CENOZOIC VERTEBRATE BIOSTRATIGRAPHY OF SOUTH CAROLINA, U.S.A., AND ADDITIONS TO THE FAUNA L. Barry Albright III, Albert E. Sanders, Robert E. Weems, David J. Cicimurri, and James L. Knight Vol. 57, No. 2, pp. 77–236 October 31, 2019 ISSN 2373-9991 UNIVERSITY OF FLORIDA GAINESVILLE BULLETIN of the Florida Museum of Natural History The FLORIDA MUSEUM OF NATURAL HISTORY is Florida’s state museum of natural history, dedicated to understanding, preserving, and interpreting biological diversity and cultural heritage. The BULLETIN OF THE FLORIDA MUSEUM OF NATURAL HISTORY is an on-line, open-ac- cess, peer-reviewed journal that publishes results of original research in zoology, botany, paleontology, archaeology, and museum science. New issues of the Bulletin are published at irregular intervals, and volumes are not necessarily completed in any one year. Volumes contain between 150 and 300 pages, sometimes more. The number of papers contained in each volume varies, depending upon the number of pages in each paper, but four numbers is the current standard. Multi-author issues of related papers have been published together, and inquiries about putting together such isues are welcomed. Address all inqui- ries to the Editor of the Bulletin. Richard C. Hulbert Jr., Editor Bulletin Committee Richard C. Hulbert Jr. Jacqueline Miller Larry M. Page David W. Steadman Roger W. Portell, Treasurer Jonathan I. Bloch, Ex officio Member ISSN: 2373-9991 Copyright © 2019 by the Florida Museum of Natural History, University of Florida. All rights reserved. Text, images and other media are for nonprofit, educational, and personal use of students, scholars, and the public. Any commercial use or republication by printed or electronic media is strictly prohibited without written permission of the museum. Publication Date: October 31, 2019 This and other recent issues of the Bulletin can be freely downloaded at: https://www.floridamuseum.ufl.edu/bulletin/publications/ Send communications about this publication to: Editor of the Bulletin; Florida Museum of Natural History; University of Florida; P.O. Box 117800; Gainesville, FL 32611-7800 USA FAX: 352-846-0287; Email: bulletin@flmnh.ufl.edu URL: https://www.floridamuseum.ufl.edu/bulletin/home/ Cover image. View of the Charleston Museum’s 1971 Chandler Bridge excavation pit (see p. 115). mailto:bulletin@flmnh.ufl.edu IN MEMORIUM ALBERT E. SANDERS (1934-2019) Just as this volume was being prepared for press, we learned that Albert E. Sanders passed away peacefully in his sleep the morning of 15 October 2019. I first met Al upon his arrival at the Charleston Museum as its new Curator of Natural Sciences in 1968. I was 11 years old at the time, and was involved in a kids group that met at the Museum every Saturday morning. The group, called The Nature Trailers, was active for nearly 40 years and served youngsters from 9 to 13 years old who had an interest in natural history and the outdoors. My fortuitous membership in this group over the same time that Al was hired and was establishing his program quite literally set the course of my career. Noticing my interest in natural history in general, and paleontology in particular, Al tucked me under his wing, further nurtured my interest in paleontology, and this re- sulted in an association that lasted over 50 years. One of the pivotal moments of my life occurred when I received a phone call from Al one summer day in 1973 asking if I’d be interested in joining a team that would search for and excavate archaeocete whales from one of the Eocene limestone quarries in South Carolina. It was as if I had won the lottery. Although known primarily for his expertise on fossil whales, particularly early odontocetes and mysticetes, Al was also what we would consider today an “old style” true naturalist. A keen observer and a careful, meticulous scientist, he knew the natural history of the South Carolina Lowcountry as few others do or ever have - the birds, the reptiles and amphibians, the plants, and more – and he was an expert on the history of SC natural history, as well. He was a remarkably gentle and soft-spoken man. It seemed that every sentence he spoke was well thought-out; he was never one to “shoot-from-the-hip.” He appeared to be always cheerful; stress apparently was not a factor in his life; he never seemed to be in a rush. In fact, Al had to “talk me off the ledge” a few times when I thought this current volume was falling too far behind the schedule I had envisioned for it. “Behind what schedule” was Al’s attitude. And he was right! Perhaps more important than Al’s careful, insightful approach to his science, was his ap- proach to life. He never forgot, nor did he let us forget—as scientists, paleontologists, natural- ists—just how exceedingly fortunate we are. He was so right about this. What we do is special and it’s wonderful and it makes for such an interesting life; as such, we never really experience the drudgery of “a job”! I could see from what would become my last visit with Al in early July, 2019, that his health was failing; but I was so hoping that he could hold on long enough to see the publication of this volume, as it represents the culmination of over ten years of work among its authors. But I had a sinking feeling it would be close. Close it was, as I received notification of his death from his wife Randa the very day before editor Richard Hulbert informed me that the page proofs were ready. Al did know, however, that this volume was “in the can,” and for that we are grateful. Al’s quirky wit, his friendship, and his life lessons will be deeply and affectionately missed. Barry Albright Albert E. Sanders (center) in November 2014 at the Mace Brown Museum of Natural History at the Col- lege of Charleston with two of his protégés, Barry Albright (left) and Jonathan Geisler (right). CENOZOIC VERTEBRATE BIOSTRATIGRAPHY OF SOUTH CAROLINA, U.S.A., AND ADDITIONS TO THE FAUNA L. Barry Albright III1, Albert E. Sanders2, Robert E. Weems3, David J. Cicimurri4, and James L. Knight5 1University of North Florida, Department of Physics, 1 UNF Drive, Jacksonville, Florida 32224 USA 2deceased 3Calvert Marine Museum, Research Associate, P.O. Box 97, Solomons, Maryland 20688 USA 4South Carolina State Museum, Department of Natural History, 301 Gervais St., Columbia, South Carolina 29214 USA 5University of South Carolina, Aiken, Department of Biology and Geology, Aiken, South Carolina 29801 USA Albright III, L. B., A. E. Sanders, R. E. Weems, D. J. Cicimurri, and J. L. Knight. 2019. Cenozoic vertebrate biostratigraphy of South Carolina, U.S.A., and additions to the fauna. Bulletin of the Florida Museum of Natural History 57(2):77–236. ABSTRACT Study of vertebrate fossils from the South Carolina Coastal Plain played a significant role in the early history of vertebrate paleontology as a scientific discipline in North America. However, a clear under- standing of the state’s vertebrate biostratigraphy has been greatly hindered by the paucity of well-exposed fossil-bearing stratigraphic sections and a complicated subsurface stratigraphy. Most units, particularly those of Neogene and Quaternary age, exist as thin veneers of marine or estuarine sediments that typically occur as infilled topographic lows or erosional remnants as determined primarily from borehole stratig- raphy. Hence, lateral continuity can be difficult to confirm over broad geographic areas often resulting in confusion insofar as vertebrate fossil provenance is concerned. The evaluation of vertebrate fossils from the South Carolina Coastal Plain presented here, and of the geologic units from which they originated, or are thought to have originated, results in the first modern biostratigraphic framework for the known Cenozoic record of the state. Results provide (1) updated correlations of many units to the most recent, astronomically-tuned marine oxygen isotope stages; (2) the first viverravid from the Atlantic Coastal Plain, Didymictis proteus; (3) new information on the timing of certain Neotropical immigrants into the Southeast during the Great American Biotic Interchange, including the oldest record of Erethizon in this region and the oldest records of capybara in the USA; (4) the possible oldest record of the microtine rodent Allophaiomys pliocaenicus in the USA; (5) new details on the unit of origin for the type specimen of the archaeocete Dorudon serratus; and (6) new details on the ontogeny of the early odontocete Agorophius pygmaeus. New taxonomic records from South Carolina include Glyptotherium texanum, Holmesina floridanus, Ondatra idahoensis, Erethizon ?bathygnathum, Leopardus amnicola, Miracinonyx ?trumani, Canis lepophagus, Canis latrans, Phocanella pumila, Callophoca obscura, Monatherium sp., Anchip- pus texanus, Subhyracodon mitis, Aphelops ?malacorhinus, Teleoceras ?guymonense, and Perchoerus sp. New specimens of previously recorded taxa are also noted. Key words: South Carolina, vertebrate paleontology, stratigraphy, biostratigraphy. 80 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) INTRODUCTION Until recently, South Carolina has not often been considered a state yielding a wealth of vertebrate paleontological resources, although the study of its fossils, particularly those from the famous “Ashley River phosphate beds” near Charleston, played an major role in the early history of vertebrate pale- ontology in North America. Louis Agassiz, Joseph Leidy, and other notable 19th century naturalists recognized the importance of fossils from the state, and collections they studied can be found in such venerable institutions as the American Museum of Natural History, the Academy of Natural Sci- ences of Drexel University (formerly the Academy of Natural Sciences of Philadelphia), Harvard’s Museum of Comparative Zoology, and the United States National Museum of Natural History, in addi- tion, of course, to the oldest established museum in the USA, The Charleston Museum.1 Missing up to this time, however, has been a clear understanding 1This point is somewhat arguable. Simpson (1942:158) con- cluded that “the first American natural history museum defi- nitely organized as such, public and independent, was ap- parently Peale’s Philadelphia Museum, in which vertebrate paleontology and the American Philosophical Society played the leading parts.” He noted that Peale’s museum was a “cabinet” of “going concern in 1770,” thereby predating The Charleston Museum by three years. and representation of the stratigraphic framework for fossil vertebrates from the state, primarily due to the reasons discussed below. As part of the southeastern USA Atlantic Coastal Plain (Fig. 1), much of the state is low lying with little topographic relief, and its warm temperate climate results in a thick vegetative cover – both factors of which severely limit expo- sure of fossil-bearing strata. Another factor to con- sider is the complicated nature of South Carolina’s subsurface geology. As generalized by Katuna et al. (1997:182), “the coastal plain of SC is underlain by a thick seaward-dipping wedge of late Creta- ceous to Holocene siliciclastic and carbonate strata deposited on Paleozoic and Mesozoic rocks along the eastern, passive continental margin of North America” (also see Gohn, 1988; Idris and Henry, 1995). Additionally, as noted by Ward et al. (1991), the deposition and geographic distribution of strata within the lower coastal plain of SC has been influ- enced primarily by two structurally positive areas, the Cape Fear Arch to the north and the Yamac- raw (aka Beaufort Arch) to the south, between which formed a coastal embayment known as the Charleston Embayment (Fig. 1A). Additional con- trols on deposition/erosion in this region were those imposed by currents of the Gulf Trough and the TABLE OF CONTENTS Introduction .............................................................................................................80 Materials and Methods ............................................................................................87 A Note on “Hobby Collecting” Vertebrate Fossils in South Carolina ....................91 Historical Background ............................................................................................93 Cenozoic Vertebrate Fossil-bearing Beds of the South Carolina Coastal Plain ......94 Additions to the Cenozoic Mammalian Fauna of South Carolina ........................153 Summary and Conclusions ...................................................................................194 Acknowledgements ...............................................................................................195 Literature Cited .....................................................................................................196 Appendix 1. All Known Non-marine Fossil Mammals from South Carolina ......225 Appendix 2. All Known Marine Fossil Mammals from South Carolina ..............229 Appendix 3. Cenozoic Non-Marine Mammalian Faunas and Fossil-bearing Strata of South Carolina ..................................................................................232 Appendix 4. Molluscan Fauna of the Ashley Formation ......................................234 ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 81 ancestral Gulf Stream during the middle Eocene to early Oligocene (Idris and Henry, 1995; Pope- noe et al., 1987). Several units deposited in this embayment, particularly those of Neogene age, have been partly to completely removed by erosion (multiple transgressions plus dissection during low stands) leaving, in many cases, only thin veneers of marine, estuarine, or (rarely) fluvial sediments that typically occur as infilled topographic lows or ero- sional remnants seen only in cores, auger cuttings, very limited outcrops, or seismic reflection and/ or resistivity profiles (e.g., Popenoe et al., 1987; Krantz, 1991; Idris and Henry, 1995; Weems and Lewis, 2002; Putney et al., 2004; Williams et al., 2009). As noted by Cronin et al. (1984:26), further limitations arise from the deposition of the same group of lithofacies (“dunes, beach sands, back- barrier muds to shelf sands”) over the course of many marine transgressions “… so that distinguish- ing the same facies from two separate transgres- sions on lithologic grounds is difficult unless inde- pendent dating of the marine facies is available.” Correlation of these units is therefore difficult over broad geographic areas and typically relies on mol- lusc and/or microfossil biostratigraphy, with data provided by ostracods, dinoflagellates, radiolari- ans, foraminiferans, and/or calcareous nannoplank- ton (e.g., Abbott and Andrews, 1979; Weems et al., 1982; Bybell, 1990; Ward et al., 1991; Campbell and Campbell, 1995; de Verteuil and Norris, 1996; Ward, 2008). Still further complications arise from the mixing of fossils of very different ages in lag deposits found at the bases of many units, again due to the numerous transgressions and regressions to which the region has been subjected. These limi- tations result in particular confusion insofar as ter- restrial (and sometimes marine) vertebrate fossil provenance is concerned. Even when provenance can be determined with a high degree of confidence, an additional problem lies in providing a refined age estimate for the unit of origin. Magnetostratigraphic anal- ysis of formations across the SC Coastal Plain is of only limited utility due to the absence of thick, exposed stratigraphic sections, which might pro- vide a discernable polarity reversal pattern that can be correlated to the global Geomagnetic Polarity Time Scale. The nonexistence of radioisotopically dateable volcanic horizons additionally precludes calibration to the time scale of sites of determin- able magnetic polarity. Furthermore, temporally calibrated range zones (plus FADs and LADs) for the biochronologically significant invertebrate groups noted above are constantly being refined as new sections and/or cores are studied, and these changes often outpace the boundaries of these zones as reflected in the most recently published geologic time scales. Thus, for an undertaking such as this one to be relevant at the time of its publi- cation depends heavily on familiarization with the most recent literature and/or results from a variety of geological subdisciplines including paleocean- ography, radioisotopic and geochemical geochro- nology, and invertebrate biostratigraphy/biochro- nology. However, despite the above noted short- comings, some geologic units are relatively well characterized from a paleontological, and to some extent a temporal, perspective. Examples include the Eocene Santee Limestone, Tupelo Bay, and Harleyville formations, plus the Oligocene Ashley and Chandler Bridge formations, with their spec- tacular records of fossil cetaceans. Pre-Pleistocene terrestrial strata, on the other hand, are essentially non-existent, and as Weems and Lewis (2002) noted, all of the middle Eocene through Pliocene lithologies underlying the Charleston area (and the SC Coastal Plain in general; Figure 2A–C) formed in marine (including coastal) depositional envi- ronments. Although rare and isolated land mam- mal fossils of late Arikareean North American Land Mammal Age (NALMA) provide a limited glimpse into the latest Oligocene-earliest Miocene terrestrial record, and obviously originated from sediments deposited in a coastal plain/fluvial set- ting, they nevertheless were recovered from much more recent marine to marginal marine sediments, having been reworked from subjacent strata. The Miocene record is particularly sparse, represented by only five formations broadly spaced across the 17.5 myr span of that series (Fig. 2B). The Pliocene is somewhat better represented, with mammal fos- 82 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) sils known from the Goose Creek Limestone and the Raysor and Duplin formations (Fig. 2C). Fos- sils from these units are providing important new information on the timing of Neotropical immi- grants, such as capybara, glyptodonts, megalony- chid sloths, and porcupines into North America during the Great American Biotic Interchange (e.g., Woodburne, 2010). Pleistocene mammals are relatively well-represented from units such as the Waccamaw, Ladson, and Wando formations, plus the submerged unnamed and geologically unchar- acterized units off Edisto and Myrtle beaches (Fig. 1A). The oldest well-dated remains of Bison in the lower 48 states were found in the upper Pleistocene Ten Mile Hill Formation, thus providing important implications for the boundary between the Irving- tonian and Rancholabrean NALMAs (Sanders et al., 2009). The primary purpose of this paper, therefore, is to provide a much needed modern chronostratigraphic and biostratigraphic foundation for the fossil vertebrates (mainly mammals) found in South Carolina. We have attempted to correlate all of the state’s lower coastal plain units with the most recent astronomically tuned marine oxygen isotope stages (i.e., those of Lisieki and Raymo, 2005; Raffi et al., 2006; Cramer et al., 2009; and Wade et al., 2011) in order to provide the highest temporal resolution currently available for the fossils found within them (Figures 3A–C). This work is not intended as an exhaustive review of all of South Carolina’s Cenozoic vertebrate fossils, although additions to the state’s fossil fauna are included, along with an updated faunal list, in Appendices 1 and 2. Appendix 1 provides a list of all known non-marine mammals from SC and the faunas and/or stratigraphic units from which they are known, plus their ages. Appendix 2 provides the same information for the known marine mammal record. Appendix 3 is provided as a quick reference list of the faunas and/or stratigraphic units from which the non-marine taxa are known, plus their ages, but also includes comparably aged faunas and/or sites in Florida that are frequently noted throughout the text. For relatively recent discussions of fossil vertebrates from South Carolina, the reader is referred to Roth and Laerm (1980), Bentley et al. (1994), Sanders (1998a), Sanders (2002), Geisler et al. (2005, 2014, 2017, 2018), Kohn et al. (2005), Chandler and Knight (2009), Fierstine and Weems (2009), Sanders et al. (2009), Cicimurri and Knight (2009a, b, c; 2019), Knight and Cicimurri (2010), Fields et al. (2012), Weems and Knight (2013), Ksepka (2014), Weems and Sanders (2014), Vélez- Juarbe and Domning (2014a, b), Cicimurri et al. (2016), Weems and Brown (2017), Babiarz et al. (2018), Boessenecker et al. (2018), Domning and Beatty (2019), and others mentioned throughout the text. Important older references include Leidy (1859, 1860, 1869, 1876a, b, c, 1877, 1890), Hay (1923), and Allen (1926). Sloan (1908) is valuable resource for older South Carolina geological literature, as is Cooke (1936). Figure 1. A, Index map of the South Carolina (SC) Coastal Plain and B, Charleston-Berkeley-Dorchester tri-county region, showing counties, towns, and major localities noted in text. A, general area where Ashley River Phosphate Beds were mined near Middleton Place; AQ, Argos Cement Quarry (formerly called the LaFarge, Blue Circle, or Gifford-Hill & Co. Harleyville quarry); C, Cross, SC; CCC1, Clubhouse Crossroads Core 1; CQ, Cross quarries; CR, Cooper River; DL, Dawson’s Landing; EB, Edisto Beach; ES, Eutaw Springs; F, Florence, SC; GC, Goose Creek, SC; GF, Givhans Ferry State Park; GQ, Giant Cement Quarry; H, Harleyville, SC; HC, Huspa Creek; HH, Holly Hill, SC; J, Jamestown, SC; JQ, Jamestown Quarry; K, Kingstree, SC; LM, Lake Moultrie; MB, Myrtle Beach; MC, Monks Corner, SC; PC, Pregnall Core (DOR-208); PL, Porters Landing, Georgia; S, Summerville, SC; and StS, St. Stephen Pit. Note: often mentioned in the text, the Camelot, Crowfield, and Walrus Ditch sites are located near the Giant Cement quarry, Goose Creek, and Summerville, respectively. Dashed line = approximate location of major structural features (Beaufort Arch, Charleston Embayment, and Cape Fear Arch) of SC Coastal Plain after Ward et al. (1991). ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 83 84 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) Figure 2A. Caption is located on pp. 86–87. ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 85 Figure 2B. Caption is located on pp. 86–87. 86 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) Figure 2. Generalized stratigraphic charts for named units of the South Carolina Coastal Plain. Geochro- nologic framework based primarily on GTS2012; boundaries of calcareous nannoplankton and forami- niferal zones follow Lourens et al. (2004), Mudelsee and Raymo (2005), Raffi et al. (2006), Anthonissen (2008), Coccioni et al. (2008), and Wade et al. (2011); NALMA boundaries follow Woodburne (2004), Flynn et al. (2005), Morgan (2005), Secord et al. (2006), Albright et al. [caption continues on next page] Figure 2C. ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 87 METHODS AND MATERIALS Institutional Abbreviations AMNH, American Museum of Natural His- tory, New York; ANSP, Academy of Natural Sci- ences of Drexel University, Philadelphia, Pennsyl- vania (formerly the Academy of Natural Sciences of Philadelphia); CCNHM, College of Charleston Natural History Museum (recently renamed the Mace Brown Museum of Natural History), Charles- ton, SC; ChM PV, vertebrate paleontology collec- tions at The Charleston Museum, Charleston, SC; CMM-V, vertebrate paleontology collections at the Calvert Marine Museum, Solomons, Maryland; FLMNH, Florida Museum of Natural History, University of Florida, Gainesville; GSM, Georgia Southern Museum, Georgia Southern University, Statesboro; McK, McKissick Museum, University of South Carolina, Columbia; MCZ, Museum of Comparative Zoology, Harvard University, Cam- bridge, Massachusetts; SCSM, South Carolina State Museum, Columbia (note: only “SC” is used as the prefix in catalogue numbers for specimens at the SCSM); UF, vertebrate paleontology collection of the FLMNH; USGS, United States Geological Survey; USNM, United States National Museum of Natural History, Smithsonian Institution, Wash- ington, D.C. Anatomical Abbreviations AP and TR refer to antero-posterior (length) and transverse (width) measurements, respectively; P or M, upper premolars and molars, respectively; p or m, lower premolars and molars, respectively. Figure 2. Continued. (2008), Sanders et al. (2009), Tsukui and Clyde (2012), May et al. (2014), and Mur- phey et al. (2018).A, Paleogene stratigraphic units; chronostratigraphic placement of formations based on Nystrom et al. (1991), Fallaw and Price (1995), Edwards et al. (2000), and Weems et al. (2016); boundar- ies of chrons C21–C23 from Tsukui and Clyde (2012); inclusion of “Jamestown beds” in Chicora Mem- ber of Williamsburg Formation from Cicimurri et al. (2016). B, Miocene stratigraphic units; DN zones based on recalibration of de Verteuil and Norris (1996) using Dybkjaer and Piasecki (2008); boundaries of chrons C4Ar–C5r from Evans et al. (2007); placement of Upland unit in upper Serravallian follows Nystrom et al. (1991), Huddlestun (1988), and Weems and Edwards (2007a). C, Pliocene and Pleistocene stratigraphic units; Pliocene/Pleistocene boundary follows Gibbard et al. (2010); Reunion I and II sub- chrons follow Kidane et al. (2007); Blancan-Irvingtonian boundary follows Morgan (2005); Hemphillian- Blancan boundary follows May et al. (2014); short dashed lines in PLEISTOCENE column separate Lower, Middle, and Upper Pleistocene; see text for further discussion. RLB = Rancholabrean NALMA. Figure 3. Stratigraphic units of the South Carolina Coastal Plain with correlation to oxygen isotope curves; geochronologic/chronostratigraphic base as in Figure 2. A, Paleogene units correlated to general- ized oxygen isotope curve of GTS2012, figure 28.11 (derived from Raffi et al. [2006] and Cramer et al. [2009]); EECO = Early Eocene Climatic Optimum; MECO = Middle Eocene Climatic Optimum; PETM = Paleocene-Eocene Thermal Maximum. B, Miocene units correlated to generalized oxygen isotope curve of Raffi et al. (2006); DN zones based on recalibration of de Verteuil and Norris (1996) using Dybkjaer and Piasecki (2008); MMCO = Middle Miocene Climatic Optimum. C, Pliocene and Pleistocene units correlated to generalized oxygen isotope curves of Lisiecki and Raymo (2005) and Raffi et al. (2006); nannoplankton and foraminiferal boundaries follow Lourens et al. (2004), Mudelsee and Raymo (2005), Raffi et al. (2006), Anthonissen (2008), and Wade et al. (2011); Reunion I and II subchrons follow Kidane et al. (2007); Pliocene-Pleistocene boundary follows Gibbard et al. (2009); Blancan-Irvingtonian Bound- ary follows Morgan (2005); Hemphillian-Blancan boundary follows May et al. (2014); Bridgerian-Uintan boundary follows Murphey et al. (2018); IRD = Ice Rafted Debris; LMGI = Late Miocene Glacial Inter- val; MPWP (Middle Pliocene Warm Period) from Dowsett et al. (2005) and Robinson et al. (2008); RLB = Rancholabrean NALMA. [images follow on pp. 88–90] 88 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) Figure 3A. Caption is located on pp. 87. ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 89 Figure 3B. Caption is located on pp. 87. 90 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) Figure 3C. Caption is located on pp. 87. ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 91 Geochronology/Chronostratigraphy The version of the global Geomagnetic Polarity Time Scale (GPTS) used for the tem- poral framework in this contribution is primarily GTS2012, i.e., that of Gradstein et al. (2012). Par- ticularly germane to our work are those portions of GTS2012 compiled by Vandenberghe et al. (2012) for the Paleogene and by Hilgen et al. (2012) for the Neogene (the latter similar to Lourens et al., 2004, in GTS2004), although the recently re-rat- ified Pliocene-Pleistocene boundary follows Gib- bard et al. (2010) in its placement at the base of the Gelasian Stage at marine isotope stage (MIS) 103 at about 2.59 Ma. The terms “Fauna” and “Local Fauna” (LF) follow Tedford (1970) and definitions in Woodburne (1987). Boundaries of North Ameri- can Land Mammal Ages (NALMAs) follow those in Woodburne (2004) with the following excep- tions: divisions of the Arikareean follow Albright et al. (2008); Hemphillian-Blancan boundary follows May et al. (2014); Blancan-Irvingtonian boundary follows Morgan (2005); and Irvingtonian-Rancho- labrean boundary follows Sanders et al. (2009). For the last glacial stage in North America, “Wiscon- sian” is used in favor of “Wisconsinan” following Pillans and Gibbard (2012). FAD, first appearance datum; LAD, last appearance datum; HO, highest stratigraphic occurrence; LO, lowest stratigraphic occurrence; Ma, mega-annum (million years), a radioisotopically calibrated numerical age; myr, millions of years, elapsed time or duration; kyr, thousands of years, elapsed time or duration. A NOTE ON “HOBBY COLLECTING” VER- TEBRATE FOSSILS IN SOUTH CAROLINA Over the course of conducting research for this project, it came to the attention of the authors that the “hobby collecting” of vertebrate fossils from SC is, in a word, thriving, although sadly to differ- ent ends. There are substantial advantages to these activities, as hobby collectors often have the time and resources required to prospect for new sites and specimens; but there is an alarming downside, as well. The trend started primarily in the 1970s as avocational fossil collectors began finding mainly sharks teeth and remains of early cetaceans in drainage ditches cut through what were then new residential neighborhoods, primarily in the Sum- merville area, which exposed the highly fossilifer- ous Oligocene-aged Ashley and Chandler Bridge formations. The trend gained additional traction as land was cleared and excavated for new highways and construction projects in the North Charles- ton and Summerville areas throughout the 1980s, in turn creating further, more easily accessible exposures of these formations. Although recently emplaced local ordinances, primarily in Dorchester County, together with the dense jungle-like growth of vegetation along the ditches, have hindered col- lecting activities, it is known as a matter of fact that fossil collecting in these settings continues today. “Hobby collecting” of SC’s fossil resources was significantly compounded by the explosion of scuba diving activities in the coastal rivers as word spread among divers that these rivers were essen- tially littered with fossils, as well as archaeologi- cal artifacts. As a member of the “second wave” of river divers in the state in the early 1970s (see below), the first author can attest to the quantity and quality of paleontological and archaeological resources that were present in the rivers then. In fact, the first author’s pursuit of vertebrate paleon- tology as a career was driven in part by his under- water explorations of the Cooper and Edisto rivers during those times. (Note: the true “first wave” of SC river divers included Drew Ruddy, Jim Batey, Steve Howard, and William Hunt of Charleston. Ruddy and Hunt were older neighbors of the first author’s, who, as a child, was fascinated by their “Sea Hunt”-like exploits; see Ruddy, 2013). The collection of paleontological and archae- ological resources from the bottoms of rivers meandering through the SC Coastal Plain began in the mid-1960s to early 1970s by only a very few adventurous divers who decided to explore the inky depths. But by the late 1970s, diving for fossils and artifacts began to rise in popularity due in large part to the early expansion of recreational scuba diving. Scuba class checkout-dives were often conducted in the regional tanic acid-stained “black water” rivers as an alternative to diving in murky offshore waters or driving several hours to 92 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) Florida’s clear water springs, sinkholes, or keys. As soon as this practice began, divers started bring- ing up 18th century black glass bottles, Native American artifacts, and what quickly became the most sought after prize of all, giant teeth of the late Miocene shark, Megaselachus megalodon (= Carcharodon megalodon, Carcharocles megalo- don, Procarcharodon megalodon, Otodus megalo- don, etc.), along with numerous other fossils. By the early 1980s, an increasing popularity in scuba diving led to an explosion of “hobby-divers” who were quite literally scouring the bottoms of nearly all the state’s coastal plain rivers. This practice was eventually regulated by the State of South Carolina because of the wealth of material, both paleonto- logical and archaeological, being recovered that was of significant cultural, historic, and scientific importance. The state had no way of knowing what was being recovered and removed from its waters without imposing certain sanctions on this growing population of enthusiasts. Some method by which an account of the resources being found needed to be maintained. Therefore, to collect any artifact or fossil from the state’s waters, divers were required to apply for a “hobby diving” permit. All that was (and still is) required by the diver was the submis- sion of a quarterly report citing the locality of their dives and the material collected. There is no doubt that exceptional and important specimens were never accounted for in these reports for fear that the state would confiscate the items – an action which prior to 2014 never occurred. Fortunately, on the other hand, the SC State Museum, The Charleston Museum, and the College of Charleston’s Mace Brown Museum of Natural History have greatly benefitted from the relatively limited cohort of “hobbyists” who, in understanding the scientific importance of their finds, have donated material to these institutions. The downside, however, has been the com- mercial exploitation of SC’s fossil resources. Starting in the 1990s and continuing at an ever increasing pace through today, a growing web- based fossil market has resulted in many spectacu- lar specimens of what are often undescribed and/or very poorly known taxa effectively being removed from scientific access – the fossils have become “trophies” that now have a bounty ascribed to them rather than specimens of scientific value that can be studied then placed on exhibit for all to share. Par- ticularly frustrating is the sale of specimens known to be new to science, but which cannot be made known to the scientific community, or the public, because of their removal from scientific/public access. Specimens held in private collections can- not be technically described (i.e., published in a peer-reviewed scientific journal); for that to occur the specimens must be curated in an accredited institution dedicated to their storage and safety, and where they can be made available for study in perpetuity by future generations of students and scientists. Exceptions can be made if it is under- stood that a private collection may be bequeathed to an institution in the future, or if the specimen is so important that it should at least be mentioned in a technical publication so as to provide a writ- ten record (as occurs in this report). However, the future of specimens held in private collections is never certain. As an example, it is known with certainty that virtually complete skulls of xenorophid ceta- ceans from South Carolina, some of the rarest and most important members of the early odontocete radiation, are currently on display in a Japanese museum, and that they were purchased by that museum, or a representative thereof, from a SC fossil dealer. Another example is a nearly complete skull of what appears to be a new, undescribed spe- cies of Eosqualodon currently on its way (as this report is being finalized) to be offered for sale at the 2019 Tucson Gem and Mineral show. Why are they not in one of South Carolina’s museums where they could be studied in the context of all other cetacean material from the region? Several reasons account for this problem. First, most muse- ums do not have the discretionary funds required to purchase such specimens. Second, most museums do not subscribe to the practice of purchasing spec- imens due to the “bounty” noted above that this places on the fossils – i.e., it immediately places a somewhat randomly assessed commercial value on an otherwise scientifically important specimen, in turn encouraging further collecting for purely com- mercial rather than scientific reasons. ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 93 Following Section 6 of the Society of Ver- tebrate Paleontology’s Code of Ethics, “the barter, sale, or purchase of scientifically significant verte- brate fossils [italics ours] is not condoned, unless it brings them into, or keeps them within, a public trust. Any other trade or commerce in scientifically significant vertebrate fossils is inconsistent with the foregoing [code of ethics] in that it deprives both the public and professionals of important speci- mens, which are part of our natural heritage.” This prompts us to add a quote from the eminent 19th century paleontologist Joseph Leidy (1877:209): “The finder and unscientific owner of fossils, igno- rant of their real importance, often retain them as curiosities, with exaggerated notions of their pecu- niary value, and no argument is sufficient to induce them to part with the specimen or place them where they may be accessible to the student.” Very fortunately, on the other hand, and as noted above, many of SC’s fossil hobbyists have made exceptionally important contributions to our understanding of the state’s paleontological resources. These individuals, realizing the scien- tific importance of so many of their finds, have very generously donated their specimens to both the Charleston and South Carolina State muse- ums, as well as to the Mace Brown Museum of Natural History, understanding that the academic study, technical description, and publication of those specimens in scientific journals is in some cases prohibited unless the specimens are curated in public repositories (accredited museums or uni- versities) where they can be accessed by students and scientists (and the public through exhibits) in perpetuity. In fact, much of what we know about vertebrate paleontology in SC today is a direct result of their dedication, hard work, and generos- ity, and we hereby acknowledge and thank those who subscribe to this view (and list those of whom we are aware in the “Acknowledgments” section at the end of this report). It should also be noted that there are some local fossil enthusiasts who purchase specimens from dealers in an attempt to keep SC’s fossil resources in the state. Some are purchased and then donated to academic institu- tions, while others are purchased and maintained in private collections. Although we do not condone the practice of purchasing fossils due to the rea- sons pointed out above, it is comforting to know that many of these purchased specimens, often of exceptional importance, have been deposited into museum collections where they can be studied and held in perpetuity. Whereas those held in private collections are not in a condition immediately con- ducive to detailed scientific study, some owners do allow access to their material, and we sincerely hope that these private collections will eventually be donated to the state’s museums or academic institutions for study and perpetual curation. HISTORICAL BACKGROUND In the period between 1750 and 1850, three great centers of learning existed in the colonies of what is now the United States of America: Boston in Massachusetts Colony; Philadelphia, in Pennsylva- nia; and Charleston, in the colony of South Caro- lina. In 1773 members of a special committee of the Charles Town Library Society, several of whom were educated in England or Scotland, had the fore- sight to propose the establishment of a natural his- tory museum, likely inspired by visits to the British Museum (Sanders and Anderson, 1999). Thus was born The Charleston Museum. Unbeknownst to, or under-appreciated by, many of today’s paleontolo- gists, however, The Charleston Museum played a pioneering role in the development of paleontology as a scientific discipline in this country. Contribu- tions to (and from) the collection have been made by some of the very founders of the field. Lardner Vanuxem conducted the first geo- logical survey of South Carolina in 1824, but lim- ited funds restricted his work to five districts in the Piedmont region (Vanuxem, 1826). Later, in 1843, Edmund Ruffin published an agricultural survey of the state, which also included descriptions of the state’s then known geological units, as well as a list of invertebrate fossils. It was the geologist Michael Tuomey, however, who, after his 1848 geological survey of South Carolina, stimulated interest in the paleontology of the state (Sanders and Anderson, 1999). Another pioneer was the Charleston native Robert Wilson Gibbes who first published paleonto- logical investigations in SC such as that describing the primitive archaeocete Dorudon serratus from 94 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) Berkeley County, north of Charleston (Gibbes, 1845). The celebrated 19th century naturalist Louis Agassiz (founder of the Museum of Comparative Zoology at Harvard University) spent considerable time in Charleston and recognized in 1847 after viewing an extensive collection of fossils accumu- lated by Francis S. Holmes “that the Charleston area contained fossil-bearing beds of major sig- nificance” (Sanders and Anderson, 1999:66). As curator of The Charleston Museum from 1851 to 1869, Holmes oversaw the publication of two now classic volumes, Pleiocene [sic] Fossils of South Carolina (Tuomey and Holmes, 1857) and Post- Pleiocene [sic] Fossils of South Carolina (Holmes, 1860), to which Joseph Leidy contributed the sec- tion on vertebrate fossils. Leidy, considered the “Father of American Vertebrate Paleontology” (Warren, 1998), studied and published on fossils from the Charleston area (e.g., Leidy, 1853, 1859, 1860, 1868a, 1869, 1876a–c, 1877, 1890), and his protégé, E. D. Cope, also published several papers on South Carolina fossils (e.g., Cope, 1867, 1868, 1883). In 1915, O. P. Hay examined specimens in the Charleston Museum and in 1923 he described Neochoerus pinckneyi, a new species of capybara, Alces runnymedensis (more recently referred to Cervalces scotti by Sanders, 2002), and the marine turtle Carolinochelys wilsoni. Major paleontological contributions continue today at The Charleston Museum, the South Caro- lina State Museum, and the Mace Brown Museum of Natural History. For example, one of the most significant Paleocene faunas from eastern North America was reported in Sanders (1998a, b, c), and Eocene marine units in the state have yielded the best North American glimpses yet into the early evolution of cetaceans, with several taxa of pro- tocetid and basilosaurid archaeocetes having been recovered (Albright, 1996; Uhen and Gingerich, 2001; Geisler et al., 2005; McLeod and Barnes, 2008; Uhen, 2008; Gibson et al., 2019). Similar explorations by AES and colleagues (particularly J. Geisler) in the above noted Ashley and Chan- dler Bridge formations (together with a host of dil- igent and concerned avocational collectors) have resulted in an unparalleled assemblage of archaic odontocetes and mysticetes of profound evolu- tionary significance (e.g., Whitmore and Sanders, 1976; Sanders, 1980; Sanders et al., 1982; Sanders and Barnes, 2002a, b; Geisler et al., 2014; Sanders and Geisler, 2015; Churchill et al., 2016; Godfrey et al., 2016; Boessenecker et al., 2017a, b). Still under study by JLK, DJC, and colleagues at the SCSM and the USNM is the spectacular Camelot Local Fauna, by far the richest and best preserved Irvingtonian flora and fauna in the state, rivaling those from Florida and California (Kohn et al., 2005; Beaty et al., 2007; Fields, 2010). Additional vertebrate assemblages under study at the SCSM include those from the Walrus Ditch, Crowfield, and Rodent Ditch localities (Chandler and Knight, 2009; Knight and Cicimurri, 2010), which are add- ing several new taxa of Blancan through Rancho- labrean age to the state’s list; a newly recognized latest Paleocene fauna has come to light, as well. There have even been exciting new discoveries of Late Cretaceous vertebrates in South Carolina, including dinosaurs (Weishampel and Young, 1996; Schwimmer et al., 2015). Although long overshadowed by the excep- tionally rich Cenozoic record from Florida (e.g., Hulbert, 2001), as well as the impressive collections from Lee Creek Mine, North Carolina (Ray et al., 2008), renewed efforts in South Carolina over the last three to four decades have resulted in a wealth of new data from sites that rival, and in some cases surpass, any others along the USA Atlantic Coastal Plain. Current studies of these sites are resulting in a much greater understanding of paleobiodiversity along the southeastern coastal plain than was avail- able only a few years ago. CENOZOIC VERTEBRATE FOSSIL-BEAR- ING BEDS OF THE SOUTH CAROLINA COASTAL PLAIN The stratigraphic foundation presented in this report is based on studies of the SC Coastal Plain by the United States Geological Survey, the South Carolina Geological Survey, and by academic and state geologists across the eastern USA. Sedi- ments representing approximately 40 named for- mations were deposited on the upper and lower Coastal Plain of the state during Cenozoic time (Figures 2A–C), with others not yet named or fully ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 95 characterized (e.g., offshore Pleistocene units). Although some are accessible in outcrops along watercourses, road-cuts, or in exposures in com- mercial quarries or construction sites, many are known only from subsurface encounters in auger holes and cores, such as the Fishburne, Parachu- cla (although exposed on the Georgia side of the Savannah River), and Ebenezer formations. Oth- ers, such as the Marks Head Formation, the Coo- sawhatchie Formation, and/or the Wabasso beds, may have exposures on river bottoms in the coastal region of Jasper and Beaufort counties, as does the Ashley Formation in, for example, the Cooper and Edisto rivers. (Note: the Marks Head Formation is also exposed on the Georgia side of the Savannah River at Porters Landing). Vertebrate fossils other than those of fish have been recovered from eight Paleogene for- mations, three Neogene formations (all Pliocene in age), and eight Pleistocene units including the unnamed terrestrial sediments now offshore, but deposited across the continental shelf during Wis- consian time prior to Holocene inundation (Table 1). The offshore deposits have yielded a rich record of Rancholabrean-aged fossils, specimens of which are regularly washed ashore along the pres- ent coasts of Myrtle and Edisto beaches – the latter being one of the best-known localities for Pleisto- cene mammal remains on the USA Atlantic Coastal Plain (Roth and Laerm, 1980; Sanders, 2002). For that reason these offshore deposits are included among the recognized stratigraphic units even though they are not currently formally named or characterized. All other units of the lower Coastal Plain are marine in origin and formally recognized by the USGS, with the exception of the Ten Mile Hill Beds – an informal name used in USGS litera- ture (e.g., Bybell, 1990; Edwards et al., 2000) for middle to late Pleistocene sediments in the Charles- ton area, but recently formalized by Sanders et al. (2009) as the Ten Mile Hill Formation. In this section, we focus primarily on those geological units from which vertebrate fossils (mostly mammals) have been recorded; but we also provide limited discussion of some units that have not yet yielded vertebrates (unequivocally), such as the lower Eocene Fishburne Formation, the middle Eocene Congaree and Warley Hill forma- tions, and the Miocene Parachucla, Marks Head, and Coosawhatchie formations. In addition to these units from the lower Coastal Plain, nearly all of the formations from the upper Coastal Plain (Figs. 1, 2A) lack a vertebrate record with the exception of fish (noted below; see Nystrom et al., 1991; Fallaw and Price, 1995; and Edwards et al., 2001). These include the following: - the lower Paleocene Sawdust Landing For- mation (NP1; updip equivalent of the Rhems Formation; unpublished shark, ray, and fish material is noted from this unit by DJC); - the middle to upper Paleocene Lang Syne and Table 1. Formally and informally named geologic units in South Carolina that have produced fossils of tetrapods (oldest to youngest descending; see Figures 2 and 3 for detailed chronology of these units). Paleogene Neogene Quaternary/Pleistocene Williamsburg Formation ?Marks Head Fm. Waccamaw Formation Santee Limestone Goose Creek Limestone Penholoway Formation Tupelo Bay Formation Raysor Formation Ladson/Canepatch Formation Parkers Ferry Formation Duplin Formation Ten Mile Hill Formation Harleyville Formation Socastee Formation Ashley Formation Wando Formation Chandler Bridge Formation Unnamed Ardis LF deposits Tiger Leap/Edisto Formation Offshore Rancholabrean deposits 96 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) Snapp formations (updip equivalents of the Williamsburg Formation); - the lower Eocene Fourmile Branch Forma- tion (NP11; updip equivalent of the Fish- burne Formation); - the lower to middle Eocene Huber Formation (NP12–14; updip equivalent of the Congaree Formation; shark, ray, and teleost material was noted from the Huber by Kite, 1982, and Nystrom et al., 1991); - the middle Eocene Warley Hill Formation (NP15/lower 16; updip equivalent of the Cubitostrea lisbonensis zone of the Santee Limestone [Harris and Fullagar, 1987]); - the upper middle Eocene McBean/Tinker Formation (NP16; updip equivalent of the Santee Limestone Cubitostrea sellaeformis zone); - the upper Eocene Clinchfield Formation (NP17(?)–18; likely updip equivalent of the Pregnall Member of the Tupelo Bay For- mation based on remains of basilosaurid archaeocetes and a fragment of brontothere tooth reported from Georgia by Westgate (2001); no vertebrate material known from this unit in SC; - the upper Eocene Dry Branch Formation (NP19–20; updip equivalent of the Parkers Ferry Formation; shark, ray, and fish material was noted from the Dry Branch by Fallaw and Price, 1995; Cicimurri and Knight, 2019); - the upper-most Eocene Tobacco Road Sand (updip equivalent of the Parkers Ferry and Harleyville formations; shark and ray teeth were noted from the Tobacco Road Sand by Nystrom et al., 1991); - the middle Miocene (late Serravalian) Upland Unit (equivalent to middle Miocene Altamaha Formation per Nystrom et al., 1991, Huddles- tun, 1988, and Weems and Edwards, 2007a; but see alternative interpretation as an updip equivalent of the upper Oligocene Chandler Bridge Formation by Katuna et al., 1997; also see Colquhoun et al., 1993); and - the “upper Miocene/lower Pliocene” Pine- hurst Formation (eolian dune deposits; con- sidered as “probably Pleistocene glacial-age dune deposits” by REW). Figures 2 and 3 accompany the following dis- cussion and provide our interpretations of the tem- poral placement of SC’s Cenozoic strata. Figure 3, in particular, is a compilation that includes corre- lation of refined calcareous nannoplankton zones, Atlantic planktonic foraminiferal zones, dinofla- gellate zones, NALMAs, and the δ18O record to the GTS2012 time scale. Both figures also provide, for the reader’s convenience, the boundary dates associated with each Epoch and Stage following GTS2012. It will be noticed that in some cases our placement of SC’s stratigraphic units differs from traditional interpretations, but this is due in large part to the highly refined, astronomically tuned nature of the marine δ18O record (Lisiecki and Raymo, 2005; Raffi et al., 2006) and its lat- est correlation to the most recent GPTS, as well as to our interpretations. This in turn has resulted in revised ages for some of these units. Although this report focuses primarily on the mammalian bio- stratigraphy of the state, certain biochronologically significant invertebrate taxa are also noted because of their utility in providing refined temporal place- ment of some of the units from which vertebrate fossils were recovered (e.g., Appendix 4). Paleocene Series (66.0 –56.0 Ma) What is known of the Paleocene vertebrate fauna of South Carolina is derived primarily from an enormous pit excavated in 1979 near the town of St. Stephen, Berkeley County, as part of a proj- ect to redivert water previously diverted from the Santee River to the Cooper River back into the Santee. The pit, reaching a depth of ~43 m (140 feet), penetrated the Williamsburg Formation of the Black Mingo Group, exposing virtually the entire Chicora Member and the top of the underlying Lower Bridge Member (Weems and Bybell, 1998). At the pit locality the Williamsburg Formation is overlain unconformably by the lower Pleistocene Penholoway Formation. The geology, paleobotany, and vertebrate fauna sampled from the St. Stephen pit and its ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 97 extensive spoil piles were analyzed in a volume edited by Sanders (1998a), providing the first glimpses of early Cenozoic paleobiology from the southern Atlantic Coastal Plain. In his summary Sanders (1998b:262) listed 31 genera and 43 spe- cies of vertebrates including sharks, rays, a saw- fish, bony fishes, turtles, a snake, crocodilians, and mammals from what is thought to be the Chicora Member. Five reptilian taxa from the Williams- burg Formation, the turtles Adocus, Agomphus, Taphrophys, Aspideretes, and possibly Bothremys (Hutchison and Weems, 1998), together with two crocodilians from Cretaceous beds in Florence and Darlington counties, Bottosaurus and Thora- cosaurus (Erickson, 1998a), provide the first evi- dence of a continuum of vertebrate taxa across the Cretaceous-Paleogene (K-Pg) boundary in South Carolina. Pollen and other paleobotanical material indicate a subtropical environment in this region in late Paleocene time (Melchior, 1998). Another site yielding fossils of Paleocene age is the Clapp Creek locality within the town limits of Kingstree, SC. This locality, first noted by Sanders (1998c:6) and by Weems and Bybell (1998:10–11), consists of two open pits immediately east of the creek that were dug by Ms. A. Baker, an avoca- tional fossil hunter, with mechanized heavy equip- ment in 1987. Each pit was originally about 10–12 meters long by 4–5 m wide, and the pit nearest the creek is separated from it by about 3 meters. Apparently abandoned for many years, current efforts to recover fossils there are difficult in that only hand tools can be used to remove the years of accumulated slumped overburden. Constant pump- ing or bailing is required to keep the water level low, but the pit is never dry and an accurate eval- uation of the stratigraphy is difficult at best. The fossils originate from a prominent lag deposit in a coarse-grained, non-indurated, muddy quartz sand unit approximately 2–2.5 meters below the ground surface. The fossils are collected by shoveling the muddy sand at the lag level into buckets, and then screen-washing the contents of the buckets in the weakly flowing creek. The most ambitious recent efforts to recover fossils from the site are currently being undertaken by Dr. C. Ciampaglio and stu- dents from Wright State University, Celina, Ohio, who are often joined by SC avocational fossil col- lectors B. Palmer (deceased) and R. Shafto. Web- sites highlighting South Carolina fossils indicate that others are collecting from the site as well. The fossils originate from a temporally mixed lag deposit that yields an abundance of croc- odilian, turtle, and fish material, plus coprolites, but also includes material of dinosaurs, mosasaurs, plesiosaurs, and Pliocene to Pleistocene mammals such as rodents, horses, tapirs, mastodons (or gom- photheres), elk, and sloth. Although Schwimmer et al. (2015) considered the Clapp Creek dino- saur material as having originated from the upper Campanian Donoho Creek Formation, more recent analysis of the Cretaceous elasmobranchs by DJC concludes that they are Maastrichtian in age, and that there are no taxa strictly indicative of the Cam- panian. The unit of origin for the dinosaurs, mosa- saurs, and plesiosaurs, therefore, is here considered to be the Steel Creek Formation – a Maastrichtian deltaic unit mapped in Georgia and South Carolina (hence the mixture of marine and worn specimens of terrestrial taxa), and correlative with the offshore deposits of the Peedee Formation (Fallaw and Price, 1995) – rather than the Donoho Creek Formation. Regarding the Paleocene, Danian sharks teeth are found at this site, apparently originating from the Rhems Formation (Santana et al., 2011), but mam- mals have yet to be recovered. Additional material of Paleocene age was noted by Cicimurri et al. (2016) from sediments immediately underlying the Santee Limestone at the Martin Marietta Materials quarry near James- town, northeastern Berkeley County (Fig. 1; detailed discussion below). Previously referred to as the “Jamestown beds” (Cicimurri and Knight, 2009a; Cicimurri, 2010) and thought to be early Eocene in age, the unit yielding the fossils, upon further study (Cicimurri et al., 2016), was found to be latest Paleocene in age and correlated with the Chicora Member of the Williamsburg Formation. In Table 2, vertebrate taxa from this locality have been added to those noted in Sanders (1998b:262) that were recovered from the St. Stephen pit. 98 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) Williamsburg Formation, Lower Bridge Member. Weems and Bybell (1998) described the Lower Bridge Member at the St. Stephen pit as lithified to semi-lithified silts to fine-grained sands and silty cristobalitic shales. They added that the only unlithified part of this unit was the uppermost two feet. Fossils recovered included “remains of sharks, rays, the bony fish Phyllodus, and small ... crocodile teeth” (Weems and Bybell, 1998:13, 25). Silty, lithified matrix adhering to a vertebra of the crocodilian Hyposaurus indicated that it, too, origi- nated from this unit. Edwards (1998), studying dinoflagellates from the Lower Bridge Member exposed in the pit, determined that its age was equivalent to calcar- eous nannoplankton zones NP3, 4, or 5 (early to middle Paleocene). The age was refined to NP5, and to reversed magnetochron C26r, upon study of a core drilled in Charleston County (Edwards et al., 1999), followed by further revision to zones NP4–5 (approx. 