodorsal base of the petrosal to a foramen opening posterior to the foramen ovale. This prominent foramen may represent an enclosure of the inferior ramus of the stapedial artery. The foramen ovale, foramen rotundum, and orbital fissure are in the same position as in other species. The basilar tubercles are excavated internally along the posterior and medial sides. Anterior to the basilar tubercles is a pair of deep sinuses in the basisphenoid, ventral to the anterior portion of the brain cavity. The zygomatic process of the squamosal narrows anterior to the glenoid fossa and is mediolaterally compressed. The ~lenoid fossa is more concave and its anterior side lies more dorsal than in M. leptostomus. No sutural connection is apparent between tha zygo- matic process of the squamosal and the jugal as in M. leidyi. The articular surface for the mandibular condyle .in: M. mathisi appears to be restricted to the lateral margin of the glenoid fossa where the zygomatic process flares laterally, whereas in M. teptostomus the articular surface appears to have extended medially almost to the lateral margin of the pterygoid. Other Hemphillian Megalont/x In addition to the above described material, a number of speci- mens have been discovered that do not warrant specific taxonomic 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 245 assignments. They are arranged below according to geographic origin. BEAR TOOTH SLIDE, NEBRASKA - A lower caniniform, FAM 77803, M. F. Skinner collected in the same channel at Bear Tooth Slide, middle fork of Deep Creek, in the middle of the Ash Hollow Forma- tion, Brown Co., Nebraska, has a long oval cross section with a flat, slightly convex, lingual border. This border is not concave as in M. curoidens, but it is far less convex than typical late Pleistocene Megalonyx. This tooth measures 22.0 x 13.8 mm. COGSWELL QUARRY, KANSAS - FAM 77805 represents an imma- ture stage of development in an upper caniniform of Megalon!/r. The upper surface is worn flat and measures only 14.1 by 9.3 mm. It expands rapidly toward the root, but is broken off 25 mm above the crown. At that point its dimensions are 15.9 by 10.4 mm. In cross section its shape approaches the long ovate form characteristic of Megalonyx. A strong lingual convexity exactly resembles that ob- served in caniniforms of typical Pleistocene Megalonyx. OPTIMA, OKLAHOMA - Savage ( 1941) described an upper and a lower caniniform tooth from the Optima Fauna. The lower, OMP 41-26-Sl, has an occlusal length of 24.8 mm. This specimen is from a mature individual as the tooth walls are parallel. A faint bulge on the lingual side suggests the more massive bulge of later species of Megalonyx. The upper caniniform, OMP 41-BE-S2 is 19 mm long and also from a mature sloth. Its strong lingual bulge is postero- lingual rather than mediolingual as in most later Megalonyx speci- mens. PINOLE, CALIFORNIA - Stirton ( 1939 ) described an upper or lower first molariform tooth, UCMP 22110. Its transverse diameter is 17.0 mm and its length 12.3. It agrees closely in size and wear pattern with a tooth of M. leptostomus from Cita Canyon, Texas ( WT 2548) described above. WESTEND BLOWOUT, OREGON - Two astragali from Hemphillian sites in Oregon ( Shotwell, 1958) pertain to Megalonyx. UO 6332 is an incomplete specimen but retains enough of the tibial and navicu- lar surfaces to reveal features characteristic of Megalonyx, most conspicuously the deep tibial notch. KREBS RANCH, OREGON - UO 8080, the second Oregon astraga- lus, is complete. It differs from astragali of Pliometanastes and re- sembles those of Megatonyx in its more nearly equal medial and 246 BULLETIN FLORIDA STATE MUSEUM Vol. 12 lateral sides, less rectangular and more convex internal calcaneal facet, and deeper navicular facet. It also differs from P. protistus in its greater size ( anteroposterior diameter 75.7 mm, transverse diameter 64.6, transverse diameter of head 44.1, length of external calcaneal facet 61.6). Pliometanastes new genus GENOTYPE: - Pliometanastes protistus. As only a single species is certainly attributed to this genus, the characterization of the genus is identical to that of the species below. ETYMOLOGY: - Greek; Plio = pliocene, meta = change, nastes = occupants; thus referring to a Pliocene wanderer. Pliometanastes protistus new species TypE: - UF 9479, a partial skull, including the posterior two- thirds of the cranium ( from the postorbital constriction posteriorly) as well as the left lacrimal, jugal, lateral portion of the maxillary containing the alveoli of Ml - MB, the right maxillary containing the alveoli of Mi - M2, the right portion of the palate containing the alveoli of M2 - MB, ETYMOLOGY: - Greek protistos - the frst. REFERRED MATERIAL: - Anterior skull fragments, UF 10337, 10338, and UF 9561; posterior skull fragment, UF 13211, anterior portion of a right mandible, UF 9450; posterior portion of a left mandible, UF 9480; left superior second of third molariform, UF 9449; left superior fourth molariform UF 9613, and UF 10341; tho- racic vertebrae, UF 9464 and UF 9468; lumbar vertebrae, UF 9465, UF 9466 and UF 9467; caudal vertebrae, UF 10342; fragmentary left humerus, UF 9445; proximal half of ulna, UF 9447; right meta- carpal II, UF 9453; left metacarpal II,UF 9454; associated right metacarpal II, UF 11524; right metacarpal III, UF 11525; right metacarpal IV, UF 11526, and right metacarpal V, UF 11527; patella, UF 9774; right fbula, UF 9446; right navicular, UF 9474; right cal- caneum, UF 9437; left astragalus, UF 9440; phalanx I of digit III of the pes, UF 9459; ungual phalanges, UF 9640, UF 9461, UF 9462, and UF 10340. TYPE HORIZON AND LOCALITY: - MeGehee Farm, about 3 miles north of Newberry, Alachua County, Florida. AIachua Formatioii, Hemphillian ( Middle PIiocene). 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 247 For many years diverse nonmarine sediments. occurring as sink- hole, cave, and fissure fillings, and as isolated accumulations in vari- ous depressions in the Ocala Limestone, have been assigned to the Alachua Formation ( Simpson, 1930; Cooke, 1945; Pirkle, 1956; and Webb, 1964). The lithologies of the sediments assigned to the Ala- chua Formation are quite heterogeneous, ranging from sinkhole fillings of reddish sandy clay, blue-gray in unweathered condition, to gray phosphatic sands in larger basins with commercially signifi- cant zones of hard-rock phosphate. The type locality of the Alachua fauna and Alachua Formation is Mixon's Bone Bed northeast of Williston, Levy County. This locality has yielded a vertebrate fauna of Hemphillian age. Vertebrate fossils found in other Alachua sediments range in age from early Miocene through Recent. Typically these fossil collections are not mixed, but occur as discrete faunas in historically distinct sinkhole fillings and solution deposits. The vertebrate-bearing deposits at McCehee lie unconformably upon an irregular surface of Ocala Limestone of Late Eocene age. Two main strata can be distinguished: a lower clayey sand lying directly upon the limestone, and above this a coarse, poorly sorted and weakly stratified gravelly sand. The greater part of the sloth ma- terial has come from the lower clayey sand layer, as has most of the associated vertebrate fauna. The upper gravelly layer contains shark teeth, ray crusher plates, gar and drum scales, turtle shells, large amounts of pebble phosphate, chunks of Ocala Limestone and black- ened, waterworn bones and teeth. Much postdepositional solution of the limestone has occurred, and in many places the deposits have dropped 10 feet or more into large solution pipes. The McCehee Farm deposits accumulated in an estuary in a coastal solution valley when the sea stood approximately 90-100 feet above its present level ( Webb, 1964). Vertebrate hard parts accumu- lated in solution pockets in the estuary. Of the 14 mammals from MeGehee Farm ( Table 10 and Fig. 9) that have been identmed to genus , 12 (.Mylagaulus, Osteoborus, Ser- ridentinus, Tapiraous, Teleoceras, Aphelops, Calippus, Nannippus, Neophipparion, Pliohippus or Dinohippus, Prosthennops, and Mega- tylopus) are known to occur in both the Clarendonian and Hemp- hillian. Tanupolama is known from Hemphillian faunas but is more typical of the Blancan and later. The species of Neohipparion from MeGehee ( near N. eurystyle ) and the species of Nannippus ( near 248 BULLETIN F.LORIDA STATE MUSEUM Vol. 12 Table 10. MCGEHEE FARM FAUNAL LIST CHONDRICHTHYEY Odontaspis cuspidata C Agassiz) 0. macrota ( Agassiz) Ist,rus hastalis C Agassiz) Carcharodon megalodon Agassiz Hemipristis serra Agassiz Carcharhinus cf. leucas C Muller and Henle) C. limbatus group Gateocerdo cut)ieri ( Peron and Le Sueur) G. aduncus Agassiz Rhizoprionodon sp Negaprion brevirostris ( Poey) Pristidae Dasyatidae Myliobatidae OSTEICHTHYES Lepisosteus Elopidaei Ariidaei Sciaenidae' AMPHIBIA Siren sp. REPTILIA Macroclemmys sp Trionyx aff: T. ferox ( Schneider) Chrysemys williamsi Weaver and Rosei C . carri Weaver and Rosei Geochelone alleni Auffenberg Geochelone aff. G. hal/i ( Sellards) Gopherus sp. Gatialosuchus americanus Alligator cf. A. mississippiensis Ophisaurus sp Farancia sp Crotalidae AVES P.halacrocorax wetmorei Brodkorb Ngcticorax fidens Brodkorb Ereunetes ragi Brodkorb MAMMALIA Talpidae Mylagaulus kinseyi Webbi Lagomorpha Ptiometanastes protistus n. gen. et sp.1 Osteoborus spp. Mustelidae' Felidaei Serridentinus ftoridanus ( Leidy) 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 249 Tapimuus sp Teleocems proterus (Leidy) Aphetops Zongipes C Leidy) Calippits sp, Dinohippus or Pliohippus Nannippus nr. N. ingenuus ( Leidy)' Neohipparion nr, N. euryst!/le (Cope V Hipparion pticatite Leidyi Prosthennops sp Protoceratidae Tanupolama sp Megat!/topus sp Antilocapridae 1 Additions lo preliminary list in Webb (1964). Megalonychidae Tanupoloma BLANCAN 1 EMPHILUAN Osteoborus ~ Prosthennops CLARENDONIAN Serridentinus Neohipparion Aphelops Protoceratidae I Pliohippus Topiravus Teleoceras Mylogoulus Calippus Nannippus Megotylopus FIGURE 9. Stratigraphic ranges of mammalian genera from McGehee Farm. N. ingenuus) are both characteristically Hemphillian species. No im- migrants from Asia such as Indarctos and Plesiogulo, valuable cri- teria for defining many Hemphillian faunas, have been found at McCehee. No sloths, here represented by Pliometanastes, have yet been reported from pre-Hemphillian localities in North America. Protoceratidae previously were not known in post-Clarendonian deposits. Taking all of these facts into consideration, the McCe}lee assemblage is assigned an early Hemphillian age. DIAGNOSIS : - Dental formula 5/4. A relatively small animal, the next larger North American megalonychid being the Hemphillian 250 BULLETIN FLORIDA STATE MUSEUM Vol. 12 Megalon!/x mathisi (new species). Mandible with long, deep, and relatively wide predental spout; posterior end of symphysis below caniniform tooth; caniniform nearly vertical in orientation, nearly circular in cross section, but Hattened lingually. Dorsal ridges not united to form a sagittal crest and diverging just anterior to the ex- ternal occipital protuberance. A pair of depressions present on either side of the median ridge on the occiput at the external occip,ital pro- tuberance. M. rectus capitus ventralis excavations lateral rather than posteromedial to the basilar tubercles. Dorsal border of occipital condyles nearly confluent with the occiput. Mastoid foramen at dor- sal margin of mastoid region. Palate with long, narrow canals for nutritive foramina. Shaft of ulna straight on anterior side. Meta- carpal II with prominent dorsoexternal process 6f proximal end. Patella proportionally longer than in Hapa/.ops or Megalon!/x Fibu- lar shaft not curved dorsomedially to the distal end in the region of attachment of the peroneus brevis muscle as in Megalon!/x Tuber calcis and its neck intermediate in development between Megalon!