60 Ma, middle Paleocene, Selandian; Figs. 2A, 3A) after study of another core drilled in Dorchester County (Edwards et al., 2000). Williamsburg Formation, Chicora Member. The Chicora Member is composed of medium to coarse-grained sands with an abundantly shelly stratum observable at the stratotype outcrop on the Santee River about 0.8 km (0.5 miles) northeast of Wilson’s Landing in northern Berkeley County (Van Nieuwenhuise and Colquhoun, 1982; Sand- ers, 1998a; Weems and Bybell, 1998). In addition to four gastropod and nine bivalve taxa (including the first Paleocene record of Pitar ovatus), the stra- totype has also yielded vertebrate remains includ- ing ray dental plate segments, sharks teeth, and a crocodilian tooth. At the St. Stephens pit, Weems and Bybell (1998) noted a prominent lag deposit at the bound- ary between the Chicora Member and the underly- ing Lower Bridge Member that included rounded phosphate pebbles and worn bone fragments. Edwards’ (1998) study of dinoflagellates from the Chicora Member at the pit resulted in an NP8 or 9 assignment, but, as with the Lower Bridge Member, additional study from the Charleston and Dorches- ter cores revised the age to NP5-9 (upper middle to upper Paleocene, upper Selandian-Thanetian). The land mammal fossils from the St. Stephen pit, referred to as the Black Mingo Fauna by Schoch (1985, 1998), were all collected from spoil piles, therefore precluding an exact determination as to their unit of origin. Schoch (1985, 1998), however, concluded that they were likely derived from the Chicora Member. Represented by only five teeth, two were described as new subspecies of the con- dylarth Phenacodus grangeri (P. grangeri mccol- lumi) and the taeniodont Ectoganus gliriformis (E. gliriformis lobdelli), and a third was referred to an enigmatic new taxon of uncertain ordinal position, Mingotherium holti. Another partial tooth was too incomplete to be referred beyond “Tribosphenida incertae sedis,” and a large caniniform tooth was thought to be representative of a pantodont, uin- tathere, or large condylarth (Schoch, 1998:238). Phenacodus grangeri is known from Tiffanian 1–5 faunas elsewhere in North America (Thewissen, 1990), and Ectoganus occurs in Ti5 through the Clarkforkian (Archibald et al., 1987). Lofgren et al. (2004) assigned the Black Mingo Fauna to Ti5, which is compatible with their derivation from the Chicora Member given the dinoflagellate correla- tion noted above, although the possibility that they originated from the unlithified upper two feet of the Lower Bridge Member cannot be discounted. In addition to the mammalian component, Erickson (1998b) reported a vertebra of a palaeophid snake from the fauna (Table 2). “Jamestown beds.” From sediments imme- diately underlying the Santee Limestone at the Martin Marietta Materials quarry near Jamestown, northeastern Berkeley County, Cicimurri and Knight (2009a:24) reported “a highly diverse verte- brate assemblage consisting of elasmobranch, oste- ichthyan, and reptilian species,” which, together with the invertebrates, indicated an Ypresian (early Eocene) age. Included in the osteichthyan assem- blage, they noted the perciform fish, Fisherichthys folmeri, originally described by Weems (1999) from the basal portion of Bed B of the Potapaco Member of the Nanjemoy Formation of Virginia, and assigned to calcareous nannoplankton zone NP11 by Gibson and Bybell (1991). Fisherichthys ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 99 folmeri is also known from the upper part of the Tuscahoma Formation in Alabama, considered earliest Wasatchian (Wa0) in age (correlative with calcareous nannoplankton subzone NP9b) on the basis of mammals recovered (Beard and Dawson, 2001, 2009). This species has also been reported from the marine portion of the Bashi Formation of eastern Mississippi (Case, 1994) deposited during the lower half of NP10 (see Cicimurri and Knight, 2009a:25). Another species reported by Cicimurri (2010) from the Jamestown deposits, plus the Bashi and Nanjemoy formations, is the extinct early Eocene ray Meridiania convexa Case, 1994. Taken together, the above noted biochrono- logical data seemed to suggest an early Eocene age. Although Cicimurri and Knight (2009a) noted that these Jamestown sediments were lithologically dif- ferent than those of the lower Eocene Fishburne Formation (see below), therefore precluding direct correlation, they considered the possibility that the two units were laterally equivalent. In a later study describing the myliobatoid ray Eorhinoptera grab- dai (here considered a junior synonym of Meridi- ania convexa), Case et al. (2011) assigned these deposits to the Fishburne Formation. However, more recent analysis of the cal- careous nannoplankton from these sediments by Cicimurri et al. (2016) resulted in a refinement of Table 2. Summary of taxa from the Williamsburg Formation, including those collected in the pit excavated for the Santee rediversion project near St. Stephen, Berkeley County (summarized from Sanders, 1998b:table 2) and from “Jamestown deposits” (see text). Chondrichthyes Osteichthyes (cont.) Myliobatis sp. Progymnodon hilgendorfi Rhinoptera sp. Chelonia Meridiania convexa Cheloniidae indet. Rhinobatos bruxelliensis Osteopygis emarginatus Pristis sp. Taphrosphys sulcus Nebrius sp. ?Bothremys sp. indet. Carcharius macrotus Agomphus pectoralis Carcharius hopei “Agomphus” sp. aff. “A.” alabamensis Odontaspis rutoti Adocus sp. indet. Otodus obliquus Aspideretes virginianus Cretolamna appendiculata Kinosternoid A Palaeocarcharodon orientalis Kinosternoid B ?Scyliorhinus sp. Squamata ?Triakis sp. Palaeophis sp. Coupatezia woutersi Glyptosaurinae gen. et sp. indet. Jacquhermania duponti Crocodilia Heterotorpedo fowleri Hyposaurus sp. Ischyodus sp. Bottosaurus sp. Osteichthyes Thoracosaurus sp. Lepisosteus sp. Eosuchus sp. ?Pycnodus sp. Mammalia Albula oweni Didymictis proteus Egertonia isodonta Mingotherium holtae Phyllodus toliapicus Phenacodus grangeri mccollumi Ostraciidae indet. Ectoganus gliriformis lobdelli Fisherichthys folmeri Tribosphenida indet. 100 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) the age of the Jamestown deposits to latest Paleo- cene subzone NP9a (i.e., uppermost Thanetian). They also determined, on the basis of lithologic and paleontologic similarity, that these beds are correla- tive with the Chicora Member of the Williamsburg Formation, not with the Fishburne Formation, and therefore assigned the Jamestown deposits to that member (Figs. 2A, 3A). Supporting the NP9a age of these James- town beds is a small partial mammal tooth found in 2007 by Charleston Museum volunteer Bill T. Palmer. The tooth was recovered from the same spoil pile in which the material of Cicimurri and Knight (2009a) and Cicimurri et al. (2016) origi- nated. It consists only of the trigonid, but has been identified as belonging to the late Tiffanian (Ti5) through earliest Wasatchian (Wa0) viverravid Did- ymictis proteus Cope, 1875 (J. Bloch and P. Morse, pers. comm. to LBA, 2015; see “Systematic Pale- ontology” section below). This specimen (ChM PV7687) represents the first viverravid known from the USA Atlantic Coastal Plain. How close the specimen is in age to the mammals of the Black Mingo Fauna, also considered as originating from the Chicora Member, is undeterminable, but the work of Cicimurri et al. (2016) points toward a late Thanetian rather than Ypresian age (Clarkforkian NALMA, approx. 56 Ma). Eocene Series (56.0–33.9 Ma) Fishburne Formation. Gohn et al. (1983) described the Fishburne Formation from the 416–440 ft interval of the Clubhouse Crossroads Corehole No. 1 (CCC1; Fig. 1B) from Dorches- ter County as a greenish-gray to pale-olive, finely crystalline, nodular, glauconitic, clayey, microfos- sil-mollusc limestone with apparent lack of bed- ding due to bioturbation. In the core, the Fishburne sharply underlies the Santee Limestone (NP16); the stratigraphically intervening Warley Hill For- mation (NP15) is absent. Although CCC1 was recovered about 40 km (24 miles) southwest of the Jamestown quarry, a much closer corehole was drilled by the USGS in 1996 on the Santee Coastal Reserve near the Santee River in northeastern Charleston County, about 20 km (12 miles) southeast of the quarry (Edwards et al., 1999). In that core, a 9.5 ft-thick section (2.9 m) of mollusc-bryozoan limestone was encountered above the upper Paleocene Chicora Member of the Williamsburg Formation – a stratigraphic position similar to that of the stratotype Fishburne Forma- tion. Edwards et al. (1999:28) observed that “This poorly recovered limestone is not the same age as the Santee and may consist of two units of differ- ent ages. It may be equivalent to the lower Eocene Fishburne Formation (Gohn et al., 1983), or the lower Eocene Congaree Formation (Fallaw and Price, 1995), or parts of both.” The best evidence for the age of this unit as determined by Edwards et al. (1999) comes from three samples of the core that yielded calcareous nannofossils indicative of uppermost zone NP9 to NP12 (Ypresian). Enig- matic as this stratum may be for formational inter- pretation, it does establish the presence of lower Eocene deposits relatively close to the Jamestown quarry at least partly correlative with the Fishburne Formation. Congaree Formation. Cooke and MacNeil (1952) elevated Sloan’s “Congaree phase” to for- mational rank, and correlated it to the Tallahatta Formation of Alabama and Mississippi on the basis of similar lithology and on the mutual occurrence of the molluscs Anadontia augustana and Ostrea johnsoni. Fallaw and Price (1995) suggested an NP12–14 (Upper Ypresian-Lower Lutetian) age for the Congaree. No vertebrates are known from this unit. Warley Hill Formation. Referred to by Sloan (1908:458) as the “Warley Hill phase,” as the War- ley Hill Marl by Cooke and MacNeil (1952:23), and as the Warley Hill Formation by Pooser (1965), these dominantly glauconitic non-calcareous sedi- ments containing the oyster Cubitostrea lisbonen- sis correlate with the lower Lisbon Formation of Alabama. The unit lies between the Congaree For- mation below and the Santee Limestone above (in the currently restricted sense, i.e., the Cubitostrea sellaeformis zone; see further discussion in the sec- tion on Santee Limestone below and on the McBean and Tinker formations in Kier, 1980, and Fallaw and Price, 1995). Kier (1980) dated the Warley Hill Formation as lower middle Claibornian on the ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 101 basis of the occurrence of the echinoid Protoscu- tella mississippiensis, and Fallaw and Price (1995) noted an NP15 age (middle Eocene, upper Lute- tian). Harris and Fullagar (1987) reported a Rb-Sr glauconitic mica isochron date of 42.0 ± 0.5 Ma for the unit, but later revised this date to 45.1 ± 1.7 Ma (correlative with the NP15 designation) based on a conventional K-Ar date, also from glauconitic micas (Harris and Fullagar, 1991). The only verte- brates known from this unit are some sharks teeth noted by DJC. Santee Limestone. This often highly indu- rated, typically gray limestone was named “the Santee white limestone” by Charles Lyell, who, in the company of Edmund Ravenel in 1842, correctly determined that oysters he observed in outcrops near Eutaw Springs along the bank of the Santee River in Orangeburg County (now submerged) were of Eocene age (Lyell, 1845). Referred to as the “Great Carolina beds” by Ruffin (1843), the “Santee beds” by Tuomey (1848), and the “San- tee marl” by Sloan (1908), it was Cooke (1936) who designated the exposures at Eutaw Springs as the type section of his Santee Limestone (Banks, 1977). The unit is widespread in the subsurface of the SC Coastal Plain, although normally exposed only in deep commercial quarries, such as those at the Martin Marietta Berkeley and Orangeburg quar- ries near Cross, SC.2 These quarries, together with another in Georgetown and a fourth in Jamestown, have yielded over 200 species of Eocene molluscs (Campbell, 1995; Campbell and Campbell, 2003). Ward et al. (1979) divided the Santee Lime- stone into a lower Moultrie Member for the highly indurated bryozoan-dominated carbonate facies with abundant Cubitostrea sellaeformis and an overlying Cross Member for the much more mol- lusc-dominated unit, which includes Crassatella alta. They designated a measured section within 2Approximately 8.9 km west of Cross, SC, straddling the north-south trending County Line Road, are the Martin Mari- etta Aggregates Cross quarries (Fig. 1B, “CQ”). Quarries on the east side of the road, now abandoned and flooded, are in Berkeley County, thus “Martin Marietta Berkeley Quarry”; those on the west side are in Orangeburg County, i.e., “Mar- tin Marietta Orangeburg Quarry” (cited as the “Southern Ag- gregates Orangeburg Quarry” in Campbell, 1995). Unless otherwise noted, reference to the “Cross quarry” implies the Berkeley quarry. the Martin Marietta Berkeley (Cross) quarry as the stratotype for the Moultrie Member, and equated this unit with lithozones I and II of Banks (1977). However, Ward et al. (1979) did not note C. lisbo- nensis in this section – the taxon that occurs strati- graphically below C. sellaeformis and biostrati- graphically characterized Banks’ (1977) lithozone I, and on which a correlation with the lower Lisbon Formation of Alabama and with the Warley Hill Formation was based (Note: Campbell and Camp- bell [2003] also did not record C. lisbonensis in the Cross quarry). Banks (1977) correlated lithozone II, disconformably overlying zone I, with the upper Lisbon Formation based on the presence of C. sel- laeformis. Thus, the Moultrie Member as conceived by Ward et al. (1979) was actually a correlate only with Banks’ (1977) lithozone II, not with zone I. An intermediate C. smithvillensis zone, recognized in the Gulf Coastal Plain, is absent. The Cross quarry was also the location for Ward et al.’s (1979) strato- type of the Cross Member. They correlated this unit with Banks’ (1977) lithozones III and IV, and with the Gosport Sand of Alabama, on the basis, in part, of the presence of Crassatella alta. Baum et al. (1980) divided the Santee Lime- stone into a lower Cubitostrea lisbonensis faunal zone (Banks’ zone I) and an upper C. sellaeformis faunal zone (Banks’ zone II; Chapel Branch Mem- ber of Powell, 1984), and recommended aban- donment of Ward et al.’s (1979) stratigraphically equivalent Moultrie Member. They similarly rec- ommended abandonment of the lithostratigraphic term Warley Hill Formation due to its correlation with their lower Santee, i.e., the Cubitostrea lisbo- nensis zone. Finally, they excluded Ward et al.’s (1979) Cross Member from the Santee Limestone, and elevated it to formational status. Powell and Baum (1982) followed the terminology of Baum et al. (1980:1100–1101), but proposed a Caw Caw Member of the Santee Limestone for “a curious updip equivalent of the typical carbonates of the Santee Limestone …,” adding that the term “Caw Caw” had priority over “McBean.” Edwards et al. (1997, 2000) followed Ward et al.’s (1979) concept of the Santee Limestone, maintaining usage of the Moultrie and Cross ‘member’ terminology; they 102 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) correlated the Moultrie Member to nannoplankton zone NP16 and the Cross Member to zones NP17– 18. More recently Geisler et al. (2005), in propos- ing their new Tupelo Bay Formation as a replace- ment for Baum et al.’s (1980) Cross Formation (fur- ther discussion below), recommended restriction of the name “Santee Limestone” to the indurated unit of NP16 age previously referred by Ward et al. (1979) to the Moultrie Member (Banks’ zone II; C. sellaeformis zone; Fig. 2A, 3A) lying disconform- ably below the Cross Member. Although the above discussion would seem to imply that some level of stability has been finally reached with respect to an understanding of the Eocene stratigraphy of South Carolina, Campbell’s (1995:121) note that “Little consensus exists on nomenclature and correlation” still holds true to some extent today, as does a simi- lar sentiment noted by Campbell and Campbell (2003) that “The Eocene stratigraphy of the [SC] region remains debated.” Regarding vertebrate fossils from the Santee Limestone, the first published account of a proto- cetid archaeocete in South Carolina was Albright’s (1996) report of three teeth (ChM PV5037, 5038, 5039) collected in the Cross quarry, presumably from the Santee Limestone (Cubitostrea sellaefor- mis zone). Although the teeth were collected on a spoil pile, their derivation from the Santee Lime- stone was based on the very similar morphology they share with those of Georgiacetus vogtlensis Hulbert et al., 1998. The Georgia type specimen was collected from the informally named Blue Bluff unit of the McBean Formation, which, based on similar molluscs (including Cubitostrea sellae- formis) and on nannoplankton indicative of zone NP16, is directly correlative with the Santee Lime- stone (Fallaw and Price, 1995). (Note: in Appendix I of Janis et al. [2008:736] the teeth are reported from the overlying Cross Formation [currently the Cross Member of the Tupelo Bay Formation; see discussion below]). As noted below, the Cross Member at the Cross quarry has yielded remains of protocetid archaeocetes, but for reasons noted above ChM PV5037, 5038, and 5039 are thought to have originated from the Santee Limestone; however, their derivation from the Cross Member is possible). Albright (1996) chose not to assign the SC teeth to a known taxon, but subsequent authors, noting the similar dental morphology, referred them to Georgiacetus cf. vogtlensis and to Georgiacetus, aff. G. vogtlensis (Uhen, 1999, and McLeod and Barnes, 2008, respectively). Based on the marine δ18O isotope record (Fig. 3A), it appears that deposition of the Santee Lime- stone coincided with the Middle Eocene Climatic Optimum (MECO) – a warm, high sea level pulse among the background of general climatic deterio- ration that was occurring through the middle and late Eocene (e.g., Edgar et al., 2010; Galazzo et al., 2014). Tupelo Bay Formation. Poorly understood for many years, two of the calcareous Eocene beds above the Santee Limestone – one exposed in the Cross quarry in Berkeley county and the other in the Giant Cement quarry in Dorchester County – have at various times been assigned, together or in part, to the Castle Hayne Formation (Cooke and MacNeil, 1952), the Santee Limestone (Sanders, 1974; Ward et al., 1979), or the Cross Formation (Baum et al., 1980). More recent work, however (Geisler et al., 2005), has shown that none of those arrangements reflected the true chronostratigraphic relationships of these two beds. Recognizing two lithologic subunits within the Cross Formation (of Baum et al., 1980) in the Pregnall core from Dorchester County (Edwards et al., 1997), Sanders and Katuna (2000) recom- mended dividing that unit into a lower Berkeley Member and an upper Pregnall Member. In propos- ing the Tupelo Bay Formation as a replacement for Baum et al.’s (1980) Cross Formation, Geisler et al. (2005) retained the name Cross Member of Ward et al. (1979) rather than “Berkeley Member” for the lower unit and maintained the previously pro- posed Pregnall Member terminology for the upper member, which comprise the deposits underlying the Harleyville and Parkers Ferry formations in the Giant Cement quarry. The two units differ biostratigraphically in their nannoplankton and cetacean content; the Cross Member falls within calcareous nannoplank- ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 103 ton zone NP17 and contains only the remains of archaeocetes of the primitive family Protocetidae, whereas the Pregnall Member is NP18 in age and has yielded only the remains of archaeocetes of the more derived Basilosauridae (Geisler et al., 2005). As seen in Figure 3A, correlation of the Tupelo Bay Formation to NP17 and NP18 indicates that it too was deposited during another transient warm interval, this time in the late Eocene across the Bar- tonian-Priabonian boundary. Tupelo Bay Formation, Cross Member. In 1994, the holotype material (ChM PV5401) of the protocetid Carolinacetus gingerichi Geisler et al., 2005, was collected from the base of the Cross Member at the Cross Quarry by a ChM party. This material consists of a partial skull, seven vertebrae, and 15 ribs. As seen in figures 3 and 4 of Geisler et al. (2005; and reproduced herein as Figure 4), the nasal opening is situated above the canine tooth – a plesiomorphic feature previously unrecorded in New World protocetids (e.g., Georgiacetus) and more typical of Old World forms such as Rodho- cetus. On that basis, Carolinacetus is currently the most primitive known cetacean from North Amer- ica (Geisler et al., 2005). Additional protocetid material from the Cross Member includes six partial skulls (ChM PV6850, PV6856, PV6950, PV8002, PV8003, PV8022) at least some of which repre- sent undescribed taxa. Collected for The Charles- ton Museum by B. Palmer (who passed away in 2018), this material represents the largest body of protocetid material from the Western Hemisphere. ChM PV6950 was recently described by Gibson, et al. (2019) as the new taxon Tupelocetus palmeri in Figure 4. Reconstruction of the holotype skull of Carolinacetus gingerichi (ChM PV5401) from the Cross Member of the Tupelo Bay Formation. A, dorsal view; B, right lateral view. Scale bar = 10 cm. Reproduced from Geisler et al. (2005) with permission of the AMNH. 104 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) honor of Palmer’s contributions. The ChM holdings from the Cross Member also include fish remains, small portions of the carapace of an undescribed leatherback sea turtle (ChM PV9102 plus material at the SCSM, the ear- liest yet known from South Carolina), a vertebra of the large Eocene snake Pterosphenus schucherti (ChM PV5766), and a large, diverse assemblage of crabs also collected by Palmer (see Blow and Manning, 1996; Bishop and Palmer, 2006; and Franţescu et al., 2010). Also from the Cross Member, but from the Martin Marietta Orangeburg quarry, is another specimen collected by Palmer that includes the partial inominate (GSM 1333) of an early sirenian conservatively referred to Protosiren sp. by Beatty and Geisler (2010), who noted that it was collected within the basal 30 cm of the member. Tupelo Bay Formation, Pregnall Member. According to Geisler et al. (2005), it is this mem- ber of their Tupelo Bay Formation that is exten- sively exposed in the Giant Cement quarry and in the nearby Argos Cement Plant quarry (formerly called the LaFarge, Blue Circle, or Gifford-Hill & Co. Harleyville quarry), both near Harleyville (Fig. 1B). In 1973, when quarrying activities were under- way in the now-abandoned southwestern portion of the Giant Cement quarry, a paleontological survey of the Eocene beds was conducted by a Charleston Museum party under the direction of AES and of which LBA was a member (Sanders, 1974). Cooke and MacNeil (1952) had assigned the lower beds in the quarry to the Castle Hayne Limestone, but Sanders (1974) provisionally referred them to the Santee Limestone. These are also the beds that Ward et al. (1979) referred to the Cross Member of the Santee Limestone, additionally noting the eroded, irregular nature of the unit’s top surface. But it was the recognition by Edwards et al. (1997) in the USGS Pregnall No. 1 core from Dorchester County (Fig. 1B) that the Cross Member spanned two calcareous nannoplankton zones (NP17 and 18) with a subtle change in lithology that prompted Geisler et al. (2005) to propose the NP17-aged Cross Member and the NP18-aged Pregnall Mem- ber as subunits of their Tupelo Bay Formation. During the 1973 Charleston Museum survey, what is now considered the Pregnall Member by Geisler et al. (2005) was measured to a depth of 9.76 m below its upper surface, and eight distinct faunal and/or lithological facies were recorded (Sanders, 1974:6). Cooke and MacNeil (1952:26) had previ- ously separated their “Castle Hayne limestone” in the quarry into two beds. Their Bed 1 equates with the lower-most zone of Sanders (1974), from 5.05 m below the top of the Pregnall to 9.76 m, and their Bed 2 is equivalent to Sander’s zones 1–7 above the 5.05 m horizon (Fig. 5). From their Bed 2 they recorded the molluscs Chlamys cookei, Chlamys n. sp., Glycymeris staminea, Chlamys n. sp. aff. C. deshayesii, Ostrea trigonalis (= Pycnodonte trigo- nalis) and the echinoid Periarchus lyelli, the latter of which they also found in Bed 1 along with C. cookei. To date, 32 taxa of gastropods and bivalves collected during the Charleston Museum survey have been identified by D. C. Campbell (pers. comm. to AES, 2014). Also revealed by the 1973 survey was the highly fossiliferous nature of the contents of the “solution pits” eroded into the upper surface of the Pregnall Member. These pits are filled with olive- gray phosphatic sediments from the phosphate zone at the base of the unconformably overlying Har- leyville Formation, and it was from these pits that all of the archaeocete material (Basilosauridae only) collected over the course of the survey was recov- ered (Sanders, 1974). However, undescribed mate- rial currently housed at the SCSM was collected in situ from the Pregnall Member, including that of basilosaurid archaeocetes and early sirenians. Regarding such material, a partial basi- losaurid skeleton dug by machinery in the Giant Cement quarry and currently housed at the SCSM (SCTC279), includes a small, elongate lumbar ver- tebra. Although its length/width ratio of 0.55 falls within the range of the B. cetoides lumbar verte- brae measured by Kellogg (1936:53), the speci- men is considerably smaller than vertebrae of that taxon. From the nearby Argos Cement quarry, the purported Pregnall Member yielded a partial ver- tebral column of another undescribed basilosaurid archaeocete (ChM PV6761) collected by B. Palmer ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 105 Figure 5. Redraft of Sanders’ (1974) figure 2 with updated stratigraphic nomenclature, inclusion of represented calcareous nannoplankton (NP) biostratigraphy, demonstration of Parkers Ferry Formation subjacent to Harleyville Formation vs. previous superjacent interpretations, and proposal to extend Harleyville Formation upward to include beds of potential NP23 age. See text for further discussion. 106 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) from near the bottom of the exposure (Geisler et al., 2005). Still another basilosaurid, Chrysocetus healyorum (SC87.195), was described from the “Cross Formation” by Uhen and Gingerich (2001) from remains found in the Holcim Inc. Holly Hill quarry (formerly the Santee Portland Cement Com- pany quarry) in Orangeburg County approximately 3.2 km (2 miles) NNE of the Giant Cement quarry (Fig. 1B). Uhen (2013) modified the stratigraphic unit of origin to the Pregnall Member of the Tupelo Bay Formation following Geisler et al.’s (2005) re- evaluation of the stratigraphy of that area. In addition to the remains of basilosaurid archaeocetes, the record of early sirenians has been expanded beyond the specimen noted above from the Cross Member at the Martin Marietta Orange- burg quarry, with several additional specimens now known from the Pregnall Member at the Giant Cement quarry. This material includes USNM 537206, a skull and partial skeleton of an archaic dugongid found by B. Palmer; SC2006.30.1–15, another skull and partial skeleton found by JLK and V. McCollum that may represent the same taxon; and ChM PV7639, a skull cap of a small, as-yet undetermined species. USNM 537206 and SC2006.30.1–15 are exceptional specimens cur- rently under study by D. P. Domning and I. S. Zalmout, who are provisionally considering them a new species of Eotheroides (D. Domning, pers. comm. to LBA, May 2015). Eotheroides was pre- viously known only from the middle to late Eocene (Lutetian to early Priabonian) of Egypt, India, and Madagascar (Zalmout and Gingerich, 2012). ChM PV7639 was first mentioned by Sanders (1974), then later reported as lost by Domning et al. (1982); but the specimen was subsequently relocated with archaeocete material in the collections of The Charleston Museum. Although ChM records indi- cate that it was collected in August of 1973 from the Harleyville Formation (from Harleyville For- mation-filled erosional pits at the top of the Preg- nall Member), Beatty and Geisler (2010) noted that it may have been derived from the Pregnall Mem- ber. Another skull cap, SC2015.65.1, collected by V. McCollum from a spoil pile near the bottom of the Giant Cement quarry is provisionally referred to Protosiren (V. McCollum, pers. com., 2016). Additional material, including that of marine rep- tiles, is represented by a partial carapace of another leatherback sea turtle (ChM PV7808) and two ver- tebrae of Pterosphenus schucherti (ChM PV5765, PV5768). Note on the stratigraphic position of the archaeocete Dorudon serratus Upper Eocene beds in Berkeley County (then in the old Charleston District) furnished the holo- type partial skull, teeth, and associated vertebrae of Dorudon serratus Gibbes, 1845 (MCZ 8763), the second archaeocete taxon made known to sci- ence following the description of Basilosaurus by Harlan (1834). In its description, Robert W. Gibbes (1845:254), a physician-naturalist of Columbia, SC, and a pioneer investigator of the paleontol- ogy of the state, reported that the type material was found “in a bed of Green sand near the Santee Canal.…on the plantation of R. W. Mazyck, Esq., about three miles from the entrance of the canal from the head waters of the Cooper river,” which was almost directly east of the small crossroads community of Moncks Corner. (Note: on the map of the Charleston District in Mills’ Atlas [1825], the Mazyck property was about 5.5 miles north of Moncks Corner). A cousin of Gibbes, Mazyck had found the bones in a marl pit on his property and reported them to Gibbes. The stratigraphic unit in which the remains were found has never been positively identified. During the 1930s, C. Wythe Cooke of the U.S. Geo- logical Survey visited the old Mazyck Place (then belonging to E. J. Dennis) and noted that “There are said to be several old marl pits, now over- grown, on the plantation,” but he did not actually observe any of them (Cooke, 1936:80). In 1940 the Mazyck property was covered by the waters of the newly-created Lake Moultrie, thus inhibiting fur- ther investigation of the stratigraphic units at this important locality. The most detailed observations of its geolog- ical and paleontological features were those of the SC State Geologist Michael Tuomey, who visited the Mazyck property about 1846 and observed that “The green sand stratum is about four feet thick, ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 107 much indurated towards the lower part, where fos- sil shells are most abundant” (Tuomey, 1848:156). He further noted that “Below this remarkable deposit is a stratum of white marl, abounding in corals” and that the green sand was exposed only in the excavations of the marl pits. Tuomey (1848:156) reported several spe- cies of molluscs from the green sand, including the bivalve Pecten perplanus and the brachiopod Ter- ebratula harlani, and stated that specimens from the site were “now in the collection of Dr. Ravenel, of Charleston.” Clearly, the reference was to Dr. Edmund Ravenel, a well-known conchologist of his day who had guided Charles Lyell to Eutaw Springs, where Lyell discovered the Santee Lime- stone. Ravenel’s collection of molluscs is in The Charleston Museum today. A search of the Rav- enel collection by the second author for some of the molluscan taxa mentioned by Tuomey resulted in a remarkable discovery – a 30 x 24 mm sample of green sand from the Mazyck site containing two single valves of two different pectenids and a brachiopod, evidently Terebratula harlani Mor- ton, 1829, the brachiopod mentioned by Tuomey (1848) as occurring in the green sand. The pectenids remain to be identified. Collected in about 1845, when interest in the green sand was at its peak, this small sample is the only remaining evidence of that stratum from the type locality of Dorudon serratus. Despite the diminutive size of the green sand sample, a small portion of it was removed without damage to the remainder and was found to contain calcareous nannoplankton referable to NP zones 16, 17, and 18 (J. Self-Trail, pers. comm. to AES, July 2009). Because the archaeocete cetaceans known to occur in deposits of NP16 (Santee Lime- stone) and NP17 (Cross Member, Tupelo Bay For- mation) age in South Carolina are all members of the primitive family Protocetidae, and because pro- tocetids do not occur in beds of NP18 age in North America, the NP16 and 17 dates for the green sand are here discounted. Moreover, Dorudon serratus is a member of the more derived family Basilo- sauridae, presently known in South Carolina only from the Pregnall Member of the Tupelo Bay For- mation of NP18 age and from the overlying (in places) Parkers Ferry Formation of NP19–20 age. Isolated remains of basilosaurid archaeocetes are also known from the Harleyville Formation-filled pits eroded into the upper surface of the Pregnall Member (e.g., Sanders, 1974). Given that the lower Harleyville has recently been determined to be of NP21 age (Cicimurri et al., 2016; Weems et al., 2016), the archaeocete material found within the pits and as part of the prominent lag deposit that rests upon the top surface of the Pregnall Member may have been reworked from the latter unit or from the Parkers Ferry Formation. However, there is no green sand stratum currently exposed in the Pregnall Member at the Giant Cement quarry. Therefore, the green sand in which the D. serratus remains were found (here provisionally assigned the informal name Moncks Corner Greensand) may represent a bed within or immediately above the Pregnall Member, but which was later eroded by the Harleyville (or Park- ers Ferry) seas (the base of the Harleyville is NP21 in age and the Parkers Ferry Formation, which also rests immediately above the Tupelo Bay Formation in some areas, is of NP19/20 age; Fig. 5). At pres- ent, however, that inference is merely speculation. Cooper Group. Established by Weems and Lemon (1984a, b), the Cooper Group consists of middle Priabonian through late Rupelian aged sed- iments that comprise, in ascending order, the upper Eocene Parkers Ferry and Harleyville formations and the middle Oligocene (upper Rupelian) Ash- ley Formation (Fig. 2B). Also included were the “Drayton Limestone beds” of Weems and Lemon (1996) for a calcarenite known only from core samples to lie stratigraphically between the Park- ers Ferry and Ashley formations in Charleston and Dorchester counties. However, because these beds were found to contain a similar suite of calcareous nannofossils to that of the Harleyville Formation (NP21), Weems et al. (2016) proposed the term Drayton Limestone Member of the Harleyville Formation for them. The base of the Cooper Group lies at the distinct, disconformable contact with what is now considered the top of the Pregnall Member of the Tupelo Bay Formation. This disconformity occurs 108 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) at the 227 ft level of CCC1 (Hazel et al., 1977), and all sediment from that level up to the 16 ft level, encompassing both the upper Eocene and Oligo- cene, were termed the “Cooper Formation.” Upper Eocene beds were noted as spanning the 227 to 180 ft interval, whereas Oligocene beds extended from that point up to the 16 ft level. Ward et al. (1979) referred the upper Eocene beds between 227 and 186 feet to the Harleyville Member of the Cooper Formation. The beds of this interval are, in turn, correlated with the interval between 255 and 227 feet in the OS-1 core (Ward et al., 1979; Weems et al., 1987) recovered 7.3 miles SE of the CCC1 (Weems et al., 2016). Parkers Ferry Formation. This “glauco- nitic, clayey, fine-grained limestone” (Edwards et al., 2000) was assigned to nannoplankton zones NP19/20 and to planktonic foraminiferal zone P17 (Hazel et al., 1977; Ward et al., 1979; Edwards et al., 2000). The stratotype is in CCC1 and it is also known “from the subsurface along the east side of the Edisto River” (Harris and Zullo, 1991), although Weems et al. (2016) noted the existence of a poorly exposed outcrop on the north side of the east branch of the Cooper River based on the earlier work of Weems and Lemon (1989). Long considered to lie stratigraphically between the Harleyville and Ashley formations, recent work by DJC and J. Self-Trail demonstrated that in the eastern region of the Giant Cement quarry (but not everywhere throughout the quarry), the Park- ers Ferry Formation lies directly upon the Pregnall Member of the Tupelo Bay Formation (Cicimurri et al., 2016; Weems et al., 2016) and beneath the Harleyville (Fig. 5). This is confirmed on the basis of calcareous nannoplankton representative of zone NP19/20 in the Parkers Ferry Formation. Although the basal-most portion of the Harleyville Forma- tion also appears to be of NP19/20, most of this unit is of NP21 age (further discussion below). Long thought to be devoid of vertebrate fossils, more recent work by DJC and JLK at the Giant Cement quarry has determined that the Park- ers Ferry Formation has a rich chondrichthyan and osteichthyan fauna. Additional vertebrate material from this formation includes a series of associ- ated vertebrae from the marine palaeophid snake Pterosphenus schucherti (SC2017.28.1 from the Argos quarry), carapace elements of leatherback sea turtles, and an abundance of material belong- ing to basilosaurid archaeocetes (e.g., SC76.6 and SC2016.31). None of this material appears to be reworked from the subjacent Pregnall Member of the Tupelo Bay Formation. Harleyville Formation. These deposits were first described as the Harleyville Member of the then-recognized Cooper Formation by Ward et al. (1979) from a 3 m thick stratotype section in the Giant Cement quarry in Dorchester County on the east side of SC Route 453, approximately 3.2 km (2 miles) NNE of Harleyville (not “3.2 miles” as stated by Ward et al., 1979:12). Weems and Lemon (1984a, b) subsequently elevated this unit to for- mational status. The Harleyville Formation uncon- formably overlies what Ward et al. (1979) then considered the Cross Member of the Santee Lime- stone, but what Geisler et al. (2005) referred to as the Pregnall Member of the Tupelo Bay Formation. Unconformably overlying the Harleyville Forma- tion, Ward et al. (1979) measured nearly 2 m of Ashley “Member” (now Formation, see below). During the 1973 Charleston Museum sur- vey of the quarry (Sanders, 1974), this unit, then included in deposits known at that time as the Cooper Marl (Cooke and MacNeil, 1952; Sanders, 1974:fig. 2), formed the lower portion of a 5.6 m section in the quarry wall (Fig. 5). The basal 36 cm of that exposure (Zone 1) consisted of olive gray (5Y3/2) clayey phosphatic sediments contain- ing profuse numbers of the small pectenid Chla- mys cocoana Dall, 1898, reported from this stra- tum by Cooke and MacNeil (1952) and confirmed as that taxon by D. T. Dockery (pers. comm. to AES, 2009). Above the basal phosphate zone the Harleyville grades into a greenish gray (5GY6/1) calcarenite and C. cocoana becomes less numer- ous; its HO is at a facies change 1.33 m above the contact with the Tupelo Bay Formation (Sand- ers, 1974:6). Chlamys cocoana is known primar- ily from the Gulf Coast, ranging throughout the Upper Eocene Moodys Branch Formation (NP17) and Yazoo Clay (NP18, NP19/20) into the lower ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 109 Oligocene Red Bluff Formation (NP21) (Dockery and Lozouet, 2003). Also known from a number of localities in Georgia, its occurrence in the Har- leyville Formation at the Giant quarry appears to be the only record of this mollusc from SC. From the location of the section measured during the 1973 survey, the blanket of Cooper Group sediments thinned easterly to a thickness of less than 0.6 m, terminating well short of the active portion of the quarry (in 2016). However, a visit to the eastern, actively worked area of the quarry by DJC in August, 2015, resulted in the determination that approximately 80 cm of Parkers Ferry Forma- tion immediately overlies the Pregnall Member of the Tupelo Bay Formation, and is in turn overlain by approximately 2 m of Harleyville Formation at that location. Revision of the stratigraphic relation- ship between the Pregnall Member of the Tupelo Bay Formation, the Parkers Ferry Formation, and the Harleyville Formation was reported in Weems et al. (2016). The top of the Harleyville Formation in the section of the quarry surveyed in 1973 was defined by a change in lithology at 2.2 m above the con- tact with the Pregnall Member (Fig. 5). Above that change, at 2.5 m, nannoplankton were not as infor- mative because of the absence of certain marker species; but present was Ericsonia formosa, a form that last appears at the top of NP21. Previously published ages for the Harleyville based on calcar- eous nannoplankton placed the unit within zones NP18 through NP19/20 (Laws, 1988; Edwards et al., 1997, 2000), but more recent analysis of sedi- ments from the type section at the Giant Cement quarry indicates that the Harleyville lies primarily within NP21, with indicators of NP19/20 in the lowermost part – particularly the calcareous nan- nofossil Isthmolithus recurvus and the coccolitho- phore Discoaster barbadiensis (J. Self-Trail, pers. comm. to AES, 2009; L. Edwards and J. Self-Trail, pers. comm. to LBA, 2012). Weems et al. (2016) placed the entirety of the Harleyville Formation within NP21, noting that the presence of I. recur- vus in the basal-most portion of the formation did not necessarily indicate an NP19/20 age because that taxon extends up into NP21, and that its pres- ence could be the result of reworking from nearby Parkers Ferry strata, which is of NP19/20 age. A sample taken 3.4 m above the Pregnall con- tact in the eastern part of the quarry visited by DJC in 2015 is suggestive of zone NP23 (Fig. 5) based on the absence of Reticulofenestra umbilicus (LAD at NP22/23 boundary and present in samples strati- graphically lower), the presence of Sphenolithus distentis (FAD in middle NP23), and the absence of marker species for NP24 and NP25 (J. Self-Trail, pers. comm. to AES, July 2009). Based on the mea- sured section of Ward et al. (1979:fig. 5), this level would be within the lower part of their “Ashley Member.” As noted by both Ward et al. (1979) and Weems et al. (2016), however, the Ashley Forma- tion includes calcareous nannofossils indicative of NP24, i.e., upper Rupelian. Perhaps these strata between the above noted lithologic change at 2.2 m above the contact with the Pregnall Member and the overlying Ashley Formation, of NP24 age, rep- resent a previously unrecognized upper part of the Harleyville Formation. Although NP21 straddles the Eocene/Oli- gocene boundary (uppermost Priabonian to low- est Rupelian; Fig. 3A), several age-diagnostic dinocysts (including Batiacasphaera baculata, B. compta, Cordosphaeridium funiculatum, Homo- tryblium plectilum, Samlandia chlamydophora, and Trigonopyxidia fiscellata) support a latest Eocene age rather than an early Oligocene age for these sediments (L. Edwards, pers. comm. to LBA, 2012; Weems et al., 2016). An accurate age for the lower portion of the Harleyville Formation, the C. cocoana zone, is particularly significant because it is from this zone, primarily from Harleyville-filled solution pits eroded into the top surface of the Pregnall Member, that all known vertebrate fossils from this unit are derived, including the isolated archaeocete elements recovered during the 1973 Charleston Museum survey (Sanders, 1974). Sanders (1974:8) noted two teeth from the Giant Cement quarry that “had been found out of place prior to the [1973 ChM] survey” that he tentatively referred to Zygorhiza. He added, how- ever, that the assignment was tentative because this taxon had not been previously recorded from 110 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) SC. More recent work by Uhen (2013) apparently confirmed the limited distribution of Zygorhiza to Priabonian aged deposits of the Gulf Coastal Plain only, i.e., this taxon has not yet been found in simi- lar aged deposits of the Atlantic Coastal Plain. Of particular interest among the archaeocete material, however, is another small, elongate, lumbar verte- bra (ChM PV7636) similar to the SCSM specimen noted above from the Pregnall Member. Although proportionate in form, Sanders (1974) hesitated to assign the specimen to Basilosaurus because its overall dimensions are smaller than those given for the lumbar vertebrae of B. cetoides by Kellogg (1936:53). However, its length/width ratio (0.46) is roughly comparable to those of that taxon (0.49- 0.55) as determined from Kellogg’s (1936:53) mea- surements of USNM 4675 and 12261. These two specimens from SC (in addition to ChM PV7637, a partial vertebra with similar proportions to ChM PV7636) appears to document the presence of an undescribed diminutive species of Basilosaurus in the late Eocene seas of the western North Atlantic. Another marine mammal element thought to have been reworked from the Pregnall Member and deposited into one of the Harleyville-filled pits is the skull cap of the small sirenian noted above in the section on the Pregnall Member (ChM PV7639 and further discussed in “Systematic Paleontol- ogy” section). In addition to the marine mammal material noted from these pits are two exception- ally important fossils representing terrestrial taxa. During the preparation of this report it came to our attention that an avocational fossil collector, Mr. J. Metts, had in his possession an exception- ally well-preserved upper molar of a large bron- tothere from the Giant quarry, as well as a palate with teeth of the primarily Chadronian to Orellan rhinoceros Subhyracodon collected in the nearby Argos Cement Plant quarry. Although both speci- mens are currently maintained in Metts’ private collection, the first author was allowed to exam- ine, measure, and photograph them. However, because they are privately held, we are unable to give them the detailed treatment in the “System- atic Paleontology” section of this report that they deserve. Efforts to obtain casts of these specimens were unsuccessful, and at this time it is not known if these specimens will ever be catalogued into a public repository.3 Both were collected from the Harleyville For- mation-filled solution pits at the top of the Pregnall Member, and this was confirmed by a sample of matrix from the Subhyracodon site provided to the first author by the collector. The sample included abundant specimens of Chlamys cocoana, and a portion was sent to L. Edwards and J. Self-Trail of the USGS for microfossil analysis in early 2012. Results indicated that the matrix was from calcare- ous nannoplankton zone NP21 (L. Edwards and J. Self-Trail, pers. comm. to LBA, 2012). The pres- ence of the large brontothere in these sediments further supports the data from the above noted microfossils that the lower Harleyville Formation is in the uppermost Eocene part of NP21 rather than in the lowest Oligocene, because brontoth- eres are not known to persist into the Oligocene of North America (Prothero and Emry, 2004; Lourens et al., 2004:fig. 20.4; M. Mihlbachler, pers. comm. to LBA, 2012). The importance of the rhino and brontothere specimens cannot be overstated, as they represent the only remains of latest Eocene terrestrial mammals (late Chadronian NALMA) yet recovered from the USA Atlantic Coastal Plain (with the exception of a brontothere tooth fragment from the upper Eocene Clinchfield Formation of Georgia [Westgate, 2001]). As noted above, NP21 straddles the Eocene/ Oligocene boundary. The dramatically different lithology of the Parkers Ferry and Harleyville for- mations compared with the outer shelf limestones on which they rest (Tupelo Bay Formation) reflects the separation of deep water deposition from shal- lower water deposition purportedly due to activa- tion of the Gulf Trough during this time (Popenoe et al., 1987). Oligocene Series (33.9 – 23.03 Ma) Ashley Formation. Described as a “homo- geneous section of calcareous, microfossiliferous, silty and sandy clays” (Edwards et al., 1997:17), 3While this paper was in review, we learned that Mr. Metts passed away. We are unaware of any plans regarding the fu- ture of his extensive collection. ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 111 the Ashley Formation underlies most of Charles- ton, Berkeley, and Dorchester counties, and a por- tion of southeastern Colleton County; it thickens seaward and is approximately 30 m thick beneath Charleston (Bybell et al., 1998). Referred to as the “Ashley marl” by Tuomey (1848), Sloan (1908) separated the unit in some areas into “Ashley” and “Cooper” marls, based on typical outcrops along these two rivers, whereas in other regions he included the two “phases” together as a single “Ashley-Cooper marl.” As Weems et al. (2016) reviewed in detail, the stratigraphic and temporal placement of the Ashley and Cooper “phases” has been long debated, and the reader is referred to that reference for the history of the debate and its clari- fication (see additional discussion in Whitmore and Sanders, 1976). The best natural outcrop of the Ashley For- mation is exposed along the eastern bank of the Edisto River at Givhans Ferry State Park (Fig. 1B), and it is this section that Ward et al. (1979) desig- nated as the lectostratotype. The unit also crops out in a number of bluffs along the Ashley and Cooper rivers, and is broadly and cleanly exposed in sub- merged outcrops in the upper reaches of the Coo- per and Edisto rivers (LBA, pers. observ.). Long thought to be of Eocene age (e.g., Ruf- fin, 1843; Tuomey, 1848), the Ashley Formation is now known to have been deposited during the late Rupelian (late early Oligocene), with calcareous nannofossils indicative of zone NP24 (Hazel et al., 1977; Edwards et al., 1997, 2000). Supporting the late Rupelian age are 87Sr/86Sr dates derived from mollusc shells that average 29.1 Ma (Weems et al., 2016). Beneath Charleston, the Ashley Forma- tion apparently rests unconformably, for the most part, on the upper Eocene Parkers Ferry Forma- tion (NP19/20), as there are no units yielding nan- nofossils indicative of the intervening upper most Eocene and lower Oligocene zones NP21, 22, or 23 (Weems et al., 2016). The absence of strata har- boring microfossils characteristic of those zones is apparently due to strong currents flowing through the Gulf Trough that swept across this region and cut down into upper Eocene units during the Rupe- lian (Popenoe et al., 1987). Interesting, however, is the report by Brain- ard et al. (2009) of a 30-to-40 foot-thick “sand lens” in the Ashley Formation beneath Daniel Island and the Cooper River. This unit was discovered during exploratory drilling for construction of the Daniel Island Extension Tunnel, which extends from the Charleston peninsula beneath the river to the Dan- iel Island wastewater treatment plant. That stratum was originally encountered beneath Daniel Island at 88 ft below ground surface and subsequently found to slope southwesterly to a depth of 150 ft beneath the edge of the Charleston peninsula (Brainard et al., 2009:fig. 8). However, those depths are greater than those recorded by Weems and Lewis (2002) for the base of the Ashley Formation in that region. Weems et al. (2016:6) interpreted this “sand lens” as “a previously unrecognized subcrop area” per- haps equivalent to their Drayton Limestone Mem- ber of the Harleyville Formation, a unit they noted was distinctive in its abundance of medium-grained bryozoan fragments. Recent work by Weems et al. (2016) has resulted in the recognition that the Ashley For- mation can be divided into three members, each separated by quartz and phosphate-rich lag depos- its which also produce prominent gamma-ray log signatures. The lower member they designated as the Gettysville Member for the part of CCC1 that ranges from -158 to -63 feet. They noted that the unit consists primarily of foraminiferal sand with a silt-clay fraction considerably less than the overly- ing members. With foraminifera indicative of zone P20 of Blow (1969) and calcareous nannoplankton of NP24 age, the Gettysville Member is approxi- mately 29.5 myr old. Overlying the Gettysville Member, in the interval of CCC1 between -63 and -51 feet is the Runnymede Marl Member. The 29.1 Ma stron- tium date noted above was derived from an oyster shell collected near the base of this member, thus supporting the 29.5 Ma age of the underlying Get- tysville Member (Weems et al., 2016). The Run- nymede Marl Member, in turn, is overlain by the Givhans Ferry Member, which in CCC1 spans the -51 to -16 foot interval (Weems et al., 2016). The section noted above at Givhans Ferry State Park 112 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) is designated the stratotype for the Givhans Ferry Member. Foraminifera from this member are indic- ative of zone P21, and three 87Sr/86Sr dates derived from mollusc shells support this, as they range from 28.75 to 28.43 Ma (Weems et al., 2016). Although the Runnymede Marl and Givhans Ferry members are separated by a distinctive and widely recognizable lag deposit that marks an unconformity between the two units, their appear- ances are similar and their lithologies are only slightly different; the Givhans Ferry is more quartz- ose and phosphatic than the Runnymede Marl, which consists primarily of foraminifera. The lag consists of “calcite-quartz-phosphate sand, with a variable content of worn shells, phosphate pebbles, occasional quartz pebbles, and fossilized vertebrate material including sharks teeth, bones and teeth of cetaceans, and marine turtle shell fragments that vary in quality from pristine to worn” (see details in Weems et al., 2016). In the vicinity of Givhans Ferry State Park the lag deposit occurs along the underwater portion of the exposures; on the Ashley River it can be seen above the low-water level dur- ing low tide, but is typically draped by an apron of tidal “pluff mud” characteristic of the salt marsh environment along that portion of the river where the unit crops out. Given (1) the nature of this lag deposit as described above, (2) the more quartzose/ phosphatic lithology of the Givhans Ferry Member, and (3) the dramatically more nearshore/lagoonal/ estuarine lithological nature of the overlying Chan- dler Bridge Formation (see below), it is our opinion that the lag deposit separating the upper two mem- bers of the Ashley Formation represents the point of maximum transgression during the late Rupelian and the beginning of the regression that occurred across the Rupelian-Chattian boundary (Pälike et al., 2006). Over the last few decades the Ashley Forma- tion has yielded numerous skulls of archaic odonto- cetes (many still undescribed), as well as the most primitive of mysticetes. But discoveries of verte- brate remains from this unit have been a common occurrence since the first settlement of Charleston in 1670. Specimens were sent back to naturalists in London, such as Sir Hans Sloane and James Peti- ver, with the earliest mention of vertebrate fossils from SC in Petiver’s (1705) brief report of two fos- sil sharks’ teeth and “some Fossil bones, as ver- tebrae, &c” received from “our ingenious friend Mr Job Lord.” Joseph (“Job”) Lord was the pastor of the church at the small settlement of Dorches- ter on the Ashley River northwest of Charleston, now protected as Fort Dorchester State Park. He probably collected the fossils along the banks of the Ashley River and nearby Eagle Creek, where fossils can still be found over 300 years later. The Ashley Formation also has been deter- mined to be the unit of origin for the holotype of Agorophius pygmaeus (Müller, 1849) of the fam- ily Agorophiidae Abel, 1914, first thought to be an archaeocete and reported by Tuomey (1847) as a “cranium of the zeuglodon.” Collected by Charles- ton naturalist F. S. Holmes (with additional ele- ments of the holotype subsequently collected by L. R. Gibbes) from what Tuomey (1847:152) referred to as “the Eocene beds of Ashley River” sometime during the mid-1840s, the specimen has not been seen since 1869 and is now considered as irretriev- ably lost (Fordyce, 1981). Although Whitmore and Sanders (1976) were unable to locate the exact site of collection, it is known to be near “Greer’s Land- ing” near Middleton Place on the Ashley River, northwest of Charleston, according to Tuomey (1848:166), from what is now known to be the Oligocene, not Eocene, aged Ashley Formation. Fortunately, however, additional skulls of Ago- rophius have since been recovered through more recent exploratory efforts, including SC2015.51.1 collected by J. Osborne from the west bank of the Ashley River in the area of the type locality and ChM PV4256 from the bottom of the Edisto River (see Godfrey et al., 2016, and Boessenecker and Geisler, 2018). These two specimens, together with the beau- tiful lithograph in True (1907) of the lost cranium, provide important new information on the changes in skull morphology during ontogeny of this early odontocete. Whitmore and Sanders (1976) con- cluded from study of a series of Eosqualodon-like specimens from the Chandler Bridge Formation (see below) that the parietals are exposed dorsally in young individuals, but in later stages of ontog- eny they become progressively covered by forward ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 113 growth of the anterior margin of the supraoccipi- tal toward the vertex of the skull. This progression appears to exist in Agorophius pygmaeus, as well. In ChM PV4256, for example, the anterior termi- nus of the supraoccipital is nearly in contact with the fronto-parietal suture (Fig. 6). In SC2015.51.1, on the other hand, there are approximately 12 mm separating these two morphological landmarks. The now lost holotype shows an even greater degree of separation, about 27 mm (“1 in. 1 l[ine]”), based on measurements in Leidy (1869:423). It should be noted, however, that some measurements of the holotype are difficult to assess because the origi- nal figure in Tuomey (1847) is not well scaled, nor is the lithograph in True (1907). Godfrey et al. (2016:fig. 1) also noted the difficulty in accurately scaling the holotype. It is clearly evident, however, that the distance between the anterior terminus of the supraoccipital and the fronto-parietal suture does vary, and that these specimens likely repre- sent three different ontogenetic stages as A. pyg- maeus transitioned from sub-adult to adult (Fig. 6). Supporting our hypothesis is the interesting obser- vation by Tuomey (1847:153) that the holotype “was evidently a young individual,” exactly what would be expected given that it shows the greatest separation. Germane to this discussion is a recently pub- lished paper by Boessenecker and Geisler (2018) who described new remains of Agorophius from the Chandler Bridge Formation. In this study of material from both the Ashley and Chandler Bridge formations, they concluded that material from the latter unit (including ChM PV4256 noted above) might possibly represent a distinct species from the specimens known from the former, with the shorter exposure of the parietals on the vertex of the cra- nium serving as one of their diagnostic characters. Rather than naming a new species, they referred the Chandler Bridge material to Agorophius sp., “pending further study.” In 1851, Robert Gibbes, who in 1845 described Dorudon serratus, described and figured six teeth from “the Eocene of Ashley River, South Carolina” as a new “mosasauroid” to which he applied the name Conosaurus bowmani (Gibbes, 1851:pl. 3, figs. 1–9). The “Eocene of Ashley River” is, as noted above, the Ashley Formation, now known to be Oligocene in age, but in either case, it is much too young to have yielded mosa- saur teeth. Joseph Leidy (1848) set the matter to rights in concluding that Gibbes’ specimens were fish teeth, not mosasaur, and establishing the name Conosaurops to replace Conosaurus. Even earlier, Agassiz (1848) applied the name Saurocetus gibbesii to a cetacean tooth (MCZ 8760) from the Charleston area that is presently of undetermined familial affinity. Subsequently, Leidy (1853) described a new odontocete, Colophonodon holmesii, from a partial tooth and fragments of five others, but that taxon is regarded as a nomen dubium by Fordyce and De Muizon (2001). The Ashley Formation has also yielded the holotype skull (USNM 11049) of Xenorophus sloa- nii Kellogg, 1923, upon which the family Xenoro- phidae was established by Uhen (2008). Xenoro- phids are archaic forms that have a land mammal- like braincase with a prominent sagittal crest (see Whitmore and Sanders, 1976:fig. 1a). Additional species of xenorophids from the Ashley Forma- tion include Albertocetus meffordorum Uhen, 2008, (see Boessenecker et al., 2017a) and the apparently suction-feeding Inermorostrum xenops Boessenecker et al., 2017b. The early baleen whale Micromysticetus rothauseni Sanders and Barnes, 2002a, (ChM PV4844) was also recovered from the Ashley Formation, as were at least three species of toothed mysticetes with archaeocete-like teeth in an archaeocete dental formula, such as the recently described Coronodon havensteini Geisler et al., 2017 (with additional material noted by Geisler et al., 2018). Of particular importance is the holotype partial skull of Ashleycetus planicapitis Sanders and Geisler, 2015, one the most primitive odonto- cetes yet reported. Two other skulls (ChM PV4824 and PV7679) represent undescribed taxa that some- what resemble the holotype skull of the odontocete Eosqualodon langewieschei Rothausen, 1968, from the Chattian age Doberg Formation of the North Sea Basin, Germany, and Eosqualodon-like skulls from the unit overlying the Ashley Formation, the Chandler Bridge Formation (discussed below). Another skull (SC2015.33.1) recovered from the Givhans Ferry Member was recently described 114 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 115 by Albright et al. (2018). Ediscetus osbornei super- ficially resembles Waipatia maerewhenua Fordyce, 1994, from the Oligocene of New Zealand. How- ever, although geologically older than the New Zealand taxon, the SC species