/x and Hapalops. Astragalus lacking V-shaped tibial notch . Metatarsals slender with restricted muscle scars. SKULL. - The skull of Pliometanastes is known from a number of fragments. The most complete specimen, UF 9479, consists of the posterior two-thirds of a cranium ( from the postorbital constriction back), as well as the left lacrimal, jugal, and lateral portion of the maxillary containing the alveoli of Mi - M3, the right maxillary con- taining the Ml - M2, the right portion of the palate containing the alveoli of M2 - Ma, and a fragment that may represent the right dor- sal projection of the jugal. No teeth are associated with this skull. Isolated skull fragments include four specimens UF 9561, 10337, 10338, and 13211, two of which appear to belong to the same indi- vidual. These specimens contain the lacrimal, the jugal, and the an- terior portion of the maxillary with the alveoli of Mi - M2, and the base of the caniniform alveolus. The other skulls, UF 9479 and 13211 appear to represent adults as all but the squamosal-supraoccipital- exoccipital sutures are obliterated. The skull of Pliometanastes is about half the size of the skull of Megalonyx ie#ersoni and is considerably smaller than that of M. leptostomus ( Figs. 10 through 13). A prominent ridge running from the dorsal margin of the fora- men magnum to the exterrial occipital protuberance has a depression on each side just ventral to the supraoccipital crest similar to the 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 251 1 20 mm 4, 1,41 l 4 I 1-3 FIGURE 10. Skull of Plionietanastes protistits, UF 9479, dorsal view. condition in Eitcholoeops fronto C Scott, 1903-1904, pl. 55, Bg. lb) . Megatonyx lacks these depressions , In P| ic)}netanastes the dorsal oc - cipital depressions are bordered laterally by a pair of semicircular tuberositiess that merge dorsally with the supraoccipital crest. Lateral to these tuberosities and medial to the suture of the exoccipital bone is a smaller and more pointed pair, each bearing a ventrally projecting ridge of bone. A deep groove lies on the suture between the supraoccipital and squamosal bones. Zoz BULLETIN FLORIDA STATE MUSEUM Vol. 12 It 20 mm f, £.f r 1 FIGURE 11. Skull of Pliometanastes protistus, U F 9479, ventral view. 1968 HIRSCHFELD/WEBB: PLIO-PLEwTOCENE SLOTHS 258 alf ~ alc ~ fr rn If cf / Pif fJ-3' FIGURE 12. Basicranial region of skull of Ptiometanastes protistus, UF 9479. Abbreviations as in figure 8, and alc, anterior lacerate canal; alf, anterior lacerate foramen; s, sinus of basisphenoid bone. Considerable variation in this region of the occiput has been ob- served in three specimens of M. le#ersoni: A cast of Owen's speci- men preserved in the Florida collections, Dickeson's specimen and a juvenile from Big Bone Cave, Tennessee, both in the collections at the ANSP. The Owen skull has a depression in the region of the squamosal-Supraoccipital suture, the squamosal bone being depressed below the level of the supraoccipital bone; just medial to the suture 254 BULLETIN FLORIDA STATE MUSEUM Vol. 12 ·- 4 20 mm FIGURE 18. Skull of Ptiometanastes protistus, UF 9479, occipital view. is a relatively f[at space, concave at its dorsal extremity, about 15 wide; medial to this a large tuberosity about 13 wide projects ven- trally from the transverse area of muscle attachment on the occiput. An entirely different situation exists in the Dickeson skull: Medial to the squamosal-supraoccipital suture, which is deeply indented, a projection of bone about 18 wide projects laterally over the suture at its dorsal end. Medial to this projection is an excavated area with the rest of the occiput rather smooth. The occiput of the juvenile from Bjg Bone Cave is quite similar to that of M. leptostomus from Cita Canyon, Texas. In the region of the suture the bony protuber- ances, prominent in the Owen and Dickeson specimens, are lacking. In overall shape the occipital region in Pliometanastes most closely· resembles that in M. leptostomus. The mastoid process is not hooklike as in M. ie#ersoni, but is rounded as in M. leptostomus. The occipital condyles are nearly confluent with the back of the skull as in Eucholeops fronto, whereas in M. ie#ersoni and M. lep- tostomus a depression separates the condyles from the general pos- terior surface of the occiput. The condyles are set farther dorsally from the basilar tubercles than in M. iefiersoni. In Pliometanastes the excavation of the basioccipital, posterior to the basilar tubercles, is not so deep as in Megalonux. The most 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 255 Table 11. MEASUREMENTS oF PNometanastes Protistus SKULL, ( UF 9479) IN MM Height of occiput from dorsal margin of foramen magnum 86.0 Height of occiput from basioccipital 67.8 'Max. width of occiput across mastoid process 108.0 Max, width across occipital condyles 60.51 Max. width of individual condyle 16.0 Max. depth of individual condyle 27.0 Transverse diameter inside foramen magnum 26.8 Dorsoventral diameter of foramen magnum 28.7 Width across the zygomatic arches at the anterior end of the glenoid fossa 182.0 Min. orbitotemporal width 71.01 Min. postzygomatic width across squamosal 94.0 Width ' of basisphenoid anterior to basilar tubercles 45.0 1 Estimate6. prominent basicranial muscle scar, for the rectus capitis muscle, is on the lateral side, whereas in Megalonyx it is on the medial side of the basilar tubercle. The area between the tubercles is nearly flat in Pliometanastes as in E. fronto. whereas it is slightly concave in M. leptostomus, and deeply concave in M. le#ersoni. The stylomastoid foramen lies at the base of the stylohyal proc- ess on its anterolateral side, as in all the other megalonychids ob- served. The position of the mastoid foramen varies considerably between individual specimens of M. ie#ersoni, as well as between that species and the unique skulls of M. leptostomus and Pliometa- nastes. In the most complete skull from MeGehee, the mastoid canal enters the squamosal bone starting at the base of the style- hyal process on its posterolateral side, ascending through the mas- toid process, passing near the posterior surface for about 16, then turning anteriorly and entering the squamosal bone at its dorsal extremity. The location of the mastoid foramen at the dorsal border of the squamosal is similar to that in E. fronto, but the canal lead- ing to the foramen is not enclosed in bone as in Pliometanastes. In Pliometanastes the stylohyal pit is widest at the posterior margin, whereas in M. fe#ersoni it is widest anteriorly. The stylo- hyal pit in Pliometanastes is not so deep as in M. le#ersoni. Plio- metanastes lacks the bony protuberances surrounding the pos- terolateral side of the foramen ovale; these are very prominent and rounded in the skull of M. leptostomus from Cita Canyon and are long, narrow, and less prominent in M. ie#ersoni. The median lacerate foramen lies farther anteriorly to the basi- 256 BULLETIN FLORIDA STATE MUSEUM V61. 12 lar tubercles in Ptiometanastes than in M. leptostomus and pattie- ularly M. ie#ersoni, but not so far forward as in E. fronto. The foramen ovale opens in the same position in relation to the glenoid fossa in Pliometanastes as it does in both Megalonyx and Eucho- loeops. The MeGehee skull is broken in the area of the opening of the foramen rotundum, but the canal leading to it is preserved. Although it i5 n6t possible to say positively where the foramen is located, it appears from the position of the canal that it lay just anterior to the glenoid fossa, ahead of its positii)n in Megalonyx and more nearly in the anterior position observed in Hapalops metimeyeri. Pliometanastes lacks the prominent sagittal crest seen in Mega- Zonyx, Eucholoeops, and Megalongchotherium. Instead a pair of low ridges extend anteriorly from the supraoccipital crest a dis- tance of 24 and then diverge. It appears that in seme juvenile specilnens of Hapalops, as illustrated by Scott ( 1903-1904 ), a pair of ridges run along th» midline of the skull, but these do not diverge in the same mailner as in this adult skull of Pliometanastes. In Pliometanastes the supmoccipital crest is low and relatively flat; in Megalonyx it rises well above the parietals and forms a V-shaped ridge. In the MeGehee crania a deep triangular depression below the external occipital protuberance marks the insertion of the nu- chal ligament. This depression is not observed in Megalonyx, Hap- alops, or Eucholoeops Large sinuses beneath the frontal and parietal bones extend farther posteriorly beneath the parietal in Pliometanastes than in a partial skull of M. cf. ie#ersoni from Hornsby Springs, Florida ( UF 4045). The distance from the posterior margin of the sinus to the supraoccipital crest is about 26 mm in the McCehee cranium, in UF 4045 it is about 63. Furthermore the parietal sinuses in the MeGehee cranium have a lateral pair of vertical partitions in ad- dition to a median partition, whereas Megalonyx ( UF 4045) ex- hibits only the median one. Megalonyx and Pliometanastes differ from Nothrotherium, which lacks a parietal sinus, and from Cho- loepus ho#mani in which the parietal sinus extends all the way posteriorly to the occiput. The posterior portion of the zygomatic arch arises as a ridge ju5t above the squamosal-supraoecipital suture. The shape of the zygomatic arch dorsal and anterior to the glenoid fossa is similar in the MeGehee cranium. to. that in the M. leptostomus, but it does 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 257 not project so far laterally as in the Cita Canyon specimen. As this part of the zygomatic arch is not preserved in the specimens of M. ie#ersoni observed, no comparison with this species can be made. The dorsal border of the zygomatic arch curves laterally as it projects from the skull in Pliometanastes but curves more medi- ally in M. leptostomus and M. ie#ersoni. The sutures of the maxillary, jugal, and lacrimal bones in Plio- metanastes lie in the same positions as in Hal)alops ruetimeyeri, the suture between the lacrimal and jugal running just posterior to the lacrimal foramen, that between the-lacrimal, jugal, and max- illary running between the anteriof opening of the infraorbital canal and the lacrimal foramen and meeting the squamosal bone at the dorsal side of the posterior opening of the infraorbital canal. The jugal in Pliometanastes is directed outward and posteriorly as in Hapalops, quite unlike that of M. leptostomus in which it projects outward and slightly anteriorly. In M. fe#ersoni the jugal is directed downward and posteriorly very close to the skull. In UF 9479 the posterior part of the jugal is fiattened mediolaterally opposite the third molariform tooth, at which point it twists and widens laterally resembling the jugal in Hapalops longiceps ( Scott 1903-1904, pl. 31). A fragment believed to represent the dorsal projection of the right jugal, associated with UF 9479, is more slender anteropos- teriorly than in either Megalonyx or Hapalops. It is flat on the lateral side, rounded on the medial side, and flattened at the dorsal extremity. The overall shape of this fragment resembles the paral- lel-sided jugal of Megalonyx and Nothrotherium and not that of Hapalops, in which the anterior side of the projection is concave and the posterior side convex. The base of the alveolus for the upper caniniform tooth is sub- circular in each of five specimens. Estimated internal measure- ments of the two specimens in which the anterior alveolar wall is preserved are: UF 10337, 10.7 mm anteroposteriorly, 7.0 medio- laterally; UF 10338, 10.4 mm anteroposteriorly and 6.7 mediolat- erally. The base of the caniniform alveolus is no larger than the base of the alveolus for the first upper molariform tooth. This is in sharp contrast to Megalonyx in which the caniniform is consider- ably larger than the first molariform tooth. A small fragment of the, right posterior side of the palate pre- served in UF 9479 differs from Megalon!/x, Hapalops, and Eucho- 258 BULLETIN FLORIDA STATE MUSEUM Vol. 12 loeops in the manner in which the nutritive foramina open into the , - ventral surface. In the MeGehee palate long narrow canals run anteroposteriorly on the ventral surface of the palate, whereas in the other genera/they run dorsoventrally so that only small round foramina appeat'ob the palatal surface. MANDIBLE. - The mandible is represented by two specimens, UF 9450 and 9480, from McGehee Farm. A third specimen from the Withlacoochee River, UF 11941, a symphysis, is referred to the same species. UF 9450, the right anterior portion, includes the Ventral part of the symphysis, caniniform alveolus, and anterior margin of the alveolus of Mi, but lacks most of the predental "spout." UF 9480 includes the posterior two-thirds of the left ramus (Figs. 14 and. 16). Table 12. MEASUREMENT OF PZiometanastes protistus MANDIBLES IN MM (UF 9450) (UF 9480) (UF 11941) Measurennent right left Depth of horizontal ramus at middle of diastema 42.0 Length of diastema 20.0 16.5 16.5 Length of symphysis from caniniform alveolus to tip 87.4 87.9 Anteroposterior diameter of caniniform alveolus 10.0 13.2 12.9 Max mediolateral diameter of caniniform alveolus 8.6 10.9 11.1 Height of condyle above ventral border of angular process 53.0 The two symphyseal specimens of Pliometanastes reveal a num- ber of distinctive features. The symphyseal spout is long, extend- ing 34.0 mm anterior of the caniniform alveoli in UF 11941 and evidently about the same in UF 9450. The predental spout is rounded at the tip . as in Pautocnus and not pointed as in Acratoc- nus. The sides of the spout are parallel, not converging anteriorly as in Paulocnus. In side view the, spout .is flat along the dorsal surface as in Paulocnus, not decurved as in Mesocnus. The sym- physis is deep with a slightly concave outline ventrally. It does not bear a keel as in Pleistocene Megatonyx. The posterior edge of the symphysis lies between the ropts of the lower caniniform teeth. In this respect Pliohietanastes agrees with Megalonyx and differs k 10 mm ~ FIGURE 14 . Mandibular symphysis of Pliometanastes protistus, UF 11941 , occlusal and lateral views. 