differs in having a significantly more advanced degree of telescop- ing, whereby the posterior frontomaxillary suture (posterior temporal crest) has progressed so far posteriorly that it has been pushed up vertically in its broad contact with the supraoccipital. This has resulted in a complete override of the parietals, a condition much advanced relative to contemporary species found in the same strata such as Xenoro- phus, Agorophius, and Ashleycetus. In addition to a diverse cetacean fauna, the Ashley Formation has also yielded an abundance of other vertebrate fossils. Sirenians are represented by Priscosiren atlantica Vélez-Juarbe and Domn- ing, 2014a, Stegosiren macei Domning and Beatty, 2019, and Crenatosiren olseni (Reinhart 1976) (see Domning, 1997;Vélez-Juarbe and Domning, 2014a, b). A single partial tooth (ChM PV9480) from the Ashley Formation may be referable to Metaxythe- rium albifontanum Vélez-Juarbe and Domning, 2014b, but all other known material of this taxon from SC is from the Chandler Bridge Formation (see discussion below on the Chandler Bridge Formation and in the “Systematic Paleontology” section). As noted by Vélez-Juarbe and Domning (2014a:951), “the early Oligocene is an important time period in sirenian evolution, because it is a transitional period between the extinction of prorastomids and protosi- renids by the end of the Eocene and the appearance of more derived sirenian lineages in the late Oligo- cene (Vélez-Juarbe, 2014).” Koretsky and Sanders (2002:17) reported the recovery of the proximal portion of a femur of a pinniped (ChM PV5713) from the Ashley Forma- tion near Summerville, in Dorchester County, not- ing that the specimen was “similar in size to those of the modern Harp seal, Pagophilus groenlandi- cus, of the family Phocidae.” Although this would appear to be the oldest known record of a true seal in the Atlantic Ocean basin, and one of the old- est records of a seal anywhere, the biogeographic analysis of pinnipeds by Deméré et al. (2003) calls into question this record and suggests that the prov- enance of the specimen may be suspect. Four additional specimens, first noted in the 19th century, are also of interest (Fig. 7). In 1856 Joseph Leidy described Phoca debilis on the basis of three teeth (now catalogued as ANSP 10322, 10324 and 10325) “from the sands of the Ashley River, South Carolina” (Leidy, 1869:415). Subse- quently, Leidy (1869) removed ANSP 10322 from the P. debilis type material and combined it with another tooth (now ANSP 10323) also from “the Ashley River deposits, South Carolina” as rep- resentative of another new species, Phoca mod- esta (Leidy, 1869:415), evidently because of the greater similarity of the crown morphology of these two teeth to each other rather than to either ANSP 10324 or 10325. Cope (1867:144) assigned P. debilis to the odontocete genus Squalodon, but Leidy’s debilis teeth are considerably smaller than those of presently recognized species of Squalodon (e.g., Squalodon calvertensis Kellogg, 1923). Not- ing Cope’s assignment, Leidy (1869:415) observed that such “may be the case, or perhaps they may belong to a Dolphin.” Kellogg (1923:13) question- Figure 6. Comparison of skulls of Agorophius pygmaeus showing inferred ontogenetic change in distance between anteriormost point of supraoccipital and fronto-parietal suture (juvenile to adult from top to bottom). Regarding scale: because the holotype is lost, any scale for the Agassiz lithograph published in True (1907) must necessarily rely on the measurements of that specimen as published in Tuomey (1847), Leidy (1869), and True (1907). Two measurements in particular can be used to provide a reasonable approximation of the size of the type. One of these is Tuomey’s (1847:153) “greatest breadth [of skull] 7 ½ in.” also reiterated in True (1907:5). But a less ambiguous measurement, and the one used here to scale the holotype, and, in turn, ChM PV4256, is Leidy’s (1869:423) “Breadth of skull at postorbitals, 7 in.” 116 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) ANSP 10323, he noted that “It seems more likely, from what is now known concerning the various described species of Squalodon and other Miocene cetaceans, that this tooth belongs to some pinni- ped.” In summary, the stratigraphic origin of all four of these teeth is uncertain, there is no evidence that any of them came from the Ashley Formation, they may be representative of more than two differ- ent taxa, and they may not be definitively represen- tative of pinnipeds! The first known representatives of a west- ern North Atlantic Oligocene avifauna are now known from the Ashley Formation, as well. These specimens, in collections at both The Charleston Museum and the SCSM, were first studied by R. Chandler who concluded that several new spe- cies are represented. Preliminary results of a more recent study by D. P. Ksepka (pers. comm. to AES, September, 2012) indicate the presence of an alba- tross (Plotornis sp.), an indeterminate genus of petrel (Procellariidae), and two species possibly representing new genera of gannets and boobies (Sulidae). Crocodilian remains from the Ashley Forma- tion consist of only three specimens, all of which are assigned to Gavialosuchus carolinensis Erick- son and Sawyer, 1996, the holotype (ChM PV4279) and paratype (SC90.93.1) of which came from the overlying Chandler Bridge Formation. Two are well-preserved associated dentaries excavated from the bottom of Dorchester Creek in Dorchester County (ChM PV4282), and the other specimen is a cranial fragment from Charleston County (ChM PV4280). Sea turtles are represented by (1) Caro- linochelys wilsoni Hay, 1923, the holotype skull of which is MCZ 1005-A; (2) Ashleychelys palmeri Weems and Sanders, 2014, the holotype of which includes a skull, shell, and limb elements (ChM PV7002), another skull (SC90.19), and additional elements of a juvenile (ChM PV6869 and PV7202); (3) Procolpochelys charlestonensis Weems and Sanders, 2014, represented by a humerus (MCZ 1005-B) originally referred to Carolinochelys wil- soni by Hay (1923), and (4) an undescribed new taxon of leatherback sea turtle (family Dermoche- lyidae) collected by JLK, DJC, V. McCollum, and Figure 7. Phoca debilis Leidy, 1856: A, ANSP 10324 and B, ANSP 10325, in lingual view; C and D, same teeth, respectively, in labial view. Phoca modesta Leidy, 1869: E, ANSP 10323 and F, ANSP 10322, in lingual view; G and H, same teeth, respectively, in labial view. (ANSP 10324 measures 21.6 mm; ANSP 10322 measures 11.1 mm). ably referred P. debilis to “Delphinodon?”, a Mio- cene odontocete genus, cautiously noting that the teeth “do not sufficiently agree with those of Del- phinodon to be definitely referred to that genus,” but that “it seems probable … that they represent some Miocene delphinid.” He then gave the strati- graphic origin of the specimens as “Edisto marl or Upper Miocene,” though he did not cite the source of that information. Both his systematic and strati- graphic referrals are problematic because upper Miocene deposits are not known from the banks of the Ashley River (e.g., Weems et al., 2016). Doubt- ful of its validity as a species of phocid, Kellogg (1923:26) also referred Leidy’s Phoca modesta to Phoca? modesta (ANSP 10322, 10323). Regarding ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 117 several volunteers (SC2007.36.1). Fierstine and Weems (2009) studied the bill- fish remains from the Ashley Formation and these include specimens of Aglyptorhynchus robustus, Aglyptorhynchus sp., the holotype material of A. palmeri and A. alsandersi, and additional speci- mens of uncertain specific allocation referred to Xiphiorhynchus. They placed both of these genera into a new subfamily, Aglyptorhychinae. Although this work focuses on the vertebrate biostratigraphy of the state, the biostratigraphic utility of molluscs prompts a discussion of them, as well. Appendix 4, therefore, provides a summary of the molluscs currently known from the Ashley Formation based primarily on recent, extensive, studies conducted by Dr. Matthew Campbell on specimens from the unit in the collections at The Charleston Museum and the SCSM. In his updated assessment he noted that calcitic taxa were pre- served as the original shell, but aragonitic taxa are preserved as internal or external molds resulting in identification only to the genus level. Chandler Bridge Formation. The Chandler Bridge Formation was named by Sanders et al. (1982) for a thin sequence of noncalcareous, arena- ceous beds that unconformably overlies the Ashley Formation with a patchy distribution in Charles- ton, Berkeley, and Dorchester counties. The unit is typically less than 2 m thick, with a maximum known thickness of about 5 m, and it was originally divided into three conformable beds numbered 1–3 in stratigraphically ascending order (Sanders et al., 1982; Weems and Sanders, 1986; Fig. 8A); a fourth bed, 1A, has been additionally recognized since original description of the unit (Fig. 8B; fur- ther discussion below). The stratotype section of the formation was described from the east side of a 21 m by 21 m (70 ft x 70 ft) pit excavated in 1970 as part of a study to recover Oligocene marine ver- tebrates from this highly fossiliferous unit (Sand- ers, 1980; Sanders et al., 1982; Fig. 9A). The pit has since been filled, but additional exposures have been described (e.g., Katuna et al., 1997; Fig. 9B). The Charleston Museum excavation was prompted by the recovery of a nearly complete skull of an unknown, albeit obviously Oligocene- aged, odontocete by an avocational fossil collec- tor in 1969. The specimen was recovered from the bank of a ditch on the north side of Ladson Road (County Road S-18-230) near Chandler Bridge Creek in Dorchester County about 23 km north- west of Charleston (Fig. 10) and donated to The Charleston Museum by Mr. R. Lambert of Sum- merville. It was the scarcity of Oligocene-aged marine vertebrate-bearing strata throughout the world that resulted in the second author’s determi- nation to excavate these beds, with funding granted under the auspices of The Charleston Museum by the Charleston Scientific and Cultural Educa- tion Fund in 1970 and by the National Geographic Society in 1971 and 1972. Like the unconformably underlying Ashley Formation, the Chandler Bridge Formation was also found to harbor a highly diverse, well-pre- served cetacean fauna. The excavation was highly successful, yielding 17 partial skeletons represent- ing six undescribed odontocete taxa recovered over three summer field seasons by an excellent crew of students from the College of Charleston. An account of the excavation methods, results, and significance of that undertaking was provided by Sanders (1980) and summarized with addi- tional results in Sanders et al. (1982). It should be noted that the “archaeocetes” mentioned in Sand- ers (1980:620) and in Sanders et al. (1982:118) as having come from the Chandler Bridge Formation have since been found to be toothed mysticetes, which have an archaeocete-like dentition. Weems and Sanders (1986) considered the Chandler Bridge Formation as representing a shal- low marine transgressive sequence, with basal Bed 1 considered indicative of an estuarine/lagoonal environment and the uppermost Bed 3 representa- tive of a beach face/shallow water marine environ- ment. The abundance of well-preserved cetacean skeletons in Bed 3 at the excavation pit, due possi- bly to stranding events, supported their beach face environmental characterization. Additional support was provided by the identification of marine mol- lusc taxa from numerous internal casts that point toward a similar environment. These include the gastropods Xenophora sp. and Apiocypraea sp., plus the bivalves Glycymeris sp., Chlamys sp., Astarte sp., Nemocardium sp., Gastrochaena sp., 118 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) Eburneopecten calvatus, and three species of Ven- eridae (M. Campbell, pers. comm. to AES, July 2015). Only Bed 3 was excavated over the entire 441 sq-m study area. Bed 2 also yielded an odonto- cete skull and postcranial material, but the presence of Callianassa-like burrows suggests a lagoonal depositional environment for this unit. Additional molluscs from the Chandler Bridge Formation include the bivalves Barbatia sp., Pecten sp., Cyc- locardia sp., Cardium sp., and the gastropod Epito- nium chamberlaini (M. Campbell, pers. comm. to LBA, August 2019) Bed 1A, as noted above, was not included in the original description of the formation. This bed was first encountered in 1981 when a skull and partial skeleton of the primitive odontocete Ago- rophius pygmaeus (ChM PV4256 noted above) was collected by three scuba divers from sandy, grayish-green sediments at the bottom of the Edisto River, the recovery at which AES was present. At a different site in the Edisto River an undescribed species of another archaic odontocete, Xenoro- phus (ChM PV4823), was recovered from similar sediments. Both Agorophius and Xenorophus were known previously only from the Ashley Forma- tion; but the sediments at the bottom of the Edisto Figure 8. A, Type section of the Chandler Bridge Formation resting disconformably on the Ashley Formation. Bottom of each “Zone” card marks the base of that bed. Described and measured on the east wall of a paleontological excavation pit (Fig. 9A) opened by The Charleston Museum in 1970 adjacent to Chandler Bridge Creek, 0.7 km NW of the confluence with Eagle Creek, in the NE quarter of the Stallsville 7.5' USGS quadrangle, Dorchester County (see Fig. 10); reproduced from Sanders et al. (1982:fig. 26). B, Diagrammatic cross-section of the Chandler Bridge Formation as exposed along a drainage ditch (Limehouse Branch) in College Park development (between Goose Creek and Summerville), Berkeley County (see LAT, LON in figure), showing stratigraphic relationship of Bed 1 and 1A. ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 119 Figure 9. A, View of Charleston Museum’s 1971 Chandler Bridge excavation pit. B, Redraft of section of the Chandler Bridge Formation measured and described by Katuna et al. (1997:fig. 7) reported at 32°59.57' N, 80°03.18' W (their site CH2). According to the Geologic Map of the Ladson Quadrangle (Weems and Lemon, 1988), the Chandler Bridge Formation crops out near this locality along the NNW- SSE trending border between Charleston and Berkeley counties. This site is approximately 4.36 km SE of the College Park cross-section locality of Figure 8B; B1, B2, and B3 refer to Beds 1, 2, and 3 of Weems and Sanders (1986). 120 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) at this locality yielded the holotype of the xenoro- phid Echovenator sandersi Churchill et al., 2016. Another xenorophid skull identified as Albertoce- tus sp. (ChM PV8680; M. Gibson, pers. comm. to LBA, June 2019) was collected at this locality, but from the underlying Ashley Formation. Billfish material is also known from Bed 1A, as are mol- luscs including the gastropods Epitonium charles- tonensis and E. cooperensis, and the bivalves Pycnodonte sp. cf. P. paroxis, Gastrochaena sp., and apparently three representatives of the pectinid genus Chlamys (these molluscs were collected by R. Patterson in July 2012, and identified by M. Campbell). In contrast to the transgressive interpretation of the unit by Weems and Sanders (1986), Katuna et al. (1997) concluded that the overall trend within the unit was that of a shallowing upward, and coars- ening-upward, regressive sequence with sediment provided from fluvial systems to the west. As sum- marized by Cicimurri and Knight (2009b), Katuna et al. (1997) recognized four sedimentary facies within the formation (Fig. 9B): (1) a basal marine facies rich in sharks teeth and well-preserved neritic nearshore marine dinoflagellates below Bed 1 of Weems and Sanders (1986) and possibly equivalent to Bed 1A (but yet to be confirmed); (2) a marginal marine facies equivalent to Bed 1 of Weems and Sanders (1986) within which acorns and hickory nuts were noted (Sanders et al., 1982); (3) a bay/ estuarine facies indicative of a restricted brackish bay or lagoonal environment with access to open ocean from which cetacean and sea turtle remains were recovered, equivalent to Bed 2 of Weems and Sanders (1986); and (4) an upper fluvial/estuarine facies lacking dinoflagellates but containing fresh- water pollen, discoidal quartz and phosphate clasts, angular quartz sand grains, and a concentration of fine grained sediments (“40% silt and clay by weight” Katuna et al., 1997:191), but also abundant cetacean remains, equivalent to Bed 3 of Weems and Sanders (1986). Further support for the regres- sive interpretation are the findings of Cicimurri and Knight (2009b:644) who concluded on the basis of the species of sharks, rays, ostracods, foraminifera, and corals collected from the lower marine facies Figure 10. Map of area in Dorchester County showing location of Charleston Museum excavation site and spatial relationship of Chandler Bridge Creek and Eagle Creek with the Ashley River. River from which these two specimens were recov- ered were not of typical Ashley lithology, in turn suggesting the probability that they were from an undescribed facies of the Chandler Bridge Forma- tion, as they did not resemble the latest Oligocene Edisto Formation either. That suspicion was subsequently verified with the discovery of an outcrop of the greenish sediments in the banks of Limehouse Branch in the College Park subdivision of North Charleston, Berkeley County, approximately 32 km (20 miles) northeast of the Edisto River locality where it was first encountered. One of the College Park locali- ties is about 240 m north of the Trinity Place Street bridge (33º 1.122’N, 80º 5.641’W), where the greenish stratum, Bed 1A, unconformably over- lies the Ashley Formation and fills an Oligocene channel eroded into the Ashley (Fig. 8B). Bed 1A is overlain by Bed 2, which in turn is overlain by upper Pleistocene sediments of the Wando Forma- tion; Bed 3 was apparently scoured away. Immedi- ately south of the bridge are two short, thin depos- its of Bed 1 in the same stratigraphic relationship as is Bed 1A, between the Ashley Formation and Bed 2 (Fig. 8B); but whether Beds 1 and 1A are directly correlative has not yet been demonstrated unequivocally. Another exposure of Bed 1A was found in the bank of a small tributary of Limehouse Branch near the Sangaree Middle School, about 1.35 km WNW of the Trinity Place bridge. Bed 1A ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 121 that it was deposited in a “shallow inner to middle neritic environment where surface water tempera- tures were between 20° and 25° C.” Added to this is the generally regressive nature of the Chattian (GTS2012:fig. 28.1). Subse- quent to opening of the Drake Passage and devel- opment of the Antarctic Circum-Polar Current around the time of the Eocene-Oligocene boundary (e.g., Miller et al., 1991, 2005; Vandenberghe et al., 2012), the Oligocene experienced several globally recognized cooling events related to eccentricity and obliquity cycles (Pälike et al., 2006). The Oi2* cooling event at 29.16 Ma prior to the Rupelian- Chattian boundary, as well as Oi2a shortly after at 27.91 Ma and Oi2b at 26.76 Ma (Fig. 3A), are associated with sea level falls estimated to be on the order of 50 to 65 meters (Wade and Pälike, 2004). However, starting at about 26 Ma there is a middle to late Chattian warming trend (Raffi et al., 2006:fig. 5). Although still punctuated by cooling events such as Oi2c in magnetochron C7 and end- ing with the Mi1 glaciation event at the Oligocene- Miocene boundary, it is within this warm interval that the Chandler Bridge Formation was likely deposited. We investigate this further below. The sediments that would eventually be described as the Chandler Bridge Formation by Sanders et al. (1982) were originally inferred to be Oligocene in age by Whitmore and Sanders (1976) on the basis of a skull from the ChM excavation first thought to be referable to Xenorophus sloanii, the holotype of which is from the Ashley Forma- tion, then regarded as upper Oligocene. This sug- gested that the Chandler Bridge Formation was not much younger than the Ashley Formation. But upon preparation of that skull, it proved to be an undescribed relative of X. sloanii. Subsequent estimates, as noted in Weems et al. (2006, 2016), resulted in an early Chattian age based on the simi- larity of some of the cetaceans from the Chandler Bridge to one from the lower Chattian sequence (“Chattian A”) in the Doberg Formation of north- western Germany. This was based primarily on comparisons by AES of three undescribed odon- tocete skulls from the Chandler Bridge excavation (ChM PV2753, 2754, 2755) with the holotype skull of Eosqualodon langewieschei. One of those skulls (PV2753), informally designated as “Genus X” by Whitmore and Sanders (1976), displays the same morphology of the skull roof as that of Eosqualo- don, i.e., the parietals are still present in the skull roof but are concealed medially by a forward thrust of the supraoccipital and are exposed only as small triangular-shaped remnants at the outer margins of the intertemporal constriction of the skull roof. That morphology represents a grade of odontocete evolution that had not been recognized previously. Noted, however, was the much smaller size of the Chandler Bridge skulls, clearly representing differ- ent taxa than Eosqualodon from the Doberg For- mation. Supporting the original Chandler Bridge- Chattian A correlation was the work of De Man et al. (2010) who reported 87Sr/86Sr dates from benthic foraminifera (27.2 ± 0.7, 27.5 ± 0.7 Ma) and a K/ Ar radioisotopic date from glauconite (27.0 ± 0.3 Ma) that placed the age of their lower Chattian (= Chattian A) of the Doberg Formation at ~27 Ma. Additionally, the Chattian A sequence lies within a correlate of calcareous nannoplankton zone NP24 (De Man et al., 2010; also see Anderson, 1961; Anderson et al., 1971; and Curry et al., 1978). Those determinations provided a time frame for the occurrence of the evolutionary grade represented by Eosqualodon and the morphologically similar specimens from South Carolina. Accordingly, the age of the Chandler Bridge was placed at approxi- mately 28 Ma (Sanders et al., 1982; Weems and Lemon, 1984a, b; Sanders and Barnes, 2002b). Recently, however, calcareous nannoplank- ton from the Chandler Bridge Formation defini- tively place the unit within zone NP25. Jean Self- Trail (pers. comm. to REW, 2014; also see Weems et al., 2016) noted the presence of Sphenolithus ciperoensis and Triquetrorhabdulus carinatus in samples from the formation. The former’s LAD occurs at the top of NP25, and the latter’s FAD occurs at the base of NP25. Further support for an NP25 assignment includes the absence in these samples of S. distentus, which last occurs at the top of NP24. These data securely bracket the age of the Chandler Bridge Formation to between 26.84 and 122 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) 23.13 Ma (GTS2012), i.e., slightly younger than “Chattian A” of the Doberg Formation in Germany. De Man et al.’s (2010) middle and upper Chattian sequences (Chattian B and C) correlate with NP25. Additional data bearing on the age of the Chandler Bridge Formation include 87Sr/86Sr dates provided by W. B. Harris from oyster shells, reported herein as Pycnodonte sp. cf. P. paroxis, collected from the same locality as were the calcareous nannoplank- ton samples (Weems et al., 2016). Analysis of these specimens yielded an average date of 24.7 ± 0.2 Ma, which falls within the ages bracketing NP25 noted above. Together these data provide an age for the Chandler Bridge Formation younger than that implied by the evolutionary grade of the cetacean fauna, and correlative with the Chattian B and C sequence at Doberg rather than Chattian A. This indicates that the unconformity between the Ash- ley and Chandler Bridge formations is apparently longer than originally considered. Based on all of the data above we revise the age of the Chandler Bridge Formation to about 25 Ma (Figs. 2A, 3A; also see following discussion on biostratigraphic utility of fossil cetaceans). Katuna et al. (1997) placed the Chandler Bridge Formation in the uppermost Chattian adjoining the boundary of the Aquitanian Stage of the lower Miocene, and as a lateral equivalent of the Upper SC Coastal Plain Upland Unit. At the time of their publication the Chattian/Aquitanian boundary was placed at about 23.8 Ma (following the time scale of Berggren et al., 1995). The revised estimate of this boundary at 23.03 Ma (GTS2012) maintains a middle-to-late Chattian age for the formation, but positions it well below the Chat- tian/Aquitanian Stage boundary. Additionally, the Upland Unit has since been shown to be no older than late middle Miocene in age by Weems and Edwards (2007a), and therefore not a stratigraphic correlative of the Chandler Bridge Formation (also see Nystrom et al., 1991). In addition to those cetaceans noted above, the Chandler Bridge Formation has also yielded two toothed mysticetes, ChM PV2778 and PV5720, both from Bed 3 at sites in Charleston and Berkeley counties, respectively. The smaller of the two, ChM PV2778, apparently belongs to the same taxon as a toothed mysticete from the Ashley Formation represented by ChM PV4745, thought to be a juvenile of Coronodon havensteini (see Sanders and Geisler, 2015:fig. 15; Geisler et al., 2017; and Hocking et al., 2017; NOTE: Hocking et al., 2017, incorrectly noted this specimen as ChM PV4645). Though not yet described, ChM PV5720 was found to share a sister taxon relationship with Coronodon by Geisler et al. (2017). In addition to having archaeocete-like teeth and dental formula, they all have other characters that define them as basal mysticetes comprising a previously unknown family, which appears to support the hypothesis that the suborder Mysticeti (the baleen whales) evolved from the basilosaurid subfamily Dorudon- tinae (Barnes and Sanders, 1996). The nearly 2 m-long skull suggests an overall length that may have approached 7.5 m, the largest of any cetacean yet recovered from the Chandler Bridge and/or Ashley formations. The Chandler Bridge Formation is also the source of the holotypes of the most primitive known baleen-bearing whales, Eomysticetus whitmorei and E. carolinensis Sanders and Barnes, 2002b, for which those authors erected the new family Eomysticetidae and the superfamily Eomysticetoi- dea. The former was recovered from Bed 3 and the latter from Bed 2 at separate sites in Dorchester County. The most common odontocetes from the for- mation are members of the family Xenorophidae. In a detailed examination of xenorophid skulls in the collections of the Charleston Museum, J. Geisler and AES found several that appear representative of taxa other than Xenorophus. For example, one yet-to-be described skull, ChM PV2775, represents a grade similar to that of Albertocetus meffordo- rum described from a partial skull (USNM 525001) from the upper Oligocene Belgrade Formation at Onslow Beach, NC (Uhen, 2008). In this taxon the anterior wall of the braincase is situated farther forward than in Xenorophus such that it extends beyond and below the level of the posterior mar- gin of the supraoccipital processes of the frontals ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 123 (Sanders, 1996). Strontium isotope analysis by W. B. Harris from mollusc shells adhering to the NC specimen yielded a date of about 26.5 Ma (Uhen, 2008). Two additional xenorophids from the Chan- dler Bridge include the early echo-locating Cotylo- cara macei Geisler et al., 2014, from Bed 2, and the above noted Echovenator sandersi from Bed 1A. Another grade of odontocete evolution is exemplified in Agorophius pygmaeus. The holo- type of this taxon was collected from the Ashley Formation, as noted above in the discussion of that unit, but it disappeared from collector F. S. Holmes’ private collection sometime during the late 19th century and has never been relocated (see Fordyce, 1981). In this evolutionary grade, the braincase is separated from the rostrum by a narrow, tabular intertemporal constriction formed by the parietals. Additional specimens from the Chandler Bridge Formation support what Boessenecker and Geisler (2018) considered to be a second species of the genus, described and referred to as Agorophius sp. Also from the Chandler Bridge is ChM PV4753, a small skull about 350 mm in length that represents the first North American specimen of the family Patriocetidae Abel, 1913. Only two species have been described, Patriocetus ehrlichi (Van Beneden, 1865) from upper Oligocene sands on the south side of the Danube River at Linz, Austria, and Patriocetus kazakhstanicus Dubrovo and Sanders, 2000, from the Karaginskaya For- mation of Kazakhstan. A third species, from the upper Oligocene Grafenberg Formation near Dus- seldorf, Germany, is currently under description by K. Rothausen and AES. In these taxa the parietals are exposed as a narrow band across the skull roof and the zygomatic processes are “pistol-shaped” (Dubrovo and Sanders, 2000). A similar cranial morphology, excluding the “pistol-shaped” zygo- matic process, is seen in skulls (e.g., ChM PV4755) that resemble Waipatia, originally described from the upper Oligocene of New Zealand (Fordyce, 1994), and which are currently being referred to as “waipatiids” until detailed study more accurately reveals their relationships. More derived cetaceans also appear to have been relatively common in coastal Chandler Bridge seas, including “Genus X” of Whitmore and Sand- ers (1976:figs. 5-8), exemplified by ChM PV2753 and discussed above as apparent relatives of Eosqualodon from the Doberg Formation of Ger- many. As in Patriocetus, the parietals in these taxa form a narrow rectangle in the skull roof, but are covered medially by the apex of the supraoccipital and are visible dorsally only as small triangles at the outer margins of the intertemporal constriction. The same morphology occurs in the much larger “Genus Y” (e.g., ChM PV2757) of Whitmore and Sanders (1976:figs. 4a, b), the largest odonto- cete yet known from either the Chandler Bridge or Ashley formations. Recovered during the 1970 ChM excavation, the axial skeleton consists of a complete skull and vertebral column, missing only the last two caudal vertebrae, with a total length of about 5.5 m. A mounted cast of ChM PV2757 is on display at The Charleston Museum, and additional specimens considered representative of “Genus Y” are housed in collections at the Mace Brown Museum of Natural History (see Boessenecker et al., 2018). Dooley (2003, 2005), who recognized only two (possibly four) valid North American spe- cies of the genus Squalodon, including S. calver- tensis Kellogg, 1923, S. whitmorei Dooley, 2005, and possibly “Squalodon” atlanticus (Leidy, 1856) and “Squalodon” tiedemani Allen, 1887, concluded that “Genus Y” closely resembles “Squalodon” tie- demani, but that neither likely represent that genus (i.e., that “Genus Y” and “Squalodon” tiedemani are likely not squalodontids; Boessenecker et al. [2018] referred to “Genus Y” as an unnamed “Ago- rophius-like giant dolphin”). Recovered during phosphate dredging operations in the Wando River, the type specimen of “S.” tiedemani has matrix adhering to it that resembles Ashley Formation, although the indurated nature of this material sug- gests to some (REW) that it may possibly represent Edisto Formation. Additional material of “Genus Y” from the Ashley Formation, however, further supports derivation of the type from the Ashley (Boessenecker et al., 2018). Regarding Squalodon, a specimen much smaller than “Genus Y” but with teeth closely resembling those of “Squalodon”atlanticus was 124 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) collected by the first author in 1973 from a spoil pile of Chandler Bridge sediment during dredg- ing operations of the NW-SE trending canal 1.15 km NE of the intersection of SC Highway 17 and Bees Ferry Road (State Road S-10-57), Charles- ton County. Further study of this specimen (ChM GPV3) is required to determine more accurately its taxonomic affinity, particularly considering that the genus Squalodon is primarily an early to middle Miocene, rather than Oligocene, form. Another grade of odontocete cranial evolu- tion is represented in ChM PV4802, a well-pre- served skull missing most of the rostrum, from Bed 2 in Berkeley County. Although in general appear- ance it resembles members of the long-snouted odontocete family Eurhinodelphinidae, such as “Rhabdosteus latiradix”4 from the lower and mid- dle Miocene of Virginia and Maryland (and pos- sibly Georgia and SC; see section below on Marks Head Formation), Geisler et al. (2011) found no relationship of this specimen with this family. Their analysis placed ChM PV4802 near the base of crown group Odontoceti. From the above discussion, it is obvious that the morphological and taxonomic diversity of whales from the Chandler Bridge Formation is truly astounding, with new specimens represent- ing even more taxa being regularly added to col- lections as the cohort of avocational collectors in the region explore previously and newly exposed outcrops. Whereas The Charleston Museum and the Mace Brown Museum of Natural History house the bulk of the cetacean material from the Chan- dler Bridge Formation, the collections at the South Carolina State Museum include a number of speci- mens, as well. In addition to the great diversity of cetaceans, however, are other marine mammals from this unit, one of which may be another very early record 4De Muizon (1988) considered Rhabdosteus latiradix Cope, 1868, as incertae sedis because of the incomplete nature of the type specimen. He also considered specimens from the Calvert Formation of Maryland referred to Rhabdosteus by Myrick (1979) congeneric with the holotype of Schizodelphis sulcatus Gervais, 1853, from France. Hence, the quotation marks, “Rhabdosteus latiradix” (see Lambert, 2004, for ad- ditional discussion). of a pinniped. In their description of the purport- edly oldest known record of a true seal from the Ashley Formation, Koretsky and Sanders (2002) also included a proximal femur from the Chandler Bridge Formation (ChM PV5712). Although this specimen, like that noted above from the Ashley Formation, is at odds with the biogeographic find- ings of Deméré et al. (2003), thus calling into ques- tion the accuracy of the stated provenance, the col- lector of the specimen, Mr. V. McCollum, vividly recalls the precise circumstances of its recovery from the Chandler Bridge Formation (V. McCol- lum, pers. comm. to LBA, January 2017). Sirenians are somewhat better represented. Several specimens in the ChM and SCSM collec- tions represent Crenatosiren olseni, Dioplotherium manigaulti Cope, 1883, and the oldest known spe- cies of the genus Metaxytherium, M. albifontanum Vélez-Juarbe and Domning, 2014b (Domning, 1989a, b, 1997; Vélez-Juarbe et al., 2012; Vélez- Juarbe and Domning, 2014a, b; M. albifontanum may also occur in the Ashley Formation [see dis- cussion in “Systematic Paleontology” section]). Knowledge of marine bird life along the Atlantic Coastal Plain during Oligocene time comes not only from the Ashley Formation, as noted above, but also from specimens recovered from the Chandler Bridge Formation. These, like those from the Ashley, are currently under study by D. P. Ksepka and include four new species within the family Sulidae (boobies and gannets) and two new species within Procellariidae (petrels). The most impressive specimen in the entire avian assem- blage from this unit is the holotype partial skeleton (ChM PV4768) of the enormous pseudodontorn Pelagornis sandersi Ksepka, 2014, a spectacular albatross-like bird with bony tooth-like projections and a wingspan of about 6.4 m. A life-size replica of the specimen, collected from Bed 2 near the Charleston Airport by AES and avocational fossil collector J. Malcolm in February, 1983, is on dis- play at The Charleston Museum (Fig. 11) together with the partial remains of another smaller speci- men (ChM PV4801) that may represent a new, but not-yet-described species. Reptilian taxa from the Chandler Bridge con- ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 125 sists of only one crocodilian, but several taxa of turtles. Specimens of Gavialosuchus carolinensis were noted above in the section on the Ashley For- mation, but the holotype skull and skeleton, ChM PV4279, was recovered in 1978 from the Chandler Bridge Formation in what is now the Trailwood subdivision off Dorchester Road in Charleston County. A mounted cast of the specimen measuring about 5.4 m long is on display at The Charleston Museum, and an isolated femur (ChM PV4281) from the ChM excavation is from an even larger individual. The paratype skull and partial skeleton, SC90.93.1, was also recovered from exposures of the Chandler Bridge Formation at the bottom of a lake that was being excavated in the Crowfield Plantation subdivision of Berkeley County. Addi- tional material of G. carolinensis is known from Dorchester County and includes two associated vertebrae (ChM PV4283) and three dissociated vertebrae (ChM PV4284, 4285, 4286). Turtle remains include the anterior half of a carapace (ChM PV7180) of Gopherus sp., the only terrestrial taxon yet recovered from the Chan- dler Bridge Formation (Franz, 2014), plus the first North American occurrence of an Oligocene side- necked turtle (family Podocnemididae), the fluvial to estuarine Bairdemys healeyorum (Weems and Knight, 2013). The latter is represented by two specimens, both from Dorchester County: SC90.16, the holotype which includes skull, shell, and skel- etal elements, and ChM PV4794, a referred speci- men consisting of a partial carapace from Bed 2. Weems and Knight (2013) also mentioned a speci- men of soft-shelled turtle (ChM PV4882), prob- ably Apalone. In the collections of The Charleston Museum is a partial carapace of a chicken turtle (Subfamily Deirocheylinae, family Emydidae) found along the bank of Eagle Creek near its junc- tion with the Ashley River in Dorchester County (Fig. 10). It was originally thought to have come Figure 11. Life-size replica of the holotype partial skeleton of Pelagornis sandersi (ChM PV4768) on display at The Charleston Museum; a spectacular albatross-like bird with bony tooth-like projections and a wingspan of about 6.4 m. Collected from Chandler Bridge Formation near the Charleston Airport by AES and avocational fossil collector J. Malcolm in February, 1983. 126 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) from the Chandler Bridge Formation, which crops out in the vicinity, but chicken turtles are known almost exclusively from Pliocene and Pleistocene localities in the southeastern USA – not from the Oligocene (Jackson, 1978). That this specimen was found together with a humerus of the extant star-nosed mole Condylura (see “Systematic Pale- ontology” section below) supports its origination from Pleistocene sediments overlying the Chandler Bridge at this locality, not from within the latter. Three species of cheloniid sea turtles are known and are most commonly represented by car- apacial elements. An exceptional specimen is the first complete carapace (ChM PV4792) of Caro- linochelys wilsoni, which was originally described from a nearly complete skull and humerus from the Ashley Formation (Weems and Sanders, 2014; also, see above). The humerus has since been referred to Procolpochelys charlestonensis, the holotype of which (ChM PV6056) is from also the Chandler Bridge Formation (Weems and Sanders, 2014). Additional, much more complete material of Procolpochelys charlestonensis (CCNHM 893 and 300.1) was recently reported by Weems and Brown (2017). Ashleychelys palmeri was also originally described from the Ashley Formation, but appears to be more abundantly represented in the Chan- dler Bridge Formation. Remains of subadults of P. charlestonensis have not yet been encountered. Leatherback sea turtles (family Dermochelyidae) are represented in the Chandler Bridge Formation by several carapace fragments in collections at both The Charleston Museum and the SCSM and by a virtually complete carapace over 2 m in length, with the vertebral column and ribs preserved (ChM PV4893). Collected near the Charleston Airport by AES and a ChM party in 1984, the partially pre- pared carapace has seven low and narrow longitu- dinal ridges, and appears to represent a new genus. Several additional specimens of sea turtles from the Chandler Bridge Formation await study at the Mace Brown Museum of Natural History. Fish remains are particularly abundant in the Chandler Bridge Formation, especially the den- tal elements of sharks, skates, and rays. Cicimurri and Knight (2009b) reported 29 different species, including the oldest known record of the whale shark, Rhincodon, and a new skate, Raja mccol- lumi. Scombroid bony fish were reported by Fier- stine and Weems (2009), with specimens referred to Xiphiorhynchus rotundus, Xiphiorhynchus sp., Aglyptorhynchus robustus, Aglyptorhynchus sp., and to Istiophoridae gen. and sp. indet. Notably absent from the formation are remains of freshwater fish and turtles (with the exception of a few carapace elements of softshell turtle, as noted above), amphibians, alligators, aquatic freshwater mammals, and terrestrial mam- mals (Weems and Knight, 2013), although a single isolated peccary tooth (ChM PV5025) here referred to Perchoerus sp. was recovered and is discussed later in this report. Tiger Leap Formation/Edisto Formation. Huddlestun (1988) divided the Parachucla Forma- tion in Georgia into a lower Tiger Leap Member and an overlying Porters Landing Member. Weems and Edwards (2001) raised the Tiger Leap Member to formational status (dividing it into four infor- mal members) and restricted the Parachucla For- mation to Huddlestun’s Porters Landing Member. In Georgia the Tiger Leap Formation consists of a “quartzose calcarenite to calcareous quartz sand” (Weems and Edwards, 2001:11) and in the Charles- ton area it is described as a “very quartzose and phosphatic, often shelly calcarenite” (Weems and Lewis, 2002:28). On the basis of dinoflagellate assemblages, the lower members were found to be of latest Oligocene age (dinoflagellate cyst zone DN1 of de Verteuil and Norris, 1996); but Weems and Edwards (2001:11) correlated the upper mem- ber with South Carolina’s Edisto Formation, or “above the highest occurrence of the genus Chirop- teridium,” which they considered lower Miocene (also see Weems et al., 2006). Originally named the “Edisto marl” by Sloan (1908), the Edisto Formation was later formalized by Ward et al. (1979:26) for the “lower Miocene sandy limestones unconformably overlying the Cooper Formation and unconformably overlain by the Raysor Formation (lower Pliocene) as exposed in the vicinity of Givhans Ferry on the Edisto River.” In contrast to the generally accepted early Miocene age, more recent work has refined that age estimate. 87Sr/86Sr isotopic analysis on samples of ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 127 molluscs (Ostrea haitiensis) from exposures of the Edisto Formation at Givhans Ferry yielded a date of 23.4 to 23.5 Ma, entirely within the late Chat- tian (Weems et al., 2006; Weems and Harris, 2008; Weems et al., 2016). Based on oxygen isotope curves (Fig. 3B), the upper Tiger Leap/Edisto formations could not have been deposited immediately preceding nor across the Oligocene/Miocene boundary, as there was a major lowstand, the Mi1 glaciation, during that interval (e.g., Miller et al., 1991; Naish et al., 2001; and Zachos et al., 1997, 2001). Therefore, they were likely deposited during high sea level associated with peak Oligocene warmth during chron C6Cr (Pälike et al., 2006) between about 23.6 to 23.2 Ma, which again is consistent with the 87Sr/86Sr dates noted above. Currently the only known terrestrial ver- tebrate fossils that may have originated from the Tiger Leap/Edisto Formation are the type specimen of the giant entelodont Daeodon mento (Allen, 1926), a massive mandibular symphysis (see “Sys- tematic Paleontology” section below), and a frag- ment of maxilla with teeth from the horse Anchip- pus texanus Leidy, 1868b. Interestingly, both of these specimens are recorded from the “Ashley River phosphate beds,” now known to be within the Wando Formation, which dates to around 100 kyr, but which is well understood to harbor reworked fossils of much greater age (e.g., Leidy, 1877; Domning, 1989b). Recent study of the entelodont specimen revealed an indurated sandy calcarenite still adhering to portions of the bone, examination of which suggests it as Edisto Formation. Although there are no known outcrops of Edisto Formation along the Ashley River in the vicinity of the old phosphate mining region, it is the detached, worn boulders of this phosphatized unit reworked into the Wando Formation there that were the focus of mining operations (Sloan, 1908; Sanders, 2002). Alternatively, that the adhering matrix may be recrystallized Givhans Ferry Member of the Ash- ley Formation rather than Edisto cannot be dis- counted given its somewhat quartzose lithology. Considered an immigrant from Asia, Daeodon first appeared in North America in the early Arikareean, then became widespread across the continent until its demise in the early Hemingfordian (Brunet, 1979; Lucas et al., 1998). Anchippus texanus is a “medial” to late Arikareean taxon morphologically transitional between early Arikareean Miohippus and early Hemingfordian Parahippus (e.g., Para- hippus leonensis). A latest Chattian age for the Edisto Formation correlates with the late Arika- reean NALMA (Figs. 2A, 3B), consistent with the temporal range of A. texanus, but the matrix within vugs of the latter specimen is not as diagnostic as that still adhering to the specimen of Daeodon. Note on biostratigraphic correlations of Paleogene cetaceans from South Carolina Although terrestrial mammal taxa have long been utilized as biostratigraphic markers to establish the framework for North American Land Mammal Ages (e.g., Wood et al., 1941; Wood- burne, 1987, 2004), marine mammals have seldom been employed as aids in determining the age of marine stratigraphic units. Marine biostratigraphy has traditionally relied upon molluscs, foramin- ifera, dinocysts, and calcareous nannoplankton for that purpose (e.g., Berggren et al., 1995). There is mounting evidence, however, that Paleogene fos- sil cetaceans can be effectively utilized in chro- nostratigraphic correlations. Because Paleogene cetaceans were rap- idly evolving forms with distinctive evolutionary grades that occurred within a limited temporal span and range of chronostratigraphic units, the presence of one of those forms in a marine forma- tion is generally a good indication of the age of that unit (to the extent that we can currently deter- mine based on limited samples sizes). For example, archaeocetes of the family Protocetidae, among the most primitive known cetaceans, are not known to occur in strata younger than middle Eocene (e.g., Gingerich et al., 1997), and members of the fam- ily Basilosauridae, in North America, are known only from the late Eocene. Those distinctions enabled Geisler et al. (2005) to use protocetid and basilosaurid archaeocete cetaceans to support the respective ages of two marine stratigraphic units in the Eocene of South Carolina, protocetids being restricted to beds of NP16–17 age (Santee Lime- 128 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) stone and Cross Member of the Tupelo Bay Forma- tion) and basilosaurids to NP18 (Pregnall Member of the Tupelo Bay Formation). Some of the most salient skeletal changes in the evolutionary history of cetaceans have been 1) the loss of hind limbs, still present in the earli- est cetaceans (e.g., Protocetidae); 2) replacement of the rotational elbow, present in all archaeoce- tes, with the non-rotational structure present in all known post-archaeocete cetaceans; and 3) the posterior movement of the nasal opening from the anterior end of the rostrum in the most primitive cetaceans to a position near the top of the skull in modern odontocetes (toothed whales) and mystice- tes (baleen whales). The latter process, tradition- ally called “telescoping,” involved the elonga- tion of the maxillary bones posteriorly until they covered most of the frontals, the reduction of the length of the frontals, the elimination of the pari- etal bones from the skull roof, and a forward thrust of the supraoccipital toward the vertex of the skull (Miller, 1923; Churchill et al., 2018). During the Oligocene, various grades of telescoping can be seen between earlier and later forms, the position of the nasal opening and the degree of reduction of the parietals being especially important evolu- tionary landmarks (e.g., Whitmore and Sanders, 1976:fig.3). Among the Oligocene cetacean taxa of Europe, two species are known only from the lower Chattian (Chattian A) Doberg Formation in the famous Doberg quarry two kilometers south of Bünde (Westfalen) in the North Sea Basin of northwestern Germany. Both are odontocetes, viz., Eosqualodon langewieschei Rothausen, 1968, and Microcetus ambiguous (Meyer, 1840), the latter known only from teeth. As discussed above in the section on the Chandler Bridge Formation, Eosqua- lodon is now considered to have relatives in the Oli- gocene beds near Charleston and in marine depos- its at Linz, Austria. As demonstrated by Whitmore and Sanders (1976:figs. 3b, 5), several undescribed Charleston Museum specimens from the upper Oli- gocene Chandler Bridge Formation represent the same evolutionary grade as Eosqualodon, share similar cranial features with E. langewieschei, and belong in the same as yet to be described family. The Charleston Museum specimens, together with specimens at the Mace Brown Museum of Natu- ral History, appear to represent at least two genera and three species. Two other undescribed genera possibly related to Eosqualodon are known from two specimens from the Ashley Formation. They are slightly more primitive than Eosqualodon and the other Eosqualodon-like forms from the overly- ing Chandler Bridge Formation in that the parietals are exposed across the entire skull roof in adults, although much narrower than in Patriocetus. It is important to point out, however, as noted above, that the so called “waipatiids” share a similar mor- phological “grade,” thus complicating the relation- ships between the Eosqualodon-like and waipatiid- like forms from the Chandler Bridge. The teeth of these small relatives of Eosqualodon from the Chandler Bridge and Ashley formations are also similar to the teeth from Doberg to which the name Microcetus ambiguous was applied (Sanders et al., 1982), suggesting that M. ambiguous may be familially related to these forms from South Caro- lina, as well. A similar pattern also exists between the mysticetes Micromysticetus tobieni Rothausen (1971) from the lower Chattian (upper Chattian A) Meeresand near Dusseldorf, Germany, and Micro- mysticetus rothauseni Sanders and Barnes, 2002a, the holotype partial skull (ChM PV4844) of which was found in the Ashley Formation near Charles- ton. Both of those stratigraphic units are of NP24 age (Rothausen, 1971; Sanders and Barnes, 2002a). A second paratype partial skull (ChM PV5933) and an isolated periotic (ChM PV7225) are also known from the Ashley Formation. Originally described as Cetotheriopsis tobieni by Rothausen (1971), the holotype skull of that taxon was found to have diagnostic characters that ally it congenerically with M. rothauseni (Sanders and Barnes, 2002a). The occurrence of Micromysticetus in upper Rupe- lian sediments in South Carolina suggests that this genus may have originated in the Western North Atlantic. A periotic of Micromysticetus (CMM-V- 5011) is also known from New Kent County, Vir- ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 129 ginia. Collected in 2004 by REW on the beach at the foot of a bluff on the south bank of the Pamun- key River near the mutual boundary of Hanover, New Kent, and King William counties, the speci- men was almost certainly washed out of the Old Church Formation at this locality, which 87Sr/86Sr dates noted elsewhere in this report place at an age very similar to that of 87Sr/86Sr dates acquired for the Ashley Formation (Weems et al., 2006). At the col- lection site, the Old Church Formation is underlain at beach level by the middle Eocene Piney Point Formation, which is far too old to contain mysti- cete remains. The Old Church Formation is over- lain by the lower Burdigalian (Lower Miocene) Popes Creek Sand Member of the Calvert Forma- tion, which is assignable to dinoflagellate zone DN2b/c and is placed at 19-19.5 Ma (Weems and Edwards, 2007b:fig. 1). Because the only known specimens of Micromysticetus from North Amer- ica are from the Ashley Formation (~29.1 Ma), it is highly unlikely that the Virginia specimen was eroded from the lower Burdigalian sediments at this locality. Thus, the Virginia specimen strongly supports the 87Sr/86Sr date of 29.15 Ma for the Old Church Formation at the Pamunkey River locality, and provides additional evidence that the Ashley and Old Church formations are correlative. CMM- V-5011 also extends the known geographic range of Micromysticetus northward from Charleston to Virginia along the Atlantic Coastal Plain. The peri- otic from Virginia is somewhat smaller than that from South Carolina, inferring that it may repre- sent a different species. In addition to the periotic is an axis vertebra (CMM-V-5010) indistinguish- able from that of the type material of M. rothauseni found by J. Osborne while diving the Pamunkey River. This specimen provides further evidence of Micromysticetus in Virginia. Recent discoveries in North Carolina provide additional correlations with Oligocene cetacean taxa in South Carolina. Uhen (2008) described a partial odontocete skull, USNM 525001, from a block of indurated matrix of the Belgrade For- mation washed ashore at Onslow Beach, Onslow County, as the new taxon Albertocetus meffordo- rum, and erected the family Xenorophidae, the type of which is Xenorophus sloanii, described by Kel- logg (1923b) from a partial skull from the Ashley Formation near Charleston. Uhen (2008) referred two other partial skulls from Onslow Beach, USNM 529238 and 529241, to A. meffordorum, as well. Apparently two additional partial skulls from Onslow Beach, USNM 529239 and 533993, also represent the Xenorophidae. Remains of this fam- ily had previously been found only in the vicinity of Charleston, where they occur abundantly in the Ashley and Chandler Bridge formations (Sanders, 1996; Boessenecker et al., 2017a). In connection with this study, samples of mollusc shells from Belgrade matrix in which the North Carolina skulls were found yielded 87Sr/86Sr dates ranging from 26.5 to 27.5 Ma. As noted above, the Chandler Bridge Formation is now known to correlate with nannoplankton zone NP25, with an age range of between 26.84 and 23.13 Ma. Although the lithology of the Belgrade samples from Onslow Beach differs consider- ably from that of the Chandler Bridge Formation, which is a fine- to medium-grained non-calcareous quartz-phosphate sand (Sanders et al., 1982), the slight overlap in ages of the Chandler Bridge and Belgrade formations, together with the abundance of relatives of the archaic odontocete Xenorophus from both areas, suggests that the Onslow Beach sediments currently referred to the Belgrade For- mation may be time equivalent with the Chandler Bridge Formation. Additional evidence of the value of Paleo- gene cetaceans as chronostratigraphic markers is indicated by examination of an apparent conflict between the dating of the Linz Sands at Linz, Aus- tria, and the evolutionary grades of cetacean skulls found during the 19th century in a sand mine on the south side of the Donau (Danube) River at Linz. Conserved at the Oberöseterreichesches Landes- museum in Linz, those specimens were assumed to have come from the Linz Sands, a marine deposit at Linz (Rothausen, 1968). Rabeder and Steininger (1975) placed the age of the Linz Sands as latest Oli- gocene (uppermost NP25) based on the presence of the foraminifera Myogypsina formosensis, which is restricted to the uppermost Oligocene. Three taxa 130 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) have been described from the Linz specimens: the odontocetes Agriocetus incertus Brandt 1874, and Patriocetus ehrlichi (van Beneden, 1865), and the mysticete Cetotheriopsis lintianus von Meyer, 1849. As noted by Sanders et al. (1982), Agriocetus is familially related to Eosqualodon from Chattian A at Doberg and its allied forms from South Car- olina. Of the two odontocetes, Patriocetus is the more primitive, the parietal bones still being vis- ible dorsally as a narrow band across the skull roof. By the early Miocene, if not by latest Oligocene time, the telescoping process had eliminated those bones from the skull roof in most odontocetes (but see further discussion below). Two other species of Patriocetus are now known: Patriocetus kazakhstanicus Dubrovo and Sanders, 2000, from the “fish bed” of the Karagin- skaya Formation of Kazakhstan (upper Chat- tian A, upper NP24,) and a new species currently under description by K. Rothausen and AES from the similarly aged Grafenberg Formation in the Lower Rhine Embayment near Krefeld, Germany. Sediment from the holotype skull of the new spe- cies contained the nassariid gastropod Hinia (Tri- tonella) pygmaea (Schlotheim, 1920), form bispi- rallis Koch and Weichmann, 1872, a transitional form toward H. (T.) schlotheimi (Betrich, 1854) that is common in the Palliolum decussatum zone of upper Chattian A. Sediment from the type local- ity of the new species of Patriocetus contained the foraminifer Cribononion subnodosum, which is typically confined to shallow marine sediments of upper Chattian A in this region (K. Rothausen, pers. comm. to AES, 2004). Thus, since two of the three known species of Patriocetus were found in lower Chattian (Chattian A) sediments, it seems highly unlikely that the nominal species, P. ehrlichi from Linz, is of latest Oligocene age. There is also an undescribed ChM specimen from the Chan- dler Bridge Formation that is familially related to Patriocetus (Dubrovo and Sanders, 2000). The mysticete Cetotheriopsis linitanus from Linz is also an archaic form and was formerly a member of the subfamily Cetotheriopsinae of the Cetotheriidae, to which Sanders and Barnes (2002a) referred Micromysticetus. Because these taxa were more primitive than the other members of the Cetotheriidae, most of which are Neogene forms, Geisler and Sanders (2003) removed the Cetotheriopsinae from the Cetotheriidae and ele- vated Cetotheriopsinae to familial rank (Cetotheri- opsidae). The three cetacean taxa from Linz – Agrio- cetus, Patriocetus, and Cetotheriopsis – appear too primitive to have been members of the cetacean fauna of the latest Oligocene, and are more char- acteristic of earlier Chattian evolutionary grades. If the beds on the south side of the Danube were of latest Oligocene age they would have yielded specimens more closely related to the cetaceans of the Aquitanian (lower Miocene) rather than to those lower in the Chattian. A possible explana- tion for the conflict between the lower Chattian evolutionary grades of the cetaceans and the latest Chattian date of the Linz Sands is indicated by the presence of the foraminifer Miogypsina formosen- sis as reported by Rabeder and Steininger (1975). In recent correspondence between F. Steininger and AES about this biostratigraphic incongru- ity, Steininger related that the sample of the Linz Sands that furnished the Miogypsina specimen that he and Rabeder used to date that unit was taken on the north side of the Danube, not on the south side in the region where the cetacean specimens were found. Steininger further stated that “there is no direct geological evidence that the Linz Sands with Miogypsina [on the northern side of the river] are the same formation of sands as the sandy deposits from which the whales are coming on the southern bank of the Danube” (F. Steininger, pers. comm. to AES, October, 2005). Consequently, we sug- gest that the sediments that furnished the cetacean skulls on the south side of the Danube at Linz com- prise an undetermined formation of lower or mid- dle Chattian age, and that specimens from this unit should no longer be