260 BULLETIN FLORIDA STATE MUSEUM Vol. 12 from Paulocnus and other West Indian megalonychids in which the symphysis extends posteriorly at least as far as the first cheek tooth. The caniniform alveolus presents a rounded to triangular cross section. The lingual wall is slightly convex, niarly flat. The apex of the triangle faces labially. Although no caniniform has been dis- covered, we are confident that the tooth wore on a near vertical anterior surface producing a sharp peint as in similar teeth of Choloepus, Eucholoeops, Pliomorphus, and Paulocnits. The alveolar ridge between the caniniform and first cheek teeth is considerably thicker in Pliometanastes than in typical Megalon!/X ( for example, UF 10348 a small Rancholabrean Megalonyx jaw from Sumter Co., Florida), but as noted above, the same broad- ened ridge is observed in Blancan specimens of Megalonyx. The postcaniniform diastema attains a length of 20.1 mm in UF 9450 and a greater but indefinite amount in UF 11941, thus exceeding the diastema in Paulocnus and approximating that in Megalonyx of similar size. UF 9450 includes the condyle but lacks the coronoid process and the posterior portion of the angle. The horizontal ramus is preserved as far forward as the alveolus of Mi and contains the lateral walls of the alveoli of M2 - MB· The poorly developed mus- cle scars and porous texture of the bone suggest that this specimen is from a young individual. The. angular process does not project ventrally as in Hapalops, but is more horizontal as in Megaton!/x The angle does not attain the dorsoventral height seen in M . ieffeF soni (see Leidy, 1855, pl. 1). The condylar process does not curve dorsally and is set in a more horizontal position than in Megatonyx. The dental canal passes through the alveolus of M3, whereas in Megalon!/x it runs outside the alveolus, being enclosed in the side of the wall of the jaw. The foramen lateral to MB is situated in the same position in the»MeGehee mandible as in Megalon!/x and Eu- choloeops, not on the ascending ramus as in Hapalops. DENTITION. - Only three isolated teeth have been found at the McCehee Farm: two left M49, UF 10341 and UF 9613, and a left upper M21 or M3, UF 9449. The two M+'s differ in size, the antero- posterior diameter of UF 10341 being 9.0 mm, while that of UF 9613 is 7.1 mm and the mediolateral diameter of UF 10341 being 12.8 mm, that of UF 9613, 10.7 mm. The fourth upper molariform, UF 10341, probably represents an older individual as the sides are 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 261 parallel. In UF 9613 the tooth tapers slightly, the crown measuring 10.7 mm in length and the root 11.2 mm. Both teeth resemble the triangular tooth of M. wheatleyi figured by Cope ( 1871), though within the whole sample of M. wheatteyi this tooth varies consider- ably in shape. As Cope ( 1899) states, "The last superior molar, however, differs considerably in form. In some individuals it is tri- angular in section; in others the section is a transverse oval. All intermediate forms occur in Megalonyx wheatleyi." The pattern of wear of the fourth upper molariform teeth in Pliometanastes is similar to that in Megalonyx. The left M2 or M3 of Pliometanastes resembles that of M. lep- tostomus in shape and wear of the crown; both have a slight con- cavity on the posterior side and a slight convexity on the anterior side. Both Pliometanastes and Megalonyx have irregular triangular- shaped second and third molariform teeth, whereas in Hapalops these teeth are squared. In these genera M# most nearly resembles that of Hapalops in shape, but here too the tooth of Hapalops is more squared. The posterior mandibular fragment, UF 9480 described above, indicates that the lower molariform teeth were of the broad trape- zoidal type observed in Megalonyx, Megalocnus, and Paulocnits, with the labial side longer than the lingual. THORACIC VERTEBRAE. - In the MeGehee collection are two thoracic vertebrae, UF 9464 and UF 9468. UF 9464 is complete, while UF 9468 lacks the epiphyses and neural spine. The exact position of these vertebrae in the column cannot be determined. In UF 9464 the neural spine slopes posteriorly, the articular surface for the tuberculum of the rib is convex, and the dorso- external angle of the postefior face of the centrum is not truncated by an articular facet. The articular surface for the capitulum is ovate and concave. The ventral part of the centrum is pierced by two pairs of nutritive foramina and numerous smaller ones. These larger foramina are considerably smaller than those observed in the lumbar vertebra. A single median foramen opens on the dorsal side of the centrum below the neural canal. UF 946S differs from UF 9464 in that the ventral part of the centrum is less rounded and has a slight keel, the concavity be- tween the prezygapophyses is deeper, the articular surface for the tuberculum of the rib is larger, and the neural canal has a slightly greater diameter. These thoracic vertebrae differ from the vertebrae of M. le#er- 262 BULLETIN FLORIDA STATE MUSEUM Vol. 12 F. ~,~ ve- 1464 a.t 1 .:4 A B /,1/+Lia A...~i C ' .1 „ D E ks.:·~:Af.*Mt,P-43 1 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 263 soni californicus figured by Stock ( 1925, pl. 18) in having relatively shorter neural spines, the dorsal end of the spine longer, less rounded, and more horizontal, the centrum relatively thicker antero- posteriorly, the prezygapophyses not sloping ventrally but more horizontal, and the articular surface for the capitulum of the rib more oval and slanting more anteriorly. LUMBAR VERTEBRAE. - Three lumbar vertebrae, UF 9465, UF 9466, and UF 9467, are available. UF 9465 lacks the transverse processes, UF 9466 lacks the transverse processes and neural spine, and UF 9467 lacks the centrum. Each of the specimens differs con- siderably from the others, suggesting that they belong to different parts of the series. In UF 9465, the posterior face of the centrum is deeper and narrower and the fragment of the transverse process is inclined more dorsally than in UF 9466. A small foramen pierces the ventrolateral side of the transverse process in UF 9465, but is lacking in UF 9466. In UF 9467 a large foramen pierces the trans- verse process posterior and slightly lateral to the prezygapophyses. The medioventral edges of the medial postzygapophyses are closest together in UF 9467, and are farther apart in UF 9465 and 9466, the distance between them being 20.6,24.2 and 26.4 mm, respec- tively. The neural spine preserved in two of the vertebrae is broad anteroposteriorly and Rattened. At the dorsal end a terminal en- largement bears two prominent posterior projections. The winglike transverse processes in UF 9467 are broad, fiattened, and intricate in construction. In dorsal view a concavity is observed on the pos- terolateral margin of each process, medial to it a prominence that projects posteriorly, and, medioventral to it, the postzygapophysis. The vertebrarterial canal opens anterior and slightly ventral to the postzygapophysis. CAUDAL VERTEBRAE. - Two caudal vertebrae are available, UF 10342 and UF 10343. In both the neural arch is missing and the transverse processes are incomplete. Both vertebrae are probably from the middle of the series. The centrum of UF 10342 is oval shape, the anterior face meas- uring 40.2 mm in width and 31.3 in height, the posterior face meas- uring 37.4 and 29.3 mm respectively. A single large foramen pierces FIGURE 15. Vertebrae of Pliometanastes protistus, A.-Anterior thoracic ver- tebra, UF 9464, lateral view. B.-Anterior view of same. C.-Caudal vertebra, UF 10342, ventral view. D.-Lumbar vertebra, UF 9465, anterior view. E.- Dorsal view of same. Scale X 1/6, 104/eab+ . - 35a. 4,«. K...f- - ....1/31 ~ . -- , A I. I.: ' 41 F 5- . I ./. -7 - 6 ·1im- Flf :655·' L r Ef'Ajrld#--, 2..'. . 3 9V D ... E 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 265 the median ventral surface of the centrum. Two pairs of promi- nences, at the anterior and posterior extremities of the ventral surface of the centrum, are present for articulation with the haemapophyses ('Fig. 15 and Table 13). CLAVICLE. - A right clavicle, UF 10344, is represented in figure 16. The compressed shaft forms a single arc between the extremities, the curvature being more arcuate than in M. idersoni but lacking the sigmoid curvature in Nothrotherium. It is much more slender, especially medial to the articular facet for the scapula, than in M. le#ersoni. The most rugose area for muscle attachment extends about halfway up the shaft from a point just lateral te the sternal facet. The shaft attains its greatest depth near the middle at the end of this most prominent muscle scar. It also broadens medially to form the articular faeet for the sternum. This surface is not so wide as in Nothrotherium, but is relatively wider than in M. le#er- soni. The head is convex, almost circular in side view, compressed, and slightly wider and deeper than the shaft. HUMERUS. - Three incomplete burneri, a left · ( UF 9445) and two rights ( UF 9444 and 9443), are represented. All lack the proximal end, and only in UF 9445 is the lateral portion of the distal end pre- served. ( Fig. 16). The humerus appears to be a relatively conservative element compared to the Santa Cruz megalonychids. The major changes ap- pear to be in the size and development of the pectoral, medial, and deltoid crests. In Hapalops the junction of the pectoral, medial, and deltoid crests stands prominently above the surrounding shaft. The pectoral crest rises posteriorly and then curves anteriorly, forming a hooklike projection on the medial side. This projection is less prominent in Pliometanastes. Along the anterolateral border of the shaft just posterior to the deltoid crest is a prominent ridge in Plio- metanastes and Megatonyx, absent in Hapalops. In both Pliometanas- tes and Megalon!/x the median ridge formed between the pectoral and deltoid crests is more prominent than in Ha'palops. In Megalonyx dorsal and slightly lateral to the entepicondylar foramen is a convex protuberance that is absent in Pliometanastes and Hapalops. FIGURE 16. Limb elements and mandible of Ptiometanastes protistus. A.- Clavicle, UF 10844. B.-Left mandible, UF 9480, lingual view, C.-Labial view of same. D.-Right mandible, UF 9450, labial view. E.-Occlusal view of same. F.-Patella, UF 9774, femoral view. G.-Left humerus, UF 9445, anterior view. Scale X %. 266 BULLETIN FLORIDA STATE MUSEUM Vol. 12 Table 18. MEASUREMENTS OF Ptiometanastes protistus POSTCRANIAL MATERIAL IN MM Thoracic Vertebrae ( UF 9464 ) ( UF 9468 ) Length of centrum 88.4 Width of centrum measured over anterior face 44.4 42.5 Depth of centrum measured over anterior face 34.4 32.8, Max. width across transverse process 84.2 84.0 Length of neural arch along middle at base of neural spine 51.0 51.7 Height measured from middle of ventral border of posterior face of centrum to end of neural spine 102.0 Dorsoventral diameter of neural canal measured at anterior end 24.9 27.8 TransverscE thickness of neural spine at middle 84.0 Transverse thickness at end of neural spine 18.0 Transverse width across anterior zygapophyses 47.2 49.0 Lumbar Vertebrae ( UF 9465) ( UF 9466) ( UF 9467) Length of centrum 45.6 42.5 Width of centrum measured across anterior face 60.4 61.0 Depth of centrum measured across anterior- facd 40.9 89.8 Width across supports for anterior zygapophyses 59.1 Width across transverse process 154.0 Max. width across posterior zygapophyses 42.6 44.0 40.8 Length of neural arch along middle and at base of neural spine 58.5 48.0 56.0 Height measured from middle 6f ventral , border of posterior face of centrum to end of neural spine 131.0 Dorsoventral diameter of neural canal measured at anterior end 81.3 27.2 Transverse thickness of neural spine at middle 52.0 Transverse thickness at end of neural spine 25.8 27.0 Clavicle ( UF 10844) Max. length 127.0 Min. depth of shaft ( medial to scapular facet) 15.7 Max. depth of shaft ( about halfway between extremities) 25.4 Long diameter of scapular facet 24.4 Humerus ( UF 9448) C UF 9444) C UP' 9445) Max. width of shaft at middle 49.2 40.7 44.7 Min. width of shaft 31.0 Thickness of shaft at end of delt6id crest 48.8 40.5 48.4 Thickness of bridge over entepicondylar foramen 22.4 [Jina ·(UF 9447) Max. mediolateral diameter of olecranon process 87.5 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 267 Width Of shaft at concavity of sigmoid cavity 40.8 DorsoventraI diameter of radial notch 80.0 Metacarpal III ( UF 11525 ) Max. length 66.5 Max. depth of proximaI end 82.9 Max. width of proximal end 82.0 Max. depth of distal end 84.9 Max. width of distal end 24.0 Min. depth of shaft 15.8 Min. width of shaft 21.4 MetacaTpal IV ( UF 11526 ) Max. length 78.5 Max. depth of proximal end 31.0 Max. width Of proximal end 19.5 Min. depth of shaft 14.8 Min. width of shaft 12.1 Metacarpal V ( UF 11527 ) Max. length 64.8 Max. depth of proximal end 28.7 Max. width of proximal ehd 25.0 Max. depth of distal end 22.4 Max. width of distal end 17.6 Min. depth of shaft 11.0 Min. width of shaft 14.1 M€tacarpal II ( UF 11524 ) ( UF 9458 ) ( UF 9454 ) Max. length 56.6 54.9 61.1 Max. depth of proximaI end 80.7 80.5 34.4 Max. width of proximal end 27.5 23.4 29.2 Max. depth of distal end 32.4 31.7 85.7 Max. width of distal end 22.2 22.0 23.4 Min. depth of shaft 14.0 13.7 15.6 Min. width of shaft 17.3 20.1 20.4 PateUa (UF 10347) ( UF 9774) Dorsoventral diameter of femoral facet 88.5 32.0 Transverse diameter of femoral facet 41.1 48.0 Thickness through femoral facet 24.0 20.5 Total length 80.0 Fibula ( UF 9448) ( UF 9446) Length through shaft 224.0 Max. anteroposterior diameter of proximal end 81.0 88.8 Transverse diameter of proximal end ( perpendicular to tibial facet) 42.2 87.6 Anteroposterior diameter at middle of shaft 19.0 Transverse diameter at middle of shaft 16.4 Anteroposterior diameter of distal end 44.5 Max. transverse diameter of distal end 47.0 Calcan€um ( UF 9487) ( UF 9442) Max. anterioposterior diameter 131.0 128.0 Max. diameter of tuber caldis 110.4 Max. thickness at posterior border 24.4 24.0 Width of neck 24.0 20.7 Thickness of neck 34.4 84.4 268 BULLETIN FLORIDA STATE MUSEUM Vol. 12 Distance from inner border of external astragalar facet to inferoexternal prominence of articulating end 58.0 50.8 Distance from dorsal border of external astragalar facet to ventral border of cuboid facet 67.2 68.5 Astragalus ( UF 9440 ) ( UF 9441 ) ( UF 10889 ) Max. anteroposterior diameter parallel to fibular facet 67.0 65.4 Max. transverse diameter measured at right angles to fibular facet 57.2 58.0 57.5 Depth of fibular facet at distal end 31.2 88.9 84.0 Max. transverse diameter of head 35.2 89.0 86.5 Anteroposterior length of external calcaneal facet 48.5 46.0 48.8 Navicular ( UF 9474) Max. diameter across cuneiform facets ( width) 42.8 Diameter taken at right angles to greatest ( dorsoplantar) diameter 41.5 Max. thickness through convexity articulating with astragalus 19.4 Phatanx II of Digit III ( UF 9459) Length through middle 27.5 Width of proximal end 37.5 Depth of proximal end 38.3 Width of distal end 35.6 Depth of distal condyles 84.4 Ungual Phalanges (UF 9460) (UF 9554) (UF 9462) (UF 10840) Max. length from posterior end of over- hanging process to tip of cIaw process 50.0 49.8 Max. depth from dorsal surface of hood to discoid area on ventral surface , 55.0 22.4 81.0 21.4 Length of subungual base 20.0 Width of ventral surface of claw process just anterior to hood 28.0 11.0 15.8 10.3 1Estinrated ULNA. - The proximal half of a right ulna, UF 9447, is illustrated in Fig. 17. The coronoid process and the area for attachment of the brachialis muscle are lacking. The McCehee ulna is intermediate between Hapalops and Mega- lon!,x in several characters. In Hapalops the anterior border of the shaft from the olecranon process to the styloid process forms a single convex arc, whereas in Megatonux this area is concave. In Prome- 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 269 tanastes the anteri6r border of the shaft is straight dorsoventrally. The lateral projection of the olecranon is considerably greater in Megatonyx than in either Pliometanastes or Hapalops. In Hapalops the posterior border of the shaft is slightly convex dorsal to the sig- moid cavity, in Pliometanastes it is nearly straight, and in Mega- lonyx it is concave, and slopes less steeply posteriorly than in either Hapalops or Pliometanastes. The olecranon process in Pliometanastes is relatively wider across the anterior border than in Hapalops. The vertical ridge along the posterior border of the olecranon process is sharper and less rounded than in Hapalops. The dorsoventral height of the radial notch is much greater in Pliometanastes than in Hapalops or even in Mega- longix. In Pliometanastes and Hapalops the radial notch is not dis- tinctly separated from the sigmoid cavity by a groove as in M. 14- fersoni californicus ( Stock, 1925). The radial notch faces almost directly anteriorly in Megalon!/x , considerably to the side in HOT)a- tops; in Pliometanastes its orientation is intermediate. MAN'US. - Four right metacarpals, numbers II, III, IV, and V, ( UF 11524 - UF 11527), were collected together at McGehee and appear to belong to one individual ( Fig. 18). Metacarpal IV is damaged and lacks the anterolateral posteromedial portion of the distal end above the carina. There are also two isolated metacarpal II, a right ( UF 9453) ( Fig. 19), and a left ( UF 9454). The corre- sponding metacarpal elements of Pliometanastes, Hapalops and Meg- alonux are all closely similar, the major differences being functions of the diferences in size. The fourth metacarpal of Pliometanastes is the longest, as in Megalonyx. The fifth metacarpal is slightly shorter than the third, whereas in Megatonyx the fifth is second in length. The available specimens of metacarpal II of M. ie#ersoni, ANSP 12507 and ANSP 12475, differ notably in the width of the shaft, length ( ANSP 12507 is shorter and more robust than ANSP 12475), and degree of concavity of the facet for metacarpal III. The trape- zoid facet of M. ie#ersoni is almost uniformly concave, whereas in Pliometanastes the surface becomes concave near the dorsal extremity of the facet, and convex ventrally as it does in Hapalops. The dorso- lateral process of the proximal end is extremely prominent, thus ac- centuating the sigmoid shape of the facet. This process is not quite so prominent in Hapalops and is almost lacking in M. ie#ersoni. The palmar extremity of the trapezoid facet cilrves anteriorly in Pliometanastes, whereas in M. fe#ersont it is flattened in the dorso- - A B f f f1 4/- fip 3 ...'9 F- 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 271 palmer plane. The dorsal end of the facet proximal to the facet f6r metacarpel III is convex in Pliometanastes, concave in ie#ersoni. The dorsolateral extremity of the trapezoid facet proximal to the facet for mc I does not extend so far dorsally as in te#ersoni. UF 9453 has a ridge from the dorsomedial end of the facet for metacarpal I to the medial projection for the articulation with phalanx I, and on either side of this ridge the shaft is concave. In UF 9454 and 11524 no ridge is apparent and the shaft is concave dorsally and convex ventrally. In both specimens of ie#ersoni the shafts are completely round. The shape of the lateral side of the shaft in UF 9453 closely resembled that of Megalonyx, ANSP 12475, but in UF 9454 the ventrolateral side of the shaft is more deeply excavated. The shape of the distal end of metacarpal II is similar in both Pliometanastes and Megalonyx. On the dorsomedial side of the distal end proximal to the projection for the articulation with phalanx I, a distinct ridge of bone runs from the dorsal end of the carina ventrally, with a depression below it. This ridge and depres- sion are much more pronounced in Megalon!/x and Hapalops than in the MeGehee metacarpal II UF 9453 and 11524, and they are even less distinct in UF 9454. Metacarpal III of Pliometanastes, UF 11525, closely resembles that of Megalonux and Hapalops. The differences lie in size and in the shape of the facets of the proximal end. In Pliometanastes, as in Hapalops, the facet for metacarpal IV consists of two parts that are continuous with one another. The dorsal end of the facet is oval with the long axis extending down the shaft. The palmar part is also oval but the long axis runs transversely; the facet is thus shaped like an English saddle in side view. In Megalon!/x the facet for meta- carpal IV consists entirely of the dorsal part with little or no palmar extension. The facet for metacarpal II is similar in shape in Pliometanastes, Megalonyx, and Hapalops. In Pliometanastes the palmar half of the facet turns medially so that it almost faces ventrally; in Megalon!/X and Hapalops the facet curves more gently and does not turn so far ventrally. The shaft in Pliometanastes is more slender and rounded than in Megalon yx. This appears to be the result of increase in size, FIGURE 17. Ulna and Ebula of Pliometanastes protistus. A, B.-Proximal half of right ulna, UF 9447, outer and radial views. C, D.-Right fbula, UF 9446, tibial and outer views. E, F, G.-Left metatarsal IV, UF 11572, left, plantar, and outer views. Scale, A-D x 56, E-F, natural size. rn : ..SA*-/ /a• / p , F .4/*--#77* - 1*JI~~ > r /4 -=*,~ ,-=-I co eu, . 17 0--4-5 CD S 0 R /#,8 > 1-r-' i- - I u , , S~C i~, 4' 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 273 as metacarpal III of Pliometanastes most closely resembles the small- er, more slender element of Megalonyx from the Hagerman beds in Idaho. Metacarpal IV of Pliometanastes, UF 11526, is a long, slender bone with a dorsopalmar expansion of the proximal and distal ends. It resembles that of Hapalops, and differs from both in the extreme mediolateral compression of the proximal end. The shaft is more slender and rounded than in Megatonjx, especially at the distal end. At the proximal end the facet for metacarpal V is similar in shape to that of Megalonyx. The facet for metacarpal III is also similar, but it is flattened against the shaft, whereas in Megalonyx and Hapalops it curves outward. The compression of the proximal end where the facet is flattened for metacarpal III is striking; the palmar portion of the proximal end in Pliometanastes is a narrow, pointed projection, whereas in the other two genera this part is wider and somewhat squared. The ilattened facet for metacarpal III brings the two metacarpals closer together, especially at the dis- tal end. The facet for articulation with the metacarpal III extends farther ventrally than in Megalonyx. The facet for the unciform is similarly shaped in both Pliometanastes and Megalonyx, except for the extreme narrowing of the palmar end in Pliometanastes. The distal end of this element closely resembles that of Megalongx Metacarpal V of Pliometanastes, UF 11527, ii slightly shorter than metacarpal III, UF 11525. This is in sharp contrast to the rel- ative lengths of these elements in Megalonyx, where mc V is con- siderably greater in length than metacarpal III. The shorter distance between the proximal and distal articular processes of metacarpal V of Pliometanastes make it more robust in appearance than that of Megalonyx. The proximal end is irregularly triangular in shape as in Megalon!/x. The distal end is considerably shorter in the dorso- palmer direction than in Megalonyx. On the proximal end the surface for articulation with the lateral extension of the cuneiform is almost square in outline and presents a relatively flat surface. Situated on the dorsomedial side of the cuneiform facet is the unciform facet which is confluent medially FIGURE 18. Associated metacarpus of Pliometanastes protistus. A.-Right met- acarpal II, UF 11524, inner view. B.-Dorsal view of same. C.-Right metacar- pal III, UF 11525, dorsal view. D.-Outer view of same. E.-Right metacarpal IV, UF 11526, outer view. F. and G.-Dorsal and inner views of same. H.-- Right metarcarpal V, UF 11527, inner view, I, and J.-Outer and dorsal views of same. All natural size. 274 BULLETIN FLORIDA STATE MUSEUM Vol. 12 I I nl f®/ A 4 0 I. 1. T F e , 7 ~+ a./. J 1 -4.-L= D E F G 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 275 with the facet for the fourth metacarpal. The unciform facet is semi- circular in outline, and the surface forms a smooth sigmoid curve with the concave portion on the dorsal and the convex portion of the palmar part. A similar sigmoid shape appears in the facet for metacarpal IV but in reverse, the convex portion dorsal and the concave palmar. On the far lateral side of the proximal end is an oval area for ligamentary attachment; the shaft is depressed on either side of it. The median part of the shaft is oval and compressed in the dot- sopalmar direction. The facet for the first phalanx at the distal eild is broad and only slightly convex, unlike the other metacarpals of the manus which have a large rounded carina. This indicates considerable restriction of movement in this last digit. Patella. - The articular facet of UF 9774 is kidney-shaped, the medial and lateral sides being almost symmetrical ( Fig. 16). In UF 10347 one side of the facet extends farther ventrally than the other. The patella has proportionally 16nger distal projection below the facet than in either Hapalops or Megatonyx; the ventral projection does not narrow below the facet as in those genera, but remains relatively wide to the distal extremity. Fibula. - One complete right fibula, UF 9446 ( Fig. 17), a proximal end of a right fibula, UF 9448, and an immature left fibula, UF 9555, lacking the epiphyses, are represented. In the MeGehee fibula the shaft is essentially straight; whereas in Megalon!/x the shaft is curved posteriorly distal to the region for at- tachment of the peroneus brevis muscle. The shaft in this region is more enlarged and rugose in Megalonyx than in Pliometanastes. The scar is considerably less distinct in Pliometanastes than in Megalonyx. In Megalon!