regarded as having come from the Linz Sands until the stratigraphic relationships between these two units on opposite sides of the Danube can be determined. The preceding examples provide evidence that correctly identified cetacean cranial elements can aid in determining the age of Eocene and Oli- gocene stratigraphic units and subunits, and when combined with 87Sr/86Sr dates, radioisotopic dates, ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 131 or planktonic and/or dinocyst age analyses, can, in some cases, provide dependable biostratigraphic correlations. It is important to point out, however, that there are certainly examples which prompt a degree of caution if relying solely on cetacean skulls for biostratigraphic purposes. The example above concerning the cetacean fossils from the Linz sands provides a good case in point. Another example involves the hiatus between the Ashley and Chandler Bridge formations. Origi- nally, on the basis of the evolutionary “grade” of their cetacean faunas and on certain taxa found in both formations, it was thought that little time was represented by the unconformity separating them, and both formations were thought to be early Chat- tian in age (e.g., Sanders, 1980; Sanders et al., 1982). Not until microfossils indicated that the Ashley For- mation was of NP24 age (upper Rupelian) and that the Chandler Bridge Formation was of NP25 age (“middle” to upper Chattian) was it realized that the hiatus between the two units was much greater than that indicated by the apparent similarities of their respective cetacean faunas. Another example is provided by the recently described odontocete Ediscetus osbornei Albright et al., 2018. Although E. osbornei is currently known only from the Ash- ley Formation (base of Givhans Ferry Member), it shows an unexpectedly derived stage of telescop- ing much advanced relative to contemporary spe- cies found in the same unit such as Xenorophus, Agorophius, and Ashleycetus, as well as signifi- cantly younger species from the Chandler Bridge Formation. In this species the parietals have been completely overridden by the anterior migration of the supraoccipital, and the posterior progression of the posterior frontomaxillary suture (posterior tem- poral crest) has progressed so far posteriorly that it has been pushed up vertically due to contact with the supraoccipital. Additional examples include the primitive archaeocete-like morphology of the vari- ous toothed mysticetes, such as Coronodon haven- steini, from the Chandler Bridge Formation – a “grade” much more typical of significantly older units. Miocene Series (23.03 – 5.33 Ma) Long and apparently frequent episodes of erosion during much of the Miocene along coastal South Carolina have resulted in a poor stratigraphic record for this interval of time, with a consequent paucity of non-marine vertebrate fossils. As Weems and Lewis (2002:36) noted for the Charleston area, “preserved patches of Miocene units lie directly on the Oligocene Ashley Formation and not on each other,” a pattern indicating “that each Miocene unit either was only deposited in isolated patches or was mostly eroded from the Charleston area before each subsequent Miocene unit was deposited.” They further noted that despite abundant evidence for repeated marine submergence of the Charleston region since deposition of the Ashley Formation in the late early Oligocene, no evidence for accumu- lation of any unit of great thickness exists. The few Miocene units that are known consist primarily of small, sparsely distributed subcrops characterized on the basis of comparisons with lithologically and biostratigraphically similar strata much better rep- resented in Georgia (e.g., Huddlestun, 1988, and Weems and Edwards, 2001). In ascending order these include the Parachucla, Marks Head, Coo- sawhatchie, and Ebenezer formations (Figs. 2B, 3B). As noted above, the upper members of the Tiger Leap (equivalent to the Edisto Formation) historically have been considered earliest Miocene, but new data indicates a latest Oligocene age. In North Carolina, Virginia, and Maryland the Mio- cene is much better represented and known primar- ily from the Pungo River, Calvert, Choptank, St. Marys, and Eastover formations. It is important to note, however, that over the last couple of years (2015–2019) occasional photo- graphs of specimens circulating among avocational fossil collectors (and seen by LBA) indicate that late Miocene fossils of both terrestrial and marine mammals are being recovered from a borrow pit or quarry of some sort in Jasper County, as well as from the bottoms of rivers in that region. This is certainly tenable, as sediments of this age would potentially be in the shallow subsurface of that region due to uplift of the “Beaufort Arch,” and fossils from nearby Brays Island support this. The authors of the current volume hope to learn more about this important development pending avail- 132 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) ability of additional information. Parachucla Formation. Originally named by Sloan (1908) for marls and shales that crop out along the Savannah River in the vicinity of Porters Landing in Effingham County, Georgia, the Para- chucla Formation of Weems and Edwards (2001) is, as noted above, equivalent to the Porters Land- ing Member of the Parachucla Formation of Hud- dlestun (1988). These “weakly lithified shales and mudstones” that Weems and Edwards (2001:11) noted are “thin or missing in the vicinity of Savan- nah on the crest of the Beaufort Arch,” occur in “very limited subcrop ... in the far western part of the Charleston area” (Weems and Lewis, 2002:29). Disconformably overlying the Tiger Leap Forma- tion, the Parachucla Formation harbors a forami- niferal assemblage indicative of late zone N4 to N5 and dinoflagellates indicative of dinoflagellate zone DN2, i.e., late Aquitanian in age (Huddlestun, 1988; Weems and Edwards, 2001; Fig. 3B). Although no vertebrate fossils are known from the Parachucla Formation in South Carolina, the White Springs Local Fauna of northern Flor- ida is thought to derive from what is considered an extension of the Porters Landing Member into the White Springs region (Morgan, 1989). The N4 to N5 correlation results in an age of about 21 Ma, and Jones et al. (1993) reported a 87Sr/86Sr date of 20.2 Ma for the Porters Landing Member at its type section in Georgia, thus supporting the upper Aquitanian correlation. Although a date of 20.2 Ma places the White Springs LF in the late Arikareean (Ar4) NALMA, the mammalian fauna is more indicative of the early Arikareean (Mor- gan, 1989, 1993; Albright, 1998; MacFadden and Morgan, 2003). Supporting this is another 87Sr/86Sr date reported by Jones et al. (1993) of 24.4 Ma on mollusc shells from the White Springs locality, and even more compelling is the mutual occurrence of the dugongid Crenatosiren olseni from the White Springs LF (type locality) and from the Chattian- age Chandler Bridge Formation in the Charleston area (Domning, 1997). Thus, and as also concluded by Domning (1997), what is considered the Porters Landing Member of the Parachucla Formation in the vicinity of White Springs, Florida, is signifi- cantly older (Chattian) than the same member in its type area along the Savannah River in Georgia (upper Aquitanian). Another dugongid originally described from the “Ashley phosphate beds,” Dioplotherium man- igaulti, is also known from the White Springs LF. It was this exceptional specimen (UF 95615) that finally shed light on the unit of origin for the Charleston holotype (ChM PV2896) and referred specimen (ChM PV2894), namely the Chandler Bridge Formation (see Domning, 1989a, b, for fur- ther discussion). Again, the Parachucla Formation in SC, of late Aquitanian age, is not age equivalent with the formation given the same name in north- ern Florida; the latter is Chattian in age and equiv- alent with the Charleston area’s Chandler Bridge Formation. Marks Head Formation. Like the Parachucla, the “Marks Head marl” was also named by Sloan (1908) for medium to coarse grained, clayey, phos- phatic and calcareous sands that crop out along the Savannah River. Although the term “Hawthorne Formation” has also been applied to these sedi- ments, Huddlestun (1988:50) used the term “Marks Head Formation.” Harboring a dinoflagellate assemblage indicative of DN2, Weems and Edwards (2001:12) noted that “the highest [informal] mem- ber probably can be assigned to dinoflagellate zone DN3.” The probable DN3 correlation is supported by a terrestrial mammalian fauna from this unit at Porters Landing, Georgia, which includes taxa indicative of the early Hemingfordian (Pratt and Petkewich, 1989). Deposition of sediments with a DN2-DN3 dinoflagellate assemblage together with mammals indicative of the early Hemingford- ian provide a date for this unit of around 18 Ma (Figs. 2B, 3B). Additional support for this age is provided by very fragmentary material of what was thought to be the lower to middle Miocene long- snouted odontocete “Rhabdosteus latiradix,” col- lected by AES and REW along the Georgia bank of the Savannah River near Porters Landing from “the basal couple of feet of the Marks Head … directly above the Parachucla” (REW, pers. observ., ca. 1978). However, a recent search for this material at The Charleston Museum, where it was taken after ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 133 it was collected, failed to relocate it. Although Weems and Lewis (2002:29) noted that the Marks Head Formation “is the most widespread unit above the Ashley [Formation] in the Charleston region,” no vertebrate fossils are known unequivocally from this unit in South Caro- lina. There is, however, a fragmentary cranium of “Rhabdosteus latiradix” in the collections of the SCSM labeled as having been recovered from the “Salkahatchie ooze” (see Sloan, 1908:472), but with no additional accompanying data. It is thought to have originated from the Marks Head Forma- tion, or perhaps the underlying Parachucla, given the recovery of this taxon from the Porters Landing section, as noted above; but its origin will likely remain a mystery. Coosawhatchie Formation. The Coo- sawhatchie clay of Heron et al. (1965), or the Coo- sawhatchie Clay Member of the Hawthorn For- mation of Abbott and Andrews (1979), was raised to formational status by Huddlestun (1988) for a diatomaceous clay unit that, in SC, crops out near Dawson’s Landing (southeast of Coosawhatchie), Jasper County (Fig. 1A). This unit is also well exposed at low tide in a small tidal creek that runs beneath US Highway 17, 1.44 km west of Gar- dens Corner (west of the intersection of US 17 and US 21) in Beaufort County, where it crops out as a chert-like lithology. This is the locality referred to by Sloan (1908:346) and Cooke (1936:111) as Huspa Creek (Fig. 1A), although they referred these silicified deposits to the “Parachucla shale.” Weems and Edwards (2001) slightly revised Huddlestun’s (1988) stratigraphy of the Coo- sawhatchie Formation, and it is only the lower Berryville Clay Member that correlates to that part of the section at Dawson’s Landing (as also noted by Huddlestun, 1988). They noted dinoflagel- lates from the lower Berryville indicative of DN5. Ernissee et al. (1977) placed the unit in upper N11 to lower N12 foraminiferal zones (about 13.5 Ma following GTS2012; Figs. 2B, 3B), and Abbott and Andrews (1979) correlated the Coosawhatchie with the Choptank Formation of Maryland on the basis of diatom assemblages. In turn, the Choptank Formation was correlated with nannofossil zone NN6 (see text-figure 3 of Abbott and Andrews, 1979), and de Verteuil and Norris (1996) correlated the lower Choptank with dinoflagellate zone DN6. Palmer (1988) studied radiolarians from the Daw- son’s Landing section and determined that it fell within the lower part of the Dorcadospyris alata zone (approx. 13.5-15 Ma). In addition to their correlation of the Coo- sawhatchie with the Choptank Formation, Abbott and Andrews (1979:228) also concluded that the unit must have formed under cool conditions. Although they suggested that this cool signal may have been due to localized coastal upwelling “rather than a general cooling of the Northern Hemisphere,” we now know that there was an abrupt global cooling event that terminated the middle Miocene Climatic Optimum (MMCO) at about 14 to 13.8 Ma (see, e.g., Cronin, 2009:107). Consideration of all these data supports our placement of the Coosawhatchie Formation in SC in the early Serravallian at about 13.4 Ma, slightly after termination of the MMCO (Fig. 3B). Although no vertebrate fossils are known unequivocally from the Coosawhatchie Formation, a single upper molar of the middle to late Miocene rhinoceros, Aphelops, currently housed in a pri- vate collection, may have originated from this unit. The tooth is reported to have been found by scuba divers in coastal Jasper County (see discussion of Aphelops in “Systematic Paleontology” section). Ebenezer Formation. The only other forma- tion of Miocene age currently recognized along the SC Coastal Plain that would have the potential to harbor vertebrate fossils is the Ebenezer Formation, a unit in the Charleston area known only from two small patches informally referred to as the Rudd Branch beds by Weems et al. (1987, 1997; also see Weems and Lewis, 2002). Originally considered a member of the Coosawhatchie Formation by Hud- dlestun (1988), Weems and Edwards (2001) raised this unit to formational rank on the basis of its late, rather than middle Miocene age and on lithological grounds as well (more sand-rich than clay-rich). In southern Georgia, Weems and Edwards (2001) divided the Ebenezer into five unconfor- mity-bounded members, four of which they cor- 134 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) related to dinoflagellate zone DN8. In the vicinity of Savannah they noted only the upper two mem- bers, the uppermost of which they considered to be the unit that Huddlestun (1988) referred to the Wabasso beds. In their assessment, the Wabasso beds do not occur in the Savannah area, because the type Wabasso beds in central Florida contain an early Pliocene foraminiferal assemblage (zone PL1), whereas those beds in the Savannah area referred to the Wabasso beds by Huddlestun (1988) contain, according to Weems and Edwards (2001), late Miocene dinoflagellates indicative of zone DN9 (late Tortonian; Fig. 2B). On the other hand, Weems and Edwards (2001:13, 14) also noted cal- careous nannoplankton from a core taken in the vicinity of Savannah that yielded taxa indicative of “the late Miocene and (or) early Pliocene time interval” (italics ours) in addition to shells of the mollusc Amusium mortoni in “the youngest Ebene- zer member.” Contrary to Weems’ and Edwards’ (2001) note that A. mortoni has a late Miocene to Plio- cene range, a Pliocene to lower Pleistocene range is more accurate, and this taxon is particularly diagnostic of the upper Zanclean (“middle” Plio- cene) Goose Creek Limestone (L. Campbell, pers. comm. to LBA, February, 2011). It is highly doubt- ful that A. mortoni occurs in sediments indicative of dinoflagellate zone DN9, as that zone spans the late Tortonian interval from about 8.7 to 7.6 Ma (Figs. 2B, 3B) – nearly 4 million years prior to its common occurrence in the Goose Creek Limestone at about 3.8 Ma. Therefore, in this report we rec- ognize an upper Ebenezer Formation member in SC of late Miocene age (DN9; late Tortonian), and place it at about 8.5 Ma, as well as the early Plio- cene Wabasso beds (Figs. 2C, 3C). Further discus- sion of the Wabasso follows. Pliocene Series (5.33 – 2.59 Ma) The Pliocene Epoch represents a nearly 3 myr long interval of Earth history characterized by dramatic variations in temperature and precipita- tion as global climate deteriorated toward the high- magnitude variability associated with Pleistocene glacial/interglacial cycles. There were intervals of the Pliocene that were relatively warmer than pres- ent, as well as short, significantly cooler intervals that marked the early stages of Northern Hemi- sphere glaciation (NHG; e.g., Sosdian and Rosen- thal, 2009). Several reasons for this variability are summarized by Haywood et al. (2009) and there- fore will not be reiterated here. Germane to the goals of this report, however, is the fact that these climatic variations impacted sea level, which is, in turn, important with respect to the preservation of continental shelf successions along the Atlantic Coastal Plain. As Williams et al. (2009:86) noted, “shelf successions tend to be preserved during global highstands,” thus recording warm climatic intervals, but lowstands induced by cooler climates “are often reflected in unconformities.” Also affect- ing depositional patterns of the sedimentary units in this region over this interval were minor tectonic adjustments which “altered the elevation of the continental shelf relative to sea level and realigned basin configurations” (Ward et al., 1991:274). More recently Rowley et al. (2013) and Rovere et al. (2014, 2015) have concluded that the pres- ent surface topographic architecture of the Atlantic Coastal Plain is due in large part to mantle flow influences (so called “DT” or dynamic topography influences) and to a lesser extent to glacial isostatic adjustments (GIA). They noted that these factors confound “attempts to use regional stratigraphic relations as references for longer term sea-level determinations” (Rowley et al., 2013:1560). Such factors provide additional complications when attempting to evaluate and understand the late Neo- gene stratigraphy of the SC Coastal Plain. To gain an understanding of the Pliocene stratigraphy of the Atlantic Coastal Plain, much work has focused on the highly fossiliferous units that underlie Maryland, Virginia, North Carolina, Georgia, and Florida (e.g., Cronin et al., 1984; Huddlestun, 1988; Scott and Allmon, 1992; Camp- bell, 1993; Petuch, 2003; Ward, 2008; Ward and Andrews, 2008). The limitations and difficulties of South Carolina’s Neogene stratigraphy have been noted previously, but there is a long history of attempts to correlate this stratigraphy, based primarily on molluscan faunas, with that of those states noted above where the record is more easily ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 135 observed and studied and, hence, better understood. Maryland, Virginia, and North Carolina have much better, more laterally extensive exposures of these fossiliferous units in stream and river banks/bluffs than does South Carolina, but the long term geolog- ical and paleontological studies of the Lee Creek Mine in North Carolina (e.g., Campbell, 1993; Ray, 1983, 1987, 2001; Ray et al., 2008), as well as the spectacular record of marine invertebrates from commercial shell pits in Florida (e.g., Petuch, 2003; Petuch and Roberts, 2007; Portell et al., 2012), have greatly added to that understanding. Krantz (1991) attempted to correlate Plio- cene-Pleistocene marine sedimentary units of the USA Atlantic Coastal Plain to the more continuous deep ocean record “on evidence of sea-level high- stands inferred from high-resolution δ18O records” (Krantz, 1991:163). This early work was successful as a first approximation, but since then the marine δ18O isotope record has been refined significantly and it also has been astronomically calibrated (e.g., Lisiecki and Raymo, 2005; Raffi et al., 2006) to the more recent GPTS of Gradstein et al. (2012). In this section we update the work of Krantz (1991), as well as that of others who have more recently used his 1991 model in a similar manner, i.e., to refine the temporal placement of South Carolina’s Plio-Pleistocene stratigraphy through correlation to sea level curves (e.g., Campbell, 1993; Camp- bell and Campbell, 1995). Additional, more recent papers germane to this discussion include those by Doar and Kendall (2014), Rovere et al. (2014, 2015), and references within those publications. Wabasso Beds. With the exception of limited subcrops of uppermost Ebenezer Formation noted above, upper Miocene sediments appear to be entirely missing from the SC stratigraphic record. The lower Pliocene, on the other hand, appears to be represented in the southeastern-most part of the state by what Huddlestun (1988) referred to as the Wabasso beds. Known only from the subsurface of coastal Florida, Georgia, and apparently southern South Carolina, Huddlestun (1988:98) described this informally named unit as a “phosphatic, cal- careous and microfossiliferous, variably argilla- ceous, silty, fine-grained to very fine-grained sand . . .” Although he reported no macrofossils from the Wabasso beds, vertebrate or invertebrate, Camp- bell (1993) and Campbell and Campbell (1995) reported that large specimens of the scallop Ches- apecten jeffersonius were collected by scuba divers from a unit exposed on the bottom of the Savannah River, which the Campbells interpreted to be the Wabasso beds. Indeed, Huddlestun’s (1988:plate 2) stratigraphic cross-section along the Savannah River indicates the presence of Wabasso beds in the subsurface of that region, and he also noted the occurrence of this unit in the shallow subsurface near Beaufort, SC, about 25 km northeast of the Savannah River. To the contrary, however, and as discussed above, Weems and Edwards (2001:13) did not recognize Wabasso beds in the Savannah region, instead referring Huddlestun’s unit to the upper member of the upper Miocene Ebenezer For- mation (“Ebenezer member #5”). Campbell and Campbell (1995) correlated the Wabasso beds to Zone 1, or the Sunken Meadow Member, of the Yorktown Formation in NC based primarily on the characterization of those beds by Chesapecten jeffersonius; Ward (2008) also shows these units as equivalent in his figure 3. Campbell (1993) and Campbell and Campbell (1995) addi- tionally considered these units equivalent on the basis of planktonic foraminifera found in both that are indicative of zone N18. Although Ward and Blackwelder (1980:D31) noted a personal com- munication from Joe Hazel confirming the occur- rence of N18 foraminifera in the Sunken Meadow Member, Huddlestun (1988) did not indicate the presence of N18 foraminifera in the Wabasso beds – he assigned the unit to the upper part of Atlantic planktonic foraminiferal zone PL1 (Huddlestun, 1988:100), a zone much less temporally restrictive than N18. In fact, few if any of the species listed by Huddlestun are restricted to N18 and several, particularly the more ubiquitous and biochrono- logically significant forms such as Globoturboro- talia (= Globigerina) nepenthes, Globorotalia margaritae, and Dentoglobigerina (= Globoquad- rina) altispira, are more indicative of N19. Ward (2008:352) provided a long list of planktonic fora- minifera from the Sunken Meadow Member, many 136 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) of which are also indicative of N19. Together, these two lines of evidence suggest an N19 age for the Wabasso beds rather than N18, if indeed the Wabasso beds and the Sunken Meadow Member are equivalent. There is one additional line of evi- dence supporting an N19 age, as well. Recovered from the Sunken Meadow Mem- ber is a mammalian assemblage originally described by Tedford and Hunter (1984), but later revised by Eshelman and Whitmore (2008), known as the Lee Creek Local Fauna. On the basis of several mam- malian species, Morgan (1994) and Eshelman and Whitmore (2008) concluded that this fauna cor- related with the latest Hemphillian (Hh4, 4.7–5.0 Ma) Palmetto Fauna of Florida (also see Morgan, 2005; Webb et al., 2008). If all the temporal data are considered, including a PL1 plus an N19 plank- tic foraminifera designation, together with a latest Hemphillian land mammal fauna, the marine high- stands over which the Wabasso beds and the Sunken Meadow Member of the Yorktown Formation were likely deposited (again – if the Wabasso beds and the Sunken Meadow Member are equivalent) are those spanning marine δ18O isotope stages T5–T7, or perhaps the interval of high sea-level bracketed by the Si6 and T4 lowstands (Fig. 3C). This pro- vides a refined age for these units of between 4.9 and 5.15 Ma. If the Wabasso beds fell within N18, this would place the unit in the uppermost Miocene rather than lower Pliocene, and within the early late Hemphillian (Hh3), rather than the latest Hemp- hillian (Hh4) NALMA (Figs. 2C, 3C). Although not a mammal, some of the most common and highly sought after vertebrate fos- sils from the southeastern Atlantic Coastal Plain (by avocational fossil collectors) are the impres- sively large teeth of the shark Megaselachus mega- lodon. As Ward (2008:275–276) noted, this taxon last occurs in the Sunken Meadow Member of the Yorktown Formation and “is common in middle to upper Miocene beds from Florida to Maryland.” Its distribution throughout a number of lag deposits typically found at the bases of various Pliocene and Pleistocene units in the SC Coastal Plain supports Ward’s (2008:276) conclusion that “the Sunken Meadow Member [and equivalent units] may have been more broadly distributed than its present pre- served limit” (see further discussion of Wabasso beds in the “Systematic Paleontology” section on Phocanella pumila below). Although Boess- enecker et al. (2019:30) did not consider specimens of Megaselachus megalodon (their Otodus mega- lodon) from South Carolina in their analysis of its extinction, they did consider this taxon’s absence from the members of the Yorktown Formation overlying the Sunken Meadow Member as “bio- chronologically real” thus reflecting the “genuine absence of this taxon.” They concluded that this largest of all sharks likely went extinct around the Zanclean/Piacenzian boundary at 3.6 Ma. Teeth of Megaselachus megalodon are also known from the Bee Ridge Fauna, Florida, of similar age (Morgan, 1994). Goose Creek Limestone. Originally named the “Goose Creek phase” by Sloan (1908), this “medium- to coarse-grained, quartzose and phos- phatic, sparsely shelly, pale-buff (wet) to chalk white (dry) calcarenite” was “revived, formal- ized, and renamed” the Goose Creek Limestone by Weems et al. (1982:H137, H140). Obviously of Pliocene age based on its fossil content, its position within SC Coastal Plain stratigraphy, as Weems et al. (1982) noted, has been controversial and the reader is referred to that publication, and also to Campbell and Campbell (1995), for details and clarification. The extent of the Goose Creek Limestone is based mainly on data from auger holes, but a few isolated outcrops are known in the Charleston area, and Weems et al. (1982) noted its extensive occurrence along the axis of the Cooper River. This has been confirmed through the efforts of scuba divers (including the first author) who have observed limited, but highly distinctive, out- crops of this unit submerged, but in shallow depths, resting unconformably on the Ashley Formation within the Cooper River. These outcrops harbor exceptionally large specimens of Amusium mortoni and Encope sp. (pers. observ., LBA). Additionally, several vertebrate fossils from the Cooper River have matrix trapped in vugs, voids, alveoli, etc., closely matching the lithology of the Goose Creek. In March 2010, LBA, AES, REW, and J. Osborne ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 137 visited the type locality of this unit on the banks of Goose Creek in Berkeley County and observed several specimens of Amusium mortoni and Euvola hemicyclicus (Ravenel, 1834). Sediment samples were collected from this locality for paleomagnetic analysis. Adding significantly to an understanding of the Goose Creek Limestone and its stratigraphic relationships is the work of Campbell (1993) and Campbell and Campbell (1995) who provided detailed studies of the unit’s molluscan fauna, as well as that of Bybell (1990) who reported on the calcareous nannofossils. The Campbells’ studies resulted in a biostratigraphic subdivision of the Goose Creek into lower and upper units. Those exposures at the stratotype section on Goose Creek in Berkeley County, at the Martin Marietta Berke- ley quarry near Cross, and at Givhans Ferry along the Edisto River (the latter of which the first author searched for unsuccessfully on numerous occa- sions) comprise the lower unit, whereas the upper unit was determined to be equivalent to the Ray- sor Formation at Canady’s Bridge on the Edisto River (see discussion below), at the Martin Mari- etta quarry (see Blackwelder and Ward, 1979, and Ward and Huddlestun, 1988), and to the stratotype section of the Bear Bluff Formation at Bear Bluff on the Waccamaw River, north of Conway, SC. It is important to note, however, that the lithology of the Raysor Formation at the above cited locali- ties is a shelly quartz sand, in stark contrast to the calcarenite typical of the Goose Creek Limestone in the Charleston area (Weems et al., 1982). Addi- tional correlations include that of the upper unit to the Buckingham Limestone of south Florida and to strata at Rice’s Pit in North Carolina; and of the lower unit to the stratotype of the Tamiami For- mation in south Florida and to the Aurora beds in North Carolina (Campbell, 1993). On the basis of the shared occurrence of Chesapecten septenarius, Euvola hemicyclicus, Amusium mortoni, Ecphora quadricostata, and several other invertebrate taxa, the Goose Creek Limestone is typically considered essentially equivalent to the Rushmere Member of the Yorktown Formation. It does not correlate to the older Sunken Meadow Member as implied by Petuch (2003:141, 144). From a vertebrate paleontological perspec- tive, the importance of an accurate age for the Goose Creek Limestone cannot be understated due to the fact that some of the mammalian remains from the unit belong to Neotropical taxa involved in the Great American Biotic Interchange (GABI). Depending on the age of the unit, remains of these taxa may represent their oldest records in North America. One example is the capybara Neochoerus pinckneyi (see discussion of capybara nomencla- ture/taxonomy in “Systematic Paleontology” sec- tion). Teeth referred to this species by Sanders (2002) from the Charles Pinckney collection at The Charleston Museum, including the holotype (ChM PV2506), were dredged from the Ashley River during phosphate mining operations at Runnymede Plantation around 1900. That they were derived from the Goose Creek Limestone is based on matrix of this unit still adhering to the specimens (Sanders, 2002:101). Additional confirmation was provided in 1979 when Doris Holt, an avocational fossil collector, recovered a capybara tooth from a borrow pit excavated on the west bank of the Ashley River about 8 km downstream from Run- nymede Plantation during construction of the Mark Clark Expressway in Charleston within which were exposures of the Goose Creek Limestone. Camp- bell and Campbell (1995:59, 64) determined that these exposures belonged to the upper unit. The presence in the Goose Creek Limestone of the calcareous nannofossils Reticulofenestra pseudoumbilica and Sphenolithus abies, noted by Weems et al. (1982), helps refine the age of this unit, as does the presence of Pseudoemiliania lacu- nosa reported by Bybell (1990). According to Raffi et al. (2006:fig. 3; also see Mudelsee and Raymo, 2005), R. pseudoumbilica ranges from approxi- mately 3.8 to 4.9 Ma, and the HO of S. abies is at about 3.55 Ma. Bybell (1990) placed the FAD of P. lacunosa in South Carolina in lower NN15, which, following the time scale of Lourens et al. (2004), occurs at approximately 3.9 Ma (late Zanclean). That the formation can be no younger than 3.8 Ma is additionally supported by an N19 foraminiferal 138 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) zone designation for Rice’s Pit (Weems et al., 1982; Akers, 1972, as noted in Campbell, 1993:13), as well as the PL3 foraminiferal zone and NN15 to middle NN14 calcareous nannofossil designation for the section of lower Goose Creek Limestone at Givhans Ferry State Park (Ward and Huddles- tun, 1988; Bybell, 1990). These lines of evidence imply an age for the Goose Creek of between 3.8 and 4.0 Ma. To satisfy these biochronologic crite- ria, we correlate the upper and lower units of the Goose Creek Limestone (and their equivalents; see Fig. 2C) to that part of the marine oxygen isotope curve bracketed by Gi20 below (a major lowstand) and Gi10 above (the HO of Reticulofenestra pseu- doumbilica). Supporting this correlation, which places the Goose Creek Limestone within magne- tochron C2Ar, is the recent paleomagnetic analy- sis by LBA suggesting reversed polarity (but not definitive) for this unit. Together these data imply that capybaras were present along the southeastern Atlantic Coastal Plain before 3.8 Ma – a conclusion similarly reached by Woodburne (2010), but sup- ported here with additional evidence (see further discussion in “Systematic Paleontology” section). Additional vertebrate remains from the Goose Creek Limestone include those of a Minke whale and a gannet (Weems et al., 1982), as well as the possible presence of three pinnipeds – the wal- rus Ontocetus emmonsi Leidy, 1859, and the seals Phocanella pumila Van Beneden, 1877, and Cal- lophoca obscura Van Beneden, 1877 (see “System- atic Paleontology” section). Cicimurri and Knight (2009c) described a partial skeleton, including the skull (SC79.65.20), of another possible Minke whale recovered by scuba divers from the Goose Creek Limestone in the Cooper River that showed evidence of scavenging by sharks. Campbell and Campbell (1995) noted “deer antler, Artiodactyl genus uncertain, porpoise, ... sperm whale,” and teeth of various sharks. Leidy (1877; also see Allen, 1926) described several species of ziphiid (beaked) whales from the Ashley River phosphate beds, nearly all specimens of which are heavily permineralized and highly water worn. It has been speculated that some of these may have been reworked from the Goose Creek Limestone, but no adhering matrix remains to confirm this and specimens from these “beds” range in age from late Eocene to Pleistocene (see Domning, 1989b, and discussion on Wando Forma- tion below). On the other hand, one of the cetacean specimens that Leidy (1877:231) described, Ceter- hinops longifrons (although not a ziphiid), included matrix still adhering to the partial skull. Personal examination of this specimen in collections at the Academy of Natural Sciences, Philadelphia, by AES resulted in the determination that the matrix belonged to the Goose Creek Limestone. A second specimen referred to Tusciziphius crispus, origi- nally described from the lower Pliocene of Italy, but reported by Post et al. (2008) from the Morgan River of Beaufort County, is also noted as having been recovered from the Goose Creek Limestone. Further discussion of these beaked whales is pro- vided below in the section on the Wando Forma- tion. Although the marine vertebrates of the Goose Creek suggest deposition under cool to temperate conditions, the Rushmere Member of the Yorktown Formation is typically considered to have been deposited under somewhat warmer con- ditions (e.g., Ward et al., 1991) not unlike those off North Carolina today (Snyder et al., 2001). If these units are indeed correlative, which the evidence supports, perhaps the Goose Creek Limestone was deposited under conditions of localized dynamic upwelling similar to those hypothesized by Snyder et al. (2001) for deposition of the Sunken Meadow Member of the Yorktown Formation. The presence of phosphate nodules in the Goose Creek lends additional support. Raysor Formation. Like the Goose Creek Limestone, the status of the Raysor Formation has a similarly confusing and complicated his- tory, which was summarized and somewhat clari- fied in Weems et al. (1982), but also discussed in several other works on Coastal Plain stratigraphy, principally Huddlestun (1988), Ward and Huddles- tun (1988), Ward et al. (1991), and Campbell and Campbell (1995). To briefly summarize, the Raysor Formation was originally named the “Raysor marl” by Cooke (1936) for exposures of Sloan’s 1908 “Upper Pee Dee Phase” along the west bank of the Edisto River near what Sloan (1908) mistakenly ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 139 believed was Raysor’s Bridge. The location has more recently been determined to have been where Canady’s Bridge crosses the river, as the physical attributes there very closely match those described by Sloan – a bluff 1200 feet below the bridge. At the Raysor’s Bridge locality, there is no bluff (see details in Huddlestun, 1988:114). The bluff below Canady’s Bridge is now referred to as Serenity Bluff as determined from the name of the road that provides access to the private property currently encompassing the bluff, i.e., Serenity Bluff Road. Furthermore, the distance from Givhans Ferry to Raysor’s Bridge according to Sloan (1908:471) was about 12 miles; this is the distance from Givhans Ferry to Canady’s Bridge. The distance from Givhans Ferry to the location of Raysor’s Bridge, based on the locality given by Huddlestun (1988:113, fig. 45), is about 16 miles. Because the original “type” locality (at the bluff below Canady’s Bridge) has long been inaccessible (overgrown and on private property), Blackwelder and Ward (1979) assigned a neostratotype section on the east bank of the Edisto River near Givhans Ferry State Park and updated the term Raysor Marl to Raysor Forma- tion. However, as Weems et al. (1982:H144–H146) and Campbell and Campbell (1995) noted, the lithology at Givhans Ferry referred to the Raysor Formation by Blackwelder and Ward (1979) con- trasts with that at the Canady’s Bridge locality and “matches well with the Goose Creek Limestone (to which Sloan, 1908 referred this locality) ...” (see above discussion under Goose Creek section). Weems et al. (1982) recommended abandonment of the Givhans Ferry section as a neostratotype for the Raysor Formation, noting the “striking” con- trast between it, which they generalized as a bio- calcarenite with a quartzose matrix, and the Goose Creek Limestone, which they referred to as a bio- calcarenite with a calcareous matrix. In what we perceive to be a glaring omission, neither Ward and Huddlestun (1988) nor Ward et al. (1991) refer- enced the findings of Weems et al. (1982) regard- ing the inadequacy of the Givhans Ferry site as a neostratotype for the Raysor Formation. Ward and Huddlestun (1988:fig. 2), in fact, placed the Goose Creek Limestone stratigraphically above the Ray- sor Formation, a likely result of Blackwelder’s and Ward’s (1979) assignment of Goose Creek expo- sures at Givhans Ferry to the Raysor, thus per- petuating the erroneous stratigraphic relationship deeper into the technical literature. Although Weems et al. (1982) and Bybell (1990) noted the problematic stratigraphic rela- tionship between the Raysor Formation and Goose Creek Limestone, Campbell and Campbell (1995) were able to shed light on this problem based on exposures of these units in the Martin Marietta Aggregates Berkeley (“Cross”) quarry. Exposures of the Raysor at this quarry had been previously noted by Ward et al. (1979). Determining that the Goose Creek Limestone is biostratigraphically divisible into upper and lower units, Campbell and Campbell (1995) found the Raysor Forma- tion resting disconformably above the lower Goose Creek unit in the quarry exposures. On the basis of mutually occurring biochronologically significant invertebrate taxa (e.g., Chesapecten septenarius, Carolinapecten eboreus var. walkerensis, Ecphora quadricostata, E. bradleyae, Encope macrophora, and others), they also determined that the Raysor was 1) laterally equivalent to the upper unit of the Goose Creek (the former an inshore, more silici- clastic facies than the latter, which is an offshore calcareous facies); 2) that both of these units, in turn, are equivalent to the stratotype section of the Bear Bluff Formation (“bed A” of DuBar, 1987) in Horry County; and 3) that these units pre-date the Duplin Formation (in the restricted sense, i.e., at the Natural Well, NC, stratotype; see further dis- cussion below). Supporting the latter correlation, Campbell and Campbell (1995) noted that Ches- apecten septenarius went extinct prior to deposi- tion of the Duplin, in turn implying a post-Rush- mere Member (Yorktown Formation) age for that formation. The concept of equivalent Goose Creek and Raysor formations and a post-Raysor (= post- Rushmere) aged Duplin Formation contrasts with the more traditional stratigraphic model (e.g., that of Cooke, 1945; Blackwelder and Ward, 1979; Cronin et al., 1984; Ward and Huddlestun, 1988; Ward et al., 1991; Ward, 2008) whereby the Ray- 140 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) sor and Duplin formations, plus the Rushmere and Morgarts Beach members of the Yorktown Forma- tion, are considered correlative. Although Ward and Huddlestun (1988) also noted the absence of C. septenarius in units above the Rushmere, Ward et al. (1991:277) cited the presence of this taxon in Duplin strata at Tarheel, NC. Mansfield (1935) and Cooke (1936) origi- nally considered the Raysor Marl equivalent with strata now referred to the Sunken Meadow Member of the Yorktown Formation. After recommending abandonment of the term “Duplin Marl or Forma- tion,” Blackwelder and Ward (1979:36) introduced the term “Raysor Formation” for calcarenite beds in southern South Carolina and southeastern Geor- gia that they considered to be biostratigraphically equivalent to “most of the Yorktown, although per- haps not the lowermost part.” Refining this corre- lation, Ward and Huddlestun (1988) correlated the Raysor to the Rushmere and Morgarts Beach mem- bers, and upon reinstating the Duplin Formation as a formal lithostratigraphic unit, Ward et al. (1991) considered it an updip lithofacies of the Raysor. Noting that Chesapecten septenarius was absent in units above the Rushmere Member, and consider- ing the Rushmere and Morgarts Beach members as “lithofacies of the same transgressive event,” Ward and Huddlestun (1988:72) attributed this pecten’s absence in the latter member to different substrate conditions. Ward (2008:360) interpreted the Rush- mere Member as the “very shelly, poorly-sorted sands of a transgression,” and the Morgarts Beach Member as “the quiet-water, fine, well-sorted sands of the high-stand that followed that transgression.” On the basis of ostracod assemblages, mollusc δ18O profiles, and benthic foraminiferal assemblages, Snyder et al. (2001:259) additionally noted that the two members were essentially indistinguishable from one another with respect to paleotempera- ture estimates. Based on comparisons with modern faunas, however, they concluded that the Morgarts Beach Member may have been deposited under slightly cooler temperatures, and under shoaling or middle neritic conditions, than the conformably underlying Rushmere Member. In contrast, Campbell (1993:7) contested the view that the Rushmere and Morgarts Beach mem- bers were deposited during a single transgression, which he supported through an attempt to corre- late statistically determined biostratigraphic inter- vals to Krantz’s (1991) δ18O model of sea level fluctuations. Although Campbell (1993:11) noted that the marine isotope curves used by Krantz were “extremely well supported” and that they could “be expected to withstand significant revision for the foreseeable future,” these curves have in recent years been further refined and recalibrated to the updated GPTS of, primarily, Lourens et al. (2004). Consequently, it is now difficult to correlate parts of Krantz’s (1991) curve, such as the segment in his figure 6 between the two major lowstands at 4.8 and 3.1 Ma, to the more refined curves of, e.g., Raffi et al. (2006). It is relatively obvious that the two aforementioned lowstands correlate to iso- tope stages Si4–6 and M2–MG2, respectively, in the more recent and refined δ18O curves (Fig. 3C). However, it is quite difficult to accurately discern to which isotope stages between those two events the “Krantz events” referred to by Campbell and Campbell (1995) as K2–K11 correlate, and on which Campbell (1993) based his model of refined Yorktown/Chowan River depositional patterns. Regarding fossil vertebrates, the only speci- men of a borophagine canid known from SC was found in what was likely the Raysor Formation in the Martin Marietta Orangeburg quarry. This speci- men, a single lower premolar (p4), has been identi- fied by Tseng and Geisler (2016) as belonging to Borophagus hilli (see “Systematic Paleontology” section), previously known from the latest Hemp- hillian of Kansas, Texas, and Florida and the early Blancan of Idaho, Washington, New Mexico, and Mexico (Wang et al., 1999). Duplin Formation. The traditional view of the Duplin Formation holds that this unit includes those beds south of the Neuse River in North Caro- lina, and across the Cape Fear Arch into northern South Carolina, that are correlative with the Rush- mere and Morgarts Beach members of the York- town Formation. The Duplin Formation also has been considered the updip, siliciclastic facies of the more calcareous Raysor Formation in South ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 141 Carolina (Blackwelder and Ward, 1979; Cronin et al., 1984; Huddlestun, 1988; Ward and Huddles- tun, 1988; Dowsett and Cronin, 1990; Ward et al., 1991; Ward, 2008). An alternative interpretation espoused by Campbell (1993) and Campbell and Campbell (1995), who subscribe to a more restricted view of the Duplin Formation, suggests that this unit is younger than the Raysor/upper Goose Creek/Rush- mere complex. They base this on a comparison of molluscs from these and equivalent sites consid- ered “pre-Duplin” with species derived specifically from the Duplin Formation’s Natural Well, NC, stratotype and from the Tearcoat Branch and Muld- row Place localities in Sumter County, SC, which they consider equivalent to the stratotype. As noted above, Chesapecten septenarius is not found strati- graphically above the Rushmere Member of the Yorktown, and in contrast to other workers who include this taxon in the Duplin fauna (see Ward et al., 1991:277), Campbell and Campbell (1995) noted its extinction prior to deposition of their con- cept of the Duplin Formation. They explain this discrepancy by noting (p. 58) that “the majority of species noted by Dall (1903) from ‘Natural well and vicinity of Magnolia’ came from the Strickland Farm rather than from Natural Well ... .” Campbell (1993:9) does, however, note the presence of C. septenarius in the type section of the Raysor, as do Ward and Huddlestun (1988). If C. septenarius is indeed absent from the highly fossiliferous Natural Well stratotype, then this would lend support to a post- Raysor/upper Goose Creek/Rushmere aged Duplin Formation, as well as to a possible Morgarts Beach Member correlation; the age of the Duplin is further investigated below. The Duplin Formation has long been consid- ered extremely fossiliferous; Campbell and Camp- bell (1995) noted over 450 molluscan species from this unit. Traditionally it has been considered to have been deposited during the same major trans- gression as that under which the Rushmere-Mor- garts Beach members of the Yorktown Formation were deposited. The geomorphic expression of this middle Pliocene shoreline has resulted in a long, nearly continuous feature, which extends from North Carolina to Florida (e.g., Dowsett and Cro- nin, 1990; Rovere et al., 2015). In South Carolina this feature is known as the Orangeburg Scarp; in Virginia as the Chippenham-Thornburg Scarp. Ward et al. (1991:276) noted that these depos- its “overlapped the entire Coastal Plain and por- tions of the eastern Piedmont,” and that “[a]long the western margin of the outcrop belt, Rushmere strata rest directly on crystalline rocks ... .” On the basis of planktonic foraminiferans, calcareous nan- nofossils, and ostracods collected from two sites, one each in SC and NC near the Orangeburg Scarp, Dowsett and Cronin (1990:436) determined that this major transgression ranged from 3.5 to 3.0 Ma. They additionally noted a correlation of the Dup- lin and its equivalents “to the uppermost part of zone N19 and part of zone N20 of Blow (1969) and to zone PL3 of Berggren (1973),” which closely matches Ward and Huddlestun’s (1988) designa- tion of the Raysor Formation to N20/PL3. Since Dowsett’s and Cronin’s (1990) publi- cation, not only has the GPTS been significantly refined, but so have LADs of several biochrono- logically significant foraminiferans and calcareous nannofossils (e.g., Lourens et al., 2004; Mudelsee and Raymo, 2005). GTS2012, in which marine oxygen isotope stratigraphy has been astronomi- cally tuned over the Pliocene, allows for a reassess- ment of the age of this transgression. A correlation to upper N19, lower N20, and PL3 would indicate an age of about 3.5 to 3.8 Ma (Fig. 3C). Importantly, however, and as seen in Figure 3C, if the Duplin is younger than the upper Goose Creek/Raysor/ Rushmere, then it cannot have been deposited prior to 3.66 to 3.63 Ma, as those points in time corre- spond to what Lawrence et al. (2009:8) referred to as “precursor glaciations” at marine isotope stages Gi4 and Gi2, respectively (i.e., precursors to inten- sification of large scale Northern Hemisphere gla- ciation [NHG] at ~2.7 Ma; see Fig. 3C). Similarly, the Duplin cannot have been deposited from 3.340 to 3.295 Ma either (marine isotope stages MG2– M2), as these also represent significant (and early) events of NHG with estimates of sea level approxi- mately 65 m below present (Lisiecki and Raymo, 2005; Mudelsee and Raymo, 2005; Dwyer and 142 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) Chandler, 2009; Lawrence et al., 2009). This con- strains the age of the Duplin to between 3.34 and 3.63 Ma; hence our placement of the Duplin during the early Piacenzian highstand at MIS MG7 at 3.5 Ma (Fig. 3C). On the other hand, there were significant events of sea-level rise within the PRISM interval, or Middle Piacenzian Warm Period (MPWP), cen- tered at about 3.2 Ma (e.g., Dowsett and Cronin, 1990; Dowsett, 2007; Dowsett et al., 2005, 2009, 2010, 2011, 2016) during which the Duplin could have been deposited, and it is in this window that L. Campbell (pers. comm. to AES and LBA, 2009) suggests the unit was formed (Fig. 3C). These high sea-level events, particularly MIS KM5 and KM3 estimated at about 25-30 m above present, fol- lowed the dramatic sea-level lowering at MIS M2, and they, in turn, were followed by another major lowstand at KM2 (40 m below present) at about 3.15 Ma (Dwyer and Chandler, 2009). In summary, the age of the Duplin remains somewhat equivocal. If it is indeed younger than the Goose Creek/Raysor/Rushmere, then it was either deposited during the MG7 sea level high- stand at about 3.5 Ma, or within the MPWP during marine isotope stages KM3 or KM5 at about 3.2 Ma. Based on the above, and keeping in mind that there are few places, if any, in SC where the strati- graphic relationships of these units are exposed in outcrop, it should be obvious that the determi- nation of whether a fragmentary vertebrate fossil originated from the Raysor or the Duplin forma- tions is a challenging exercise. However, on the basis of the more siliciclastic nature of the Duplin Formation inasmuch as it represents a more shore- ward facies than the Raysor (regardless of age), the fossilized remains of terrestrial mammals are more likely to have derived from the former than the lat- ter in those areas of SC where mammals are being recovered from distinctively sandy/shelly units of appropriate age. These include the remains of Phugatherium and Glyptotherium from Florence and Dorchester counties, respectively. In the Charleston area, Ward et al. (1991:277) noted that the Duplin Formation grades laterally into a “calcarenite that is relatively free of silici- clastics” which they further noted was mapped as the Raysor Formation following the recom- mendation of Blackwelder and Ward (1979). This description prompts the question as to whether the siliciclastic-free calcarenite might more accurately be the Goose Creek Limestone, and if their Dup- lin is actually the Raysor, particularly considering Blackwelder’s and Ward’s (1979) mistaken assign- ment of Goose Creek exposures at Givhans Ferry to the Raysor Formation, and Weems et al.’s (1982) assessment of the Raysor in the Charleston area as a biocalcarenite with a quartzose matrix (and the Goose Creek as a biocalcarenite with a calcareous matrix). The interpretation that the Raysor Forma- tion is the down-dip facies of the Duplin Forma- tion is certainly tenable between Charleston and the NC border given the more calcareous nature of the former relative to the more siliciclastic nature of the latter; but it is our opinion that Ward et al. (1991) erred in their concept of a relatively clas- tic-free Raysor Formation in the Charleston area. This siliciclastic-free unit is likely the Goose Creek Limestone. Cypresshead Formation. Although described and characterized primarily in Wayne and Effing- ham counties, Georgia, Huddlestun (1988:122) also noted that the Cypresshead Formation extends “at least as far north as the vicinity of Summerville in Dorchester County, South Carolina ....” These burrowed, bioturbated, non-phosphatic, horizon- tal and cross-bedded sands are considered to have been deposited in a coastal beach/sound environ- ment. Stratigraphic and limited paleontological data indicate a late Piacenzian (late Pliocene, PL5/ N21) age (Huddlestun, 1988). No vertebrate fossils are known from this unit. Pleistocene Series (2.59 – 0.0117 Ma) In the most recent synthesis of Pleistocene mammalian faunas of SC, Sanders (2002) added taxa previously unknown from the state, updated the identification of some known specimens, and also included discussions and clarifications of the geologic units from which many of the specimens were recovered. Of particular importance are his discussions on the “Ashley River phosphate beds,” from which so many of Charleston’s fossils were ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 143 recovered, and which he determined were derived from the middle portion of the Wando Formation of McCartan et al. (1980). The following discussion reiterates information provided in Sanders (2002), but also provides further details on discoveries that were just coming to light as that work was being published, such as the spectacular assemblage of beautifully preserved mammals that constitute the Camelot Local Fauna, the rich and highly diverse Walrus Ditch Local Fauna, and the equally impor- tant Crowfield Local Fauna. The fossils from these three assemblages are curated at the SCSM. Waccamaw Formation (Lower Pleistocene). Originally named the “Waccamaw beds” by Dall (1892) for shelly sands exposed along the Wacca- maw River in Horry County, it was Blackwelder (1979) who formalized the term “Waccamaw For- mation.” Ward et al. (1991:282) noted the unit’s limited distribution as occupying “an area on the southeastern flank of the Cape Fear arch.” They also noted that the Waccamaw sites along the Intra- coastal Waterway in Horry County had a “some- what different” molluscan fauna than the sites in Columbus and Brunswick counties, NC, thus con- cluding that the two beds were deposited during separate transgressions. This, in turn, led to the concept of an upper and lower Waccamaw. Accord- ing to Ward et al. (1991) the upper Waccamaw is represented by those beds at Calabash, NC, and along the Intracoastal Waterway in Horry County, SC, whereas the lower Waccamaw is represented by sites in Columbus and Brunswick counties, NC, such as the Old Dock, Shallotte, and Walkers Bluff localities. Campbell and Campbell (1995) recog- nized both the upper and lower units at Calabash noting their separation by an unconformity. (Also see abstracts by Graybill et al., 2009; Kelley et al., 2010a, b; Badyrka et al., 2010). Several attempts have been conducted to accurately date the Waccamaw Formation, includ- ing the use of calcareous nannoplankton, plank- tic foraminiferans, molluscs, ostracods, 87Sr/86Sr analysis, amino acid racemization, uranium series dating, and magnetostratigraphy (e.g., McCartan et al., 1982; Cronin et al., 1984, Bybell, 1990; Ward et al., 1991; Campbell and Campbell, 1995; Gray- bill et al., 2009; Appleby et al., 2010; Badyrka et al., 2010). Most of these have resulted in a general- ized “late Pliocene-early Pleistocene” age assign- ment for the unit. Note, however, that the recent establishment of the Pliocene-Pleistocene bound- ary at the base of the Gelasian Stage (Gibbard et al., 2010) results in placement of the unit wholly within the lower Pleistocene. According to Campbell and Campbell (1995:66), “the lower Waccamaw dates from 2.4 Ma based on the planktonic foraminifera, and from 2.53 Ma by U-He coral dates from correla- tive deposits in southern Florida (Bender, 1973).” They added that the molluscan fauna was domi- nated by mid-shelf species, many of which Ward et al. (1991:288) noted grew to large size “suggest- ing optimal [subtropical to tropical] conditions...” Between 2.42 and 2.52 Ma are three significant events of sea level lowering at marine isotope stages 96, 98, and 100 (Lisiecki and Raymo, 2005), therefore precluding deposition of the lower Wac- camaw over that interval. Preceding those events are two highstands at stages 101 and 103 (2.57 Ma and 2.59 Ma, respectively), but more significant are the intervals of high sea level at stages 93 and 91, at about 2.38 and 2.34 Ma, respectively; it is to these stages that we correlate deposition of the lower Waccamaw Formation (Fig. 3C). Edwards et al. (2000) determined that the Waccamaw Formation extends farther south into SC (into Dorchester County) than previously con- sidered, although separation into upper and lower units is not as easily determinable as in its more typical area to the north. Bearing on the age of this unit is the Walrus Ditch Local Fauna, also from Dorchester County (Downing and White, 1995; Fields et al., 2012). Although the sediments from which the fauna was recovered yielded both marine and terrestrial species, their gravelly, poorly sorted, coarse-grained nature failed to yield microfossils that might help refine its age. However, referral of these beds to the Waccamaw Formation is sup- ported by the nearly identical description of that unit by Edwards et al. (2000) from several cores drilled in Dorchester County, who noted that the unit weathers to a dark-yellowish-orange to red- dish-brown color – exactly that seen at the Walrus Ditch locality. 144 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) The terrestrial species include several indic- ative of a late Blancan to early Irvingtonian age, some of which have long ranges, but others of which are biochronologically significant (Table 3). The presence of the porcupine Erethizon limits the maximum age of the fauna to about 2.6 Ma, whereas the presence of Nannippus places a young- est limit on the fauna of about 2.1 Ma (Bell, 2004). The occurrence of Allophaiomys represents one of the oldest records of this taxon in North America. These records provide a late Blancan age for the fauna (also see Fields et al., 2012), again compat- ible with our correlation to marine isotope stages 93 and 91. Correlation of the Walrus Ditch LF to the late Blancan also supports placement of the lower Waccamaw Formation in chron C2r2r, the lower- most reversed interval of the Matuyama Chron, spanning 2.2-2.58 Ma. Although Liddicoat et al. (1981) reported a reversed polarity for the Wac- camaw beds in the Charleston area, which would provide further support for our correlation, later interpretation of those beds by Weems and Lemon (1988) resulted in a Penholoway Formation assign- ment (the unit overlying the Waccamaw). Never- theless, the variety of data from the lower Wac- camaw converges on a date of about 2.3-2.4 Ma (Figs. 2C, 3C). In addition to the terrestrial mam- mals listed in Table 3, the fauna also includes ceta- ceans representative of four families (Delphinidae, Kogiidae, Pontoporiidae, and Mysticeti), a dugon- gid, an undescribed taxon of walrus (the fauna’s namesake), and several species of marine birds and marine/estuarine fish. Blancan aged mammals from the highly mixed assemblage recovered from the earlier noted Clapp Creek locality in Kingstree, SC, may also be derived from the lower Waccamaw, as these are represented by teeth of Ondatra idahoensis, Holmesina floridanus, capybara, Nannippus pen- Table 3