/x fibulae a ridge running from the tibial facet to the dor- sal margin of the muscle scar gives the shaft a triangular slope, whereas in the MeGehee fibula this ridge is lacking and the shaft is round. In Pliometanastes the calcaneal facet is distinctly concave with a slight papilla at the junction of the astragalar and calcaneal facets. The astragalar facet is likewise concave, whereas it is flat in Megalonyx fibulae, which lack the raised area at the junction of the two facets. FIGURE 19. Podial elements of Pliometanastes protistus. A.-Right metacar- pal II, UF 9458, inner view. B.-Right navicular, UF 9474, astragalar surface. C.-Cuneiform surface of same. D.-Left astragalus, UF 9440, tibial view. E. and F.-Fibular and calcaneal views of same. G.-Right calcaneum, UF 9437, Outer view. H.-Inner view of same. Scale X M. 276 BULLETIN FLORIDA STATE MUSEUM Vol, 12 In Pliometanastes and Megalonyx the proximal end of the fibula is shorter and broader than in Hapalops. A deep depression appears on the anteromedial side of the proximal end in these genera but not in Hapalops·; a depression in Hapalops posteromedial to the tibial facet of the proximal end is absent in the other genera. Calcaneum. - Four calcanea ( Fig. 19) are represented, three right, UF 9437, 9438, and 9442, and one left, UF 9439. Only UF 9437 is complete; UF 9439 from an immature individual lacks the epi- physis. The calcaneum of Pliometanastes retains several primitive fea- tures associated with Hapalops and its allies, the major differences re- suit from an increase in size. The length of the tuber calcis is intermediate between those of Hapalops and Megalonyx. An adult calcaneum of a small Megalon!/T from Sumter County, Florida, is no larger than the McCehee calcan- eum yet has considerably longer tuber calcis, The McGehee calcan- eum also differs from the Sumter County Megalon!/x specimen in lack- ing an anterodorsal projection of the tuber calcis Pliometanastes differs from all species of Megalon!/x observed including the Sumter County specimen, in having the dorsal region of the tuber calcis proportion- ally thicker. Apparently the degree of proximity of the astragalar facets and the depth of the grooves separating them vary considerably within the genus Megalonyx. In Plioinetanastes, as in the Blancan M. lepto- stomus, the grooves do not appear to be so deep as those described by Stock ( 1925) and Leidy ( 1855) for M. ie#ersoni. In Pliometanastes the two astragalar facets are almost confluent, whereas in Hapalops, M. leptosto,nus and the Sumter County Mega- lonyx the large external astragalar facet is set ventrally below the level of the smaller internal astragalar facet. In the McCehee calcaneum the cuboid and internal astragalar facets are closer than in Hapalops and almost touch: The distance between - them varies in the different species of Megalonux and appears to be greatest in M. leptostomus; the distance is greater in all Megal.on!/x than in Pliometandates. The shape of the external astragalar facet in Pliometanastes is identical to that of Hapalops, both sides being mirror images of each other. In Megalon!/X the side of the facet dorsal to the internal astrag- alar facet is longer than the side dorsal to the cuboid facet. The shape of the cuboid facet is similarly almostj circular and slightly concave in Hapalops, Pliometanastes, and jMega/onux, but Pliometanastes does not have the inner lip flattened almost at right 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 277 angles to the rest of the surface as Stock ( 1,925) describes for M..lef- fersoni californicus. This inner lip is also absent in M. leptostomus and the Sumter County Megalonyx. The internal astragalar facet in Pliometanastes is more ovate than that of Hapalops, which is almost round. It differs from that of Mega- lon!/x mainly in the dorial margin, which is convex and curves toward the external astragalar facet as in Hapa/ops. fn Megalonyx the dorsal margin is concave and forms a distinct separation between the two facets. The lateral area for attachment of the peroneus muscle posterior to the cuboid facet varies in shape and thickness in Pliometanastes, Hapatops, and Megalonyx. The McGehee calcaneum has a distinct and relatively wide groove immediately lateral to the external astrag- alar facet. This groove is less well defined in M. lepton!/x, and is only a slight depression in M. leptostomus and Hapalops. In Plionieta- nastes, as in Hapalops, the projection for attachment of the' peroneus is large, relatively long, and wide; in idersoni, in M. leptostomus, and the Megalonux from Hagerman, Idaho, it is considerably smaller, shorter, and narrower. The area for attachment of the peroneus thus decreases with increase of size in Megalonux, but in Pliometanastes more closely resembles Hapalops than the small species of Megalonux Astragalus. - Three left astragali, UF 9440, 9441, and 10339, in the collection from McGehee ( Fig. 19) all are complete, well preserved, and appear to represent adult individuals. The most conspicuous difference between a MeGehee astragalus and those of M. ie#ersoni and M. wheatle!/i is its more rounded shape. The medial side is shorter than the lateral side in Pliometanastes whereas in M . iefjersoni and wheatle !/ i the medial and lateral sides are more nearly equal. The McCehee astragalus more closely resem- bles Hapatops in this condition. This increase in the width of the medial side of the stragalus appears to be related to increase in size, as a specimen of Megalon!/X from the Santa Fe River in Florida, which has an astragalus smaller than the MeGehee astragalus, is relatively narrow on the medial end as in Pliometanastes. All specimens 6f Megaton!/x examined have a deep, V-shaped notch between the tibial. and navicular facets to receive a process of the tibia. UF 10339 and 9440 have only a vague depression, UF 9441 a very shallow notch. Although no tibia is represented in the McCehee collection, the lack of a deep notch in the McCehee astragali implies the lack of a very well developed tibial projection. This is of consid- erable interest, because the tibia Stirton ( 1939) described from the 278 BULLETIN FLORIDA STATE MUSEUM Vol. 12 Mulholland Fauna in the San Francisco Bay Region has a well devel- oped and quite prominent astragalar spine on the distal end of the, tibia. In Hapatops, PU 15913, a relatively deep notch is present. The internal calcaneal facet is .not separated from the external calcaneal facet by a wide, distinct furrow as in all specimens of Mega- lonyx observed. The internal calcaneal facet in UF 9440 is connected to the external calcaneal facet by a low ridge. In the other two speci- mens there is only shallow depression between the two facets. In Hapalops, PU 15913, as in Megalonyx, a distinct separation exists. The internal calcaneal facet varies considerably in shape in the specimens of Megalonyx available, but in all cases it is either rela- tively flat or slightly convex. In Pliometanastes it is slightly concave and somewhat diamond-shaped, with apex curving up toward the ex- ternal calcaneal facet. The shapes of the external calcaneal facet and flbular facet are· similar in Pliometanastes, Hapalops and Megalon!/x The navicular facet in the McCehee astragalus is ovate, having almost a tear-drop shape, whereas in Megalonyx it is considerably deeper in a dorso-plantar direction. NAVICULAR. - A right navicular, UF 9474, is shown in Figure 17. The McCehee navicular exhibits the condition seen in Hapalops where the mesocuneiform facet is raised above the ectocuneiform facet, whereas the opposite condition exists in M. le#ersoni and the Hagerman megatonyx where the mesocuneiform is lower than the ectocuneiform facet. The two cuneiform facets do not meet each other as a distinct ridge in Pliometanastes as they do in Hapalops, M. le#ersoni, M. sier- rensis ( Sinclair, 1905), the megalonyehid fr6m the Snake Creek Beds of Nebraska described by Sinclair ( 1915), and the Megalonyx frorn Hagerman, Idaho. Instead in the MeGehee navicular the mesocunei- form facet slopes gently to merge with the ectocuneiform facet. In the navicular of Pliometanastes and Hapalops, the mediolateral diameter ( width across the cuneiform facets ) is greater than the dorsoplantar diameter. This condition is also found in the navicular, PU 12970, from the Snake Creek Beds ( Sinclair, 1915 ). In M. ie#er soni, M. sierrensis and the specimen from Hdgerman, Idaho the mediolateral diameter across the cuneiform facets is less than the dorsoplantar diameter. The mammillate Surface in the McGehee navicular is less rounded than in M. je#ersoni. The navicular in both Pliometanastes and Hapalops lacks the deep notch on the median side of the ectocuneiform facet, between the 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 279 ectocuneiform and mesocuneiform facets, which is prominent in the later species of Megalonyx and in M. curuidens from the Snake Creek beds. Instead, in the MeGehee navicular, as in Hapalops, the medial border of the ectocuneiform facet slopes gently upward to meet the median ridge surrounding the medial border of the mesocuneiform facet. METATARSAL IV. - A nearly complete left metatarsal IV, UF 11572, is represented in the MeGehee collections. The element lacks the dor- sal projection of the proximal end, broken slightly ventral to the level of the dorsal part of the facet for articulation with metatarsal III, as well as lacking the exteriodorsal projection of the distal end above the carina. Metatarsal IV of Pliometanastes is slightly shorter than in Mega- lonyx from Hagerman, Idaho and about the same length as the mega- lonychid from the Mulholland site II, California, described by Stirton ( 1939). The greatest length of the McCehee metatarsal measures 82.5 mm while the Hagerman Megalonyx element measures 87.3, and the Mulholland mqtatarsal measures 82.7. Metatarsal IV of Pliometanastes resembles Hapalops in retention of a long and slender shaft and thus differs from the Hagerman Mega- lonyx in which the element is short and robust with relatively less distance between the more massive proximal and distal extremities. The portion of the cuboid facet that is preserved is quadrilateral in shape and extends to the posteroventral margin of the element as in Hapalops and the Hagerman Megalonux but differs in the latter aspect from the Mulholland specimen in which the cuboid facet slopes sharply toward the facet for the metatarsal III. The facet of metatarsal IV for articulation with metatarsal III is elliptical in outline and slightly convex in Pliometanastes. In this character it is similar to the Mulholland metatarsal IV and differs from the Hagerman Metalonux and Hapalops in which the facet is concave. This character, however, appears to vary considerably within the genus Megalonux. The facet for articulation with the metatarsal V is parallel to the shaft and is subtriangular in shape as in the Hagerman Megalonyx and the Mulholland specimen. The facet of Pliometanastes differs from the Mulholland specimen in that the lower portion of the facet of Pliometanastes does not flare outward but is flat and very slightly de- pressed into the shaft. The region of the proximal end immediately posteroventral to the facet for the metatarsal V is depressed in Pliom€tanastes, unlike 280 BULLETIN FLORIDA STATE MUSEUM Vol. 12 the situation in the Hagerman Megalonux and the Mulholland meta- tafsal in which this region is Ievel with the facet or in the case of Megaton!/x, slightly bulging. Pliometanastes differs from the Hagerman Megalon!/x and the Mui- holland specimen in the posteroventral view of the proximal end. In Pliometanastes the dorsal area of muscle attachment is relatively small, somewhat square in shape, and projects outward from the shaft, but not nearly so prominently as a projection in Hapolops. Where the shaft curves upward t6 meet this region of attachment lies a concavity that is more distinctly demarcated in Pliometanastes than in the Ha- gerrnan Megaton!/x or the metatarsal from Mulholland. In the Hager- man specimen, the region for muscle attachment covers a much greater area of the proximal end and shaft than in Pliometanastes or the Mulholland metatarsal. The Mulholland metatarsal differs from those of Pliometanastes and the Hagerman Megalon!/x in that the cuboid facet does not. extend all the way to the ventral margin of the proximal surface .but terminates just ventral to the level of the ventral margin of the faeet for metatarsal III. A specimen of M. cf. wheatle!/i from Florida, discussed by Simpson ( 1928, Figure 8) exhibits a similar condition with respect to the foreshortened cuboid facet and depres- sion at the extreme ventral side of the proximal end. The metatarsals of Pliometanastes are notable for their slender proportions and for their terminally restricted muscle scars, In the small Blancan Megalonyx from Hagerman, Idaho for example, the least depth of the shaft of metatarsal IV is 27.7 mm and the least width is 23.5, whereas in P. protistus the corresponding dimensions are 16.4 and 18.5 mm. Although the areas of muscular and tendinous attachment at the extremities of this element are closely comparable in these two taxa, they are smaller and occupy proportionally less of the shaft in Ptiometanastes. Although the genus Megalonux encom- passes considerable variation in these same features-compare the stocky specimen of M. ie#ersoni californicus Stock ( 1925, fig. 49) with the slender one of M ie#ersoni in Leidy ( 1855, pl. XIII)-Plio- metanastes falls well beyond the scope of variation observed in Mega- lonyx. PHALANX I OF D=rm.-A single well preserved right phalanx I of digit III of the pes, UF 9459, is represented. Its distal end differs considerably from the pes of Hapalops and Megalonux in the shape of the condyles, the depth of the median groove between the condyles, and the proximity of the condyles to each other. The distal condyles of the McCehee phalanx I are considerably 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 281 more rounded.than in Hapalops or Mega/on!/x, where they are quite flattened. In these two genera the lateral wall of each condyle is flattened to form a ridge, whereas in- Plioinetanastes the lateral wall is smoothly rounded with no ridge. The groove between the condyles is relatively deeper and nar- rowers the condyles being closer together in the McGehee phalanx 1 than in either Megalonyx or Hapalops. A ridge around the dorsal mar- gin eneloses a pit at the dorsal end of the groov.e between the con- dyles in Pilometanastes. This ridge is poorly developed in Hapalops and is absent in Megalonyx Both Pliometannstes and Hapalops lack the dorsal lip above the proximal articular surface for metatarsal lII, which is well developed in Megalonux. The proximal articular facets for the metatarsal III in Plionietanastes are similar in size to those of Hapalops and Megalon!/x. In all three genera the medial facet is larger than the lateral and the proximoplantar facet on the medial side is about twice as large as on the lateral side. The proximoplantar facets in the MeGehee phalanx I are most similar to those in the same element of Megalonyx. In both genera the medial facet is trapezoidaI; the lateral facet is triangular in Megalonyx and nearly so in Pliometanastes. In Hapalops these facets project dis- tally, but not so in these two genera. In both Pliometanastes and Megalon!/x the proximoplantar border is nearly flat, whereas in Hapalops it is distinctly concave. The proxi- mal facets tor metatarsal III agree with those in Hapalops and differ from those in Megalonyx in having the broad articular facets meet the facet for the carina abruptly at the right angles , whereas in Mega- lon!/x the broad facets turn anteriorly and grade into the carinate facet. In Pliometanastes phalanx I lacks the hook-like projection above the medial facet for mt III in Megalonux and Hapalops. The distal articular surface of the first phalanx is relatively well rounded in Plio- nietanastes, indeed even the phalanges of Hapalops are more flattened on this surface. This contrasts sharply with the fiattened articulation in Megalont/x which restricts movement between the first and second phalanges and leads to fusion of these elements in late Pleistocene time. UNGUAL PHALANGES.-Five ungual phalanges are preserved in the MeGehee collection. They are readily recognized as megalonychid claw cores by their strong lateral compression, with a triangular cross-section as in Megalonyx and the Santa Cruz megalonychids, 282 BULLETIN FLORIDA STATE MUSEUM Vol. 12 The largest claw in the collection, UF 9460 is believed to repre- sent phalanx 3 of digit III of the pes. The bony sheath is partially pre- served on one side, and the very anterior tip is lacking. The base is relatively more robust than in any of the other McCehee claws. On the anterior half of the base is a well-marked discoid area for tendon attachment. Posterior to this area two nutritive foramina, equal in size, open to the distal part of the claw. The dorsal surface of the elaw is broadly convex at the proximal end and narrows about halfway to the tip of the phalanx. From this point to the tip it has a narrow dor- sal edge and is triangular in cross section. The two articular concavi- ties are symmetrical, of equal width and depth. At the dorsal and plantar extremities of the median ridge between them the ridge is squared instead of pointed as in the other claws. The posterior dorsal surface above the median ridge is deeply indented. The smallest claws, UF 10340 and UF 9554, are complete and lack only the bony sheath. The area for tendon attachment on the base is a small, raised area anterior to the two nutritive foramina, but it is not a well-marked discoid area as in phalanx 3 of digit II of the pes. The dorsal surface is not so broadly convex at the proximal end as in the large claw of the pes. The dorsaI surface does not narrow as in that element to form a sharp ridge, but remains more nearly convex to the anterior tip. The two articular concavities are not divided by a distinct median ridge as in the phalanx 3 of digit III. A slight rise be- tween the two concavities fades out comfletely at the dorsal half of the inner surface. In UF 9554 the dorsal and plantar extremities of the proximal end are more pointed than in UF 10340, and the dorsal border of the proximal end is not indented as in phalanx 3 of digit III of the pes. The claw UF 9462 is intermediate in size between the claws dis- cussed above and lacks the bony sheath and anterior tip. It differs from the others in that the articular concavities are not so symmetri- cal, and one concavity extends farther proximally than the other. The median ridge between the two concavities is well defined. This median ridge is pointed, at both dorsal and plantar proximal extremities, un- like the squared condition in the phalanx 3 digit III. On the dersal border, above the articular surface for phalanx II, is an indentation, but it is not so deep and well defined as in phalanx 3. A claw from the small Sumter County, Florida Megalon!/X is slightly larger than the McGehee claw UF 9462. The base of the Sumter County claw is mote robust and the dorsal projection at the proximal end is longer with a greater dorsoplantar height than in 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SIX)TIIS 288 Pliometanastes. Along the entire dorsal length of the phalanx the claw core of the Sumter County specimen is less compressed than in Pliometanastes. However, these may not represent the same digital elements. ?Pliometanastes galushai, new species Typ~.-FAM 77811, right half of mandibular symphysis with ca- niniform root and part of right ramus containing two anterior molari- form teeth, ETYMOLoGY.-Patronymic for Ted Galuslia of the American Mu- seum of Natural History, in recognition of his extensive contributions to vertebrate paleontology in general, and to North American gravi- grade collections in particular. I)IAGrosis.-A small megalonychid with long, relatively deep sym- physeal spout, posterior end of symphysis below caniniforms, triangu- lar caniniforms, and transversely broadened, subrect:ingular molari- forms. AGE AND LOCALITY.-Hemphillian, froin horizon stratigraphically between Round Mountain and San Juan Quarry. Collected by Ted Galusha in 1947 below lower Tuffaceous Zone, San Juan locality, north of Espanola, Lyden Quadrangle, New Mexico. DE.scRIPTION.-The type and only specimen is from a subadult, as indicated by the slight taper along the teetli and by the porous nature of the bone. ( Fig. 20). . 40 - , 9.-r j 20mm FIGURE 20. Mandible of Pliometanastes galushai, FAM 77811, lateral and oc- clusal views. The symphyseal region produces a long spout that reaches at least 14 min anterior to the caniniform. Whether the spout was pointed or parabolic at the tip is not determinable. The anteroventral border presents a convex to Halt profile. As in Ptiometanastes the symphysis terminates posteriorly between the caniniform roots, not farther pos- terior as in Acratocm,s and most West Indian megalonychids. The symphyseal depth at this point is 24.5 mm. 284 BULLETIN FLORIDA STATE MUSEUM Vol. 12 As in Pliometanastes and several West Indian megalonychids, the caniniform tooth presents a subtriangular cross section. The lingual wall is slightly convex; the apex of the triangle is labial. The post caniniform diastema is 13.1 mm long; it is longer than in Paulocnus and approximates that of Plioinetanastes protistus. The two molariforms are wider than long ( 10.0 x 5.3 mm). Unlike cheek teeth of Pliometanastes and Megatongx, they are subrectangular rather than subtrapezoidal in cross section. The first and second cheek teeth closely resemble one another. In their shape and small size, they present a strong resemblance to those of Acratocnus or Hapalops. The unique combination of features in the type clearly separates this species from other North American and West Indian megalony- chids, but the limited material makes it difficult to decide to which described genus it has closest affinities. If comparable juvenile speci- mens of P. protistus were known, the two species might be seen to be closely related. The structure of the symphysis and caniniform both agree more closely with those in P. protistus than with those in other known megalonychids, but the geometry of the cheek teeth suggests fundamentally different and more primitive relationships. The post- caniniform constriction may be a less important, possibly ontogenetic difference. Further interpretations must await better material of these taxa. Other Hemphillian Pliometanastes DUNNELLON PLANT 6, FLORIDA. - A well preserved black second left metacarpal ( V-2877, Florida Geological Survey), collected from Plant 6, Dunnellon Phosphate Company near Dunnellon, Marion Co., Florida, is almost identical in size and features to that described from McGehee Farm. On that basis and because of its geographical prox- imity it is assigned to Pliometanastes protistus. Box-T, HIGGINS, OKLAHOMA. - M. F. Skinner collected three iso- lated megalonychid teeth in 1939 from Pit 1 of the Box T quarries. A pair of lower caniniforms, FAM 77809, and a lower right first molariform, FAM 77808, probably belong to a single individual. The caniniforms clearly would fit the alveoli in P. protistus. They are subtriangular in cross section. with the presumed lingual side slightly convex, the apex of the triangle labial. At the occlusal surface the tooth measures 11.7 mm anteroposteriorly and 10.4 transversely. The shearworn surface slopes anteriorly, but the angle is only about 20 degrees below the horizontal, as compared with an angle of over 50 \L 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 285 degrees in Chotoepus. Pliometanastes caniniforms do not change shape in any important way with age, as the cross sectional shape remains unchanged through a length of over 48 mm in these speci- mens. A faint ovate wear facet occupies the posterior vertical wall of the caniniform. The first molariform is slightly longer on the labial than on the lingual side. It measures 12.6 long by 15.0 wide. Faint grooves occupy the lingual and anterior faces. NORTH SANTA CLARA CANYON, NEW MEXICO. - The posterior por- tion of a braincase, FAM 77810, was collected in 1946 by Ted Galusha from the third promontory in Santa Clara Canyon north of Pueblo San Juan. Its stfatigraphic position lies between those of the San Juan and Leyden quarries. The specimen agrees in size and detailed morphology with the cranial material of Pliometanastes from McGehee Farm. In particular this specimen exhibits the low occiput, absence of sagittal crest, protrusive otcipital condyles and nuchal depres- sions characteristic of Pliometanastes. KLIPSTEIN RANCH, CALIFORNIA. - From the Klipstein Ranch 3, Upper Caliente formation of the Caliente Range of California, UCMP V-5684, are several fragments of an upper dentition plus Ml - Ms of the right side and Mi - M2 of the left side ( Table 14). The teeth of the right side are all broken below the occlusal surface, except the caniniform which has a sharp-pointed wear surface. Table 14. MEASUREMENTS OF UPPER TEETH OF Pliom€tanastes FROM KLIPSTEIN RANCH, CALIFORNIA IN AIM Right Left Anteroposterior Mediolateral Anteroposterior Mediolateral 13.4 9.0 18.5 9.3 Mt 12.2 16.4 12.61 17.8, 18.2 19.5 14.4 19.8' M" 11.3 19.5 1Measured at 111: o.chisal surfaces: all others measured at Ipperniost m#gil. The caniniform is subtriangular approaching trihedral as in Plio- metanastes. It measures 13.4 mm anteroposteriorly and 9.0 mm mediolaterally and is smaller than any of the molariform teeth, unlike the situation in Megalonux where the caniniform is larger than the molariform teeth. The narrowest part of the caniniform is anterior, opposite that of the caniniform alveolus of the mandible from McCehee Farm. The medial side is slightly convex forming an acute angle at the posteromedial side. On the posterolateral side the tooth bulges outward completing the triangle. The shape --- 286 BULLETIN FLORIDA STATE MUSEUM Vol. 12 of the caniniform is the only feature that distinguishes this specimen from Megatonyx. The molariform teeth are smaller than those of Megalonyx Zeptostomus but do not differ significantly in shape. Other Hemphillian Megalonychidae Jnf SWAYZE QUARRY, KANsAs. - A left intermediate upper molar- iform tooth, FAM 77804, was collected by M. F. Skinner from the Jim Swayze Quarry of Hemphillian age. It has a heavily worn anterior slope, is narrower labially than lingually, and has a shallow concavity on the labial side. Its dimensions are 19.7 by 12.5 mm. It cannot be determined with certainty whether this tooth pertains to Megalonyx or to a large individual of Pliometanastes. GUYMON, OKLAHOMA. - A mature left intermediate upper molar, FAM 77806, is from the Guymon Quarry. Its dimensions are 14.0 by 9.8 mm. It is heavily worn anteriorly. In occlusal view the posterior side is slightly convex. The lingual side of the tooth is considerably wider than the labial side. HEMME HILLS, CALIFORNIA. - An essentially complete left fibula was collected from the Hemphillian Hemme Hills Site, UCMP V-5053, Contra Costa County, California. This fibula is small, the greatest length measuring 235.7. The shaft is distinctly curved distally and has a prominent area for attachment for the peroneus brevis muscle, as in Megatonyx. MT. EDEN, CALIFORNIA. - Frick ( 1921) described a mature inter- mediate upper molariform tooth from the Mt. Eden fauna. Its dimen- sions are 18.8 by 12.3 mm. It is slightly larger than the corresponding tooth of Pliometanastes protistus, but is an inadequate basis for a generic assignment to Megalonyx. MULHOLLAND, CALIFORNIA. - Four posteranial elements are de- scribed by Stirton ( 1939) from Mulholland, site 2, of California. They include a left tibia, a third and a fourth metatarsal and a median phalanx. The tibia from Mulholland has a prominent astragalar spine and it must be assumed that the astragalus possessed a deep notch to receive it. The astragalus of Pliometanastes lacks a prominent V- shaped notch and presumably a prominent spine on the tibia as well, but until the variability of this character can be assessed, a generic assignment on this basis in unwarranted. RATTLESNAKE CREEK, OREGON. - Sinclair ( 1906 ) described a mega- lonychid claw core for which Ameghino ( 1912) proposed the name Sinclairia oregoniana. The claw cannot be distinguished from those of Megalon!/x or Pliometanastes, so that the name is a nomen dubium. 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 287 Sinclair's supposition that this specimen was from Miocene Mascall deposits seems unlikely; a more probable interpretation is that it came from the PIiocene Rattlesnake beds of the same area. Orm BASIN, OREGON. - A very small proximal phalanx, UO 23275, measuring only 24.7 mm wide, 30.1 long and 22.1 deep, was collected from the Hemphillian fauna of Otis Basin ( Shotwell, 1963). KREBS RANCH, OREGON. - A proximal third manual phalanx, UO 8074, and two large, and a small second phalanx, UO 8057, 8058 and 8060 cannot be identified generically but further attest to the ubi- quity of megalonychids in the Hemphillian of North America. WESTEND BLOWOUT, OREGON. - Three elements of a moderate sized megalonychid, a third metacarpal, UO 9812, a proximal third manual phalanx, UO 9814, and a partial second phalanx, 9813, occur in this late Hemphillian site. This material and that from the previous locality were alluded to by Shotwell ( 1958). RELATIONSHIPS The fundamental subdivisions of sloth phylogeny become evident by Santacrucian time in South America. Among megalonychoids the megatheriids ( including nothrotheres) reduce and then lose their anterior sets of caniniform teeth, at the same time expanding their symphyseal regions into great predental scoops. The true megalony- chids, by contrast, develop a secant, self-sharpening pair of canini- forms, in front of which only a modest predental "spout" is retained. Other major differences, such as the twisted astragalus, subsequently develop between the megatheriids and the megalonychids. Many such differences are still reflected in the contrasts between Bradypus and Choloepus, the two genera of living tree sloths. Eucholoeps represents the Santacrucian megalonychid stock ( Scott, 1903-04). Its arrangement of shearing caniniforms and. moderate sym- physeal spout have been retained by most later megalonychids: by Ptiomorphus and Ortotherium in South America, by Act'atocnus, Miocnus, Synocnus, Paulocnus and most genera in the West Indies, and by Pliometanastes in North America. The exceptional genera in which this system has been considerably modified are Megalocnus in the West Indies, and Megalon!/x in North America. Another primitive feature observed in Eucholeops is the nearly rectangular shape of its molariform teeth. This pattern is retained in Mioenus, S!/nocnus and Acratocnus of the West Indies and in ?Pliometanastes galushai of North America. Most of the late genera develop subtrapezoidal molariform teeth. 288 BULLETIN FLORIDA STATE MUSEUM Vol. 12 Important transitional genera from Eucholeops to later forms are Pliomorphus Ameghino and Ortotherium Ameghino from the late Miocene of South America. In both genera the caniniform teeth remain subtriangular, the symphyseal spout is mederately shortened, and the molariforms subrectangular. The West Indian and North American megalonychid radiations stemmed from some Miocene South American sloths closely allied to Pliomorphus. Paulocnus and Ptiometanastes also lie very near that common stock. Although the West Indian sloths are known only from Pleistocene deposits, their evident diversity implies a much longer occupation of the region, possibly from late Miocene time. It is not yet clear whether this diversity represents one, two or even more independent invasions by South American pioneer stocks. In the most recent study of the group de Paulo Couto ( 1967) avoids explicit discussion of this question; however, his classificatory headings, separating the megalocninae from the ortotheriinae, might imply a double invasion. The simplest interpretation, pending further evidence, remains a single late Miocene invasion of the West Indies followed by an insular radiation. We have compared the various West Indian genera with P/iomet- anastes in search of some common ancestral pattern. As indicated below, however, the comparisions are relatively remote. Megalocnus is the most specialized of the West Indian genera. Some of the major differences from Pliometanastes are as follows: rodent-like specialization of the anterior teeth, palate greatly depres- sed in relation to the basicranial axis, mandible lacking a pre-dental spout and being considerably more robust posterior to the diastema, pectoral and deltoid crests of the humerus wholly separate, and tuber calcis of the calcaneum lacking the degree of posterior expansion observed in calcanea of Pliolnetanastes. Mesocnus differs from Pliometanastes in its meniscoid caniniform, relatively shallower and more slender mandibular symphysis, and the absence of an entepicondylar foramen on the humerus. Megaloc- nus and Mesocnus, however, resemble Pliometanastes and Megalonux in their obliquely set, subtrapezoidal cheek teeth. Microcnus has a shorter, even smoother cranial roof than Plin- metanastes. And it has a shorter narrower symphyseal spout. Its most remarkable specializations are its very sma·ll size and the arhoreal modifications of its skeleton, particularly evident in the: distinct neck of the astragalus. These bradypodid-like features distinguish Microc- nus from all other megalonychid ground sloths. 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 289 The genera Acratocnus, Miocnus and Synocnus have subtriangular caniniform teeth that seem at first to ally them with Pliometanastes ( de Paulo Couto, 1967) but the symphysis in each of these genera is fundamentally different from the Florida genus: the symphyseal spout is shallower, narrower and it is abruptly inflected anterior to the caniniform; the caniniform teeth are set well anterior to the posterior edge of the symphysis, are separated laterally from the symphysial groove by a distinct ridge, and are inclined strongly anteriorly, whereas in Pliometanastes they are nearly vertical. These genera also have more rectangular, transversely implanted teeth than Pliometanastes, Megalon!/x, Megatocnus and Mesocnus in which the teeth are set obliquely and are subtrapezoidal in shape. Postcranial differences separating Acratocnus from Pliometanastes include the long neck of the calcaneum (Anthony, 1916), the relatively small tuber calcis with little posterior expansion, and the long slender proportions of the limb elements. Perhaps the closest comparison to Pliometanastes may be made with Paulocnus petrifactus from eastern Curacao ( Hooijer, 1962, 1964). It is about the same size as the Florida genus and similarly lacks a sagittal crest. The lower caniniform tooth approaches the nearly vertical orientation of that tooth in Pliometanastes, and the symphyseal spout is relatively broader than. in the other West Indian forms. Also the distance between the caniniform and the next tooth is relatively short in Pautocnus as in PNometanastes. However in Paulocnis, the lower caniniforms are not flattened lingually, the symphyseal spout is shorter, and the symphysis extends farther posteriorly, underlying Mi. Comparison of limb elements is rendered exceedingly difficult by the poor preservation of the Curacao material. ' However, it is clear that the neck of the calcaneum is more con- stricted in Paulocnus than in Pliometanastes. From these comparisons we conclude that of all the West Indian genera Paulocnus most resembles Pliometanastes. These two genera probably resemble the ancestral Mio-Pliocene stock that lived in northern South America. Curaca6 lies so close to the mainland of South America that Paulocnus is unlikely to bear any immediate relationship to the radiation (s) that occurred in the Greater Antilles. These other West Indian forms appear to be divided into two broad groups: (1) Mesocnus and Megalocnus, both with subtrapezoidal molariform teeth; and (2), the remaining genera with more rectang- ular molariform teeth. Possibly these stem from two separate inva- 290 BULLETIN FLORIDA STATE MUSEUM Vol. 12 sions; more probably they represent a complex radiation from a single stock allied to Paulocnus· and Pliometanastes. Pliometanastes is clearly the more primitive of the North American megalonychids, but it does not follow that this genus is directly ancestral to Megalonyx. Evidently, Megalonijx already existed along with Pliometanastes in the Hemphillian. By late Hemphillian time the genus had attained its essential progressive features. Although M. mathisi presumably represents a primitive stage within the evolu- tion of Megalon!/x, it occurs too late in the Hemphillian to be part of the original transition. The common ancestor of the North American genera must have been pre-Hemphillian. It presumably ranged south of the United States at that time. The course of development of the caniniform of Megatongs has an important bearing on this phylogenetic reconstruction. Scott ( 1903- 04 ) attempted to recognize an early phylogenetic line with Mega- Zonyx-like canines in his Megalon!/chotherium from the Santa Cruz beds. However, its teeth are in fact no different from those of Eucholeops; we. therefore, support Kraglievich's ( 1925 ) view that these two genera are synonymous. Moreover the new records of Megalonyx suggest that it developed the characteristic canines during or just prior to the Hemphillian in North America. Pliomorphus shows some striking similarities to Megalonyx in the configuration of the skull, particularly the basicranial region. The major differences are found in the low dorsoventrally constricted nasal region of Pliomorphus and its triangular caniniform teeth. In order to understand the significance of the long, ovate, blunt- worn caniniforms of Megalonyx, it is necessary to analyze their masticatory function. In his excellent study of mastication in tree sloths, Sicher ( 1944) emphasized the pattern of dental attrition facets as the key to a functional analysis. In Brad!/Pus the Rat glossy facets occupy the posterior slopes of the lower molariforms and the anterior slopes of the uppers, indicating that the principal masticatory stroke has an anterior component. In Chotoepus the opposite occurs: the principal stroke has a posterior component, even though a lesser protractive closing phase also occurs when the caniniforms are used for cutting. This different emphasis in the two Recent sloth genera is reflected in the zygomatic and mandibular construction ( Sicher, 1944). In Megatonyx, Pliometanastes, and Paulocnus the heavy attrition facets occupy the posterior edges of the lower and the anterior edges of the upper nidlariform, teeth.> The oppositd edges of the teeth are 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 291 more or less flat. The principal masticatory stroke presumably had an anterior component as in Brad!/pus. The same muscular complex could also have served to engage the self-sharpening triangular caniniform teeth in Pliolnetanastes and Paulocnus (Fig. 21). The transitjon from a sharp triangular to a blunt ovate caniniform in Megalonyx is represented by M. mathisi. Placement of the caniniforms more directly above one another would produce blunt horizontal occlusion rather than sharp Vertical occlusion, with no major change in jaw movements. The adaptive value of this change may be associated with a shift from tropical to more temperate forests where the more sclerophyllous leaves were more readily stripped off by the broad blunt caniniform teeth. After this pattern became established, the caniniforms broadened considerably as evidenced by the change from Megalonyx curoidens to M. leplostomus to 1 ater species. As this trend continued, the vertical shearing surface of the anterior caniniforms lost its effectiveness. A longer more horizontal occlusal surface between the canines, however, came to function as an additional grinding mill. 9 POOO Power shear-up and forward -A Roll power grind-up and back CHOLOEPUS stopper MOUU Power stroke -05 13[3 n up ond forward MEGALONYX FIGURE 21 . Diagram of dental wear surfaces in Choloepus and Megalonyx Shear surfaces solid, grinding surfaces dashed. t O LI G O C EN E M IO CE NE PL IO CE NE PL EI ST O C EN E Nuevo Leon Greater AntilleonMegolonyx megalonychid Poulocnus megalonychids nothrotheres megotheres =SY=Sf - "011=00= $==5*§ :.y.=.. 2*.->>*·S. Megalonyx Pliomeionastes Pliomorphus - Ortotherium Complex =~ f.::M......%*:./:......::9:bll:/./.:~ *j...:..~2== **8~-....: M~£.....:... ~~ ~*Re#f 0%80*3: Eucholoeops and allies Hapolops ond ollies Planops ond allies Complex with emphasis on Complex with emphasis on reduction shearing coniniforms or loss of caniniforms Complex Rodiation FIGURE 22. Phylogenetic diagram of the relationships of megalonychid sloths, 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 293 Accompanying enlargement of the caniniforms in Megalonyx, the symphyseal spout was greatly reduced and the bone between and behind the upper and lower caniniforms considerably thickened. M. mathisi represents a primitive stage in this respect as in the little enlarged caniniform teeth. Analogous modifications evidently occur- red in Megalocnus. No other changes in the jaw of Megalonyx have been observed. The articulation and musculature of the jaw and the occlusion of the molariform teeth appear to remain the same in Pliometanastes and Megatonyx. Reducti6n of the symphyseal spout and buttressing of the al- veolar bone behind the caniniforms probably represent structural changes to support heavier wear on the caniniforms. Moreover, the grasping function of the spout may have been supplanted in part by the modified caniniforms. Futhermore the caniniforms of Megatonyx project little above the level of the cheek teeth and thus allow the tongue greater freedom of movement. Suggested relationships between known megalonychids are illus- trated in figure 22. 294 BULLETIN FLORIDA STATE MUSEUM Vol. 12 LITERATURE ClTED Alt, D, and H. K. Brooks. 1965. Age of the Florida marine terraces. J. Geol, 72: 406-411. Ameghino, F. 1912. L'age des formationes sedimentaires tertiares de l'Argentine, en relation avec l'antiquite de l'homme. An. Mus. Nac. Buenos Aires, 22: 45-75. Anthony, H. E. 1916, Preliminary report of fossil mammals of Porto Rico, Ann. New York Acad. Sci. 27: 193-203. Cooke, C. W. 1945. Geology of Florida. Fla, Geol. Surv. Bull. 29: 398. Cope, E, D. 1871. Preliminary report on Vertebrata discovered in the Port Kennedy bone cave. Proc. Amen Phil. Soc. 12: 78-102. ---. 1898. A preliminary report on the vertebrate paleontology of the Llan o Estacado. 4th Ann. Rept. Texas Geol. Surv., 186 pp. ---. 1899. Vertebrate remains from Port Kennedy bone deposit. J. Acad. Nat. Sci. Phila., ser. 2,11: 198-267. Frick, C. 1921. Extinct vertebrate faunas of the Badlands of Bautista Creek and San Timoteo Canyon, Southern California. Univ. Calif. Publ., Bull, Dept. Geol. 12: 820-321,849-350. Gazin, C. L. 1985. Gravigrade sloth remains from the late Pliocene and Pleis- tocene of Idaho. J. Mamm. 16: 52-56. Hay, 0. P. 1919. Descriptions of some mammalian remains from Florida of probably Pleistocene age. Proc. U. S. Nat. Mus. 46: 105-112. --. 1921. Description of species of Pleistocene vertebrata, types of specimens most of which are preserved in the United States National Museum. Proc. U.S. Nat. Mus. 59: 599-642. ---. 1927. The Pleistocene of the Western region of North America and its vertebrated animals. Cam. Inst. Wash. Publ. 822 b, 846 pp Hoffstetter, R, 1952. Xenarthra. In Traite de Palcontologie, ed. by Jean Pive- teau. Paris: Masson et Cie. Tonie VI., 2: 585-686. Hooijer, D. A. 1962. A fossil ground sloth from Curacao, Netherlands, Antilles. Proc. Koninkl. Ned. Akad. v. Wet. Amsterdam, ser. B. 65: 40-60. --. 1964. The snout of Paulocnus petrifactus C Mammalia Edentata). Zool. Med. Mus. Leiden, 89: 79-84. Johnston, C. S., and D. E. Savage. 1955. A survey of various late Cenozoic vertebrate faunas of the panhandle of Texas. Part I. Univ. of Calif. Publ. in Geol. Sci., 81: 27-59. Kraglievich, L. 1925. Un nuevo eslabon en la serie fieogenetica de la sub- familia Nothrotheriinae: Senetia mirabilis, nuevo genero y especie de la for- macion entrerriana. An. Mus. Nac. Hist. Nat. Buenos Aires, 83: 177-198. Leidy, J. 1855, A memoir on the extinct sloth tribe of North America. Smith- son, Contrib. Knowl. 7: 1-68. ---. 1860. Remarks on the structure of the feet of Megalon!/x Trans. Amer. Phil. Soc. 11: 107-108. --. 1871. Report on the vertebrate fossils of the Territory .formations of the West. U.S. Geol. Surv. Wyoming. 2nd. Ann. Rept. ( Hayden) : 340-870. Lindahl, J. 1892. Description of a skull of Megaton!/x teid!/i . Trans. Amer. Phil. Soc. 17: 1-10. 1968 HIRSCHFELD/WEBB: PLIO-PLEISTOCENE SLOTHS 295 Lyon, Gretchen M. 1988. Megalongx miNeri, a new Pleistocene ground sloth from Southern California. Trans. San Diego Soc. Nat. Hist. 9: 15-80. Marsh, 0. C. 1874. Notice of new Tertiary mammals III. American J. Sci,, 1874: 581-534. Matthew, W. D. 1918. Contributions to the Snake Creek fauna with notes upon the Pleistocene of Western Nebraska. Bull. Amen Mus. Nat. Hist. 88: 188-229. ---. 1924. Third contribution to the Snake Creek fauna. Bull. Amer. Mus. Nat. Hist. 50: 149-150. ---. 1909. (with H. J. Cook). A PIiocene fauna from western Nebraska. Bull. Amer. Mus. Nat. Hist. 26: 861-414. ---. 1959. ( with C. de Paula Couto). The Cuban edentates. Bull. Amer. Mus. Nat. Hist. 117: 1-56. Meade, G. E. 1945. The Blanco fauna. Univ. Texas Bull. 4401: 509-542. Merriam, J. L., C. Stock, and C. L. Moody. 1925. The Pliocene Rattlesnake Formation and fauna of eastern Oregon with notes on the geology of the Rattlesnake and Mascall deposits. Carnegie Inst. of Wash. Publ. 847: 43-92. Owen, R. 1840-1845. Odontography London: Hippolyte Bailliere, Publ. Paulo Couto, Carlos de. 1967. Pleistocene Edentates of the West Indies. Amer. Mus. Novs., 2804: 1-55. Pirkle, E. C. 1956, The Hawthorne and Alachua Formations of Alachua County, Florida. Quart. j. Florida Acad. Sci. 19: 197-240. Piveteau, Jean ( ed.) 1952. Traite de Paleontologie. Tome VI, vol. 2. Paris: Masson et Cie. - Puri, H. S., and R. 0. Vernon. 1964. Summary of the Geology of Florida and a Guidebook to the Classic Exposures. Tallahassee: Fla, Geol. Surv., Spec. Publ. 5: 1-812, Savage, D. E. 1941. Two new Middle Pliocene carnivores from Oklahoma with notes on the Optima fauna. Amer. Midl. Nat. 25: 692-710, Scott, W. B. 1908-1904. Reports of the Princeton University expeditions to Patagonia 1896-1899. In Paleontology, Mammalia of the Santa Cruz Beds, 5: 161-864, Shotwell, J A. 1956. Hemphillian mammalian assemblage from northeastern Oregon. Bull. Geol. Soc. Amer. 67: 717-789. ---. 1958. Inter-community relationships in Hemphillian ( Mid-Pliocene) Mammals. Ecology 39: 271-282. --. 1968. The Juntura Basin: Studies in earth history and paleoecology. Trans. Amer. Phil. Soc. 58: 1-77. Sicher, Harry. 1944. Masticatory apparatus of the sloths. Field Mus. Nat. Hist., Zool. Ser., 29: 161-168. Simpson, George G. 1928. Pleistocene mammals from a cave in Citrus Co., Florida, Amer. Mus. Novitates, 828: 1-16. ---. 1980. Tertiary land mammals of Florida. Bull. Amer. Mul Nat. Hist., 59: 149-211, ---. 1945. The principles of classification and a classification of mammals. Bull, Amer, Mus. Nat. Hist, 85: 69-75. ---. 1948. The beginning of the age of mammals in South America. Bull. Amen Mus. Nat. Hist., 91: 1-232. ---. 1959. Appendix to The Cuban edentates, by W. D. Matthew and C. de 296 BULLETIN FLORIDA STATE MUSEUM Vol. 12 Paulo Couto, Bull. Amer. Mus. Nat. Hist. 117: 51-54 Simpson, G. G., A. Roe and R. C. Lcwontin. 1960. Quantitative Zoology. Har- court Brace, New York. Sinclair, W. J. 1905. New mammalia from the Quaternary caves of California. Univ. Calif. PubI., Bull, Dept. Geol., 4: 145-161. --. 1906. Some edentate-like remains from Mascall beds of Oregon. Univ. Calif. Publ.,Bull. Dept. Geol., 5: 65-66. ---. 1915. Additions to the fauna of the Lower Pliocene Snake Creek beds ( results of the Princeton University 1914 expedition to Nebraska). Proc. Amen Phil. Soc., 54: 78-95. Stirton, R. A. 1986. Succession of North American continental -Pliocene mam- malian faunas. Amen Jour. Sci,, ser. 5, 82: 160-206. ---. 1989. Cenozoic mammal remains from the San Francisco Bay region. Univ. Calif. Publ. Geol. 24: 389-410. Stock, C. 1913. Nothrotherium and Megalonf/x from the Pleistocene of southern California. Univ. Calif. Publ., Bull. Dept. Geol. 7: 841-858. ---. 1925. Cenozoic gravigrade edentates of western North America. Publ. Carnegie Inst. Wash. 831, 206 pp ---. 1942, A ground sloth in Alaska. Science 95: 552«558. Stovall, J. W. 1940. Megalon!/x hokani, a new species of ground sloth from Could, Oklahoma. Amer: Jour. Sci., 238: 140-146. ---. and W. N. McAnulty. 1950. The vertebrate fauna and geologic age of Trinity River terraces in Henderson County, Texas. Amer. Midi. Nat., 44: 222-250. Vernon, R. 0. 1951. Geology of Citrus and Levy Counties, Florida. Fla. Geol. Surv. Bull., 83: 1-186. Webb, S. D. 1964. The Alachua Formation, Guidebook 1964 Field Trip, Soc. Vert. Paleo., Central Florida, Univ, Fla. and Fla, Geol. Sum: 22-29. Contributions{ to .the BuLIffIN OF TifE. FLORiDA STATE MUSEUM may be in any Beld of biology. Mapusclipti de~link With natifal history or-systematic problems involying the southeastern United States dr th* Caril;bean area afe solicited especially.. Manuscripts should be of medium length-50 to 200 pages. Exam*Lation for suitability is made by an Editorial Board. The BuLLE*IN is distributed worldwide ,through instituti6nal subscriptions and exchanges only. It is considered. the respohsibility of the, author to distribute, his paper to all interested individuals. To aid in this, fifty copies afe furnished the author without cost, PREPARATION OF MANUSCRIPT Hi*hly recommended a5 a guide is the volume: Conference of Biological Editors, Committee on Forin and Style. 1960. Style manualf for biological journals. Amer. Inst. Bi61. Sci., Washington. 92 p. Manuscripts should be typewritten with double spacing throughout, with ample margins, and on only one side of the paper. The author should keep a copy; the original and a carbon. must be submitted. Tables and legends of figures should be typed on sheets separate from: the text. Several legends or tables may be placed[ on a single sheet. Illustrations, including maps and photographs, should be referred to as "figures." All illustrations are redueed to a makinium of 4-1/4 by 7-1/8 inches. Size scales, wherever they are necessary, should be incorporated into the figure. References to literature should conform with the, bibliographic style used in fecent numbers of the BULLETIN. Spell out in full the titles of non-English serials and places of publication. - Footnote material should be kept to a minimum. However, provide. copy for a footnote detailing the title, affiliations, and address of the author (see recent numbers of the BULLETIN). Manuscripts must be accompanied by a syn6psis-a brief and factual summary (not a 'mere description) of the contents and conclusions, which points out the presence Of any new information and indicates its relevance. In it list all new organisms described and give their rahges; indicate all taxonomic changes pro- posed. The synopsis, written in full sentences, should be concise, but completely intelligible in itself without references to the paper, tbereby e*abling the busy reader to decide more surely than he can from the title alone whether the paper merits his reading. The· syhtlpsis will be published with the paper. It does not replace the usual conclusions or summary sections. It may alsd serve as copy for the ab5tracting services. Manuscripts and all editorial matters should be addressed to: Managing Editor Of the BOLLETIN Florida State Museum Seagle Building Gainesville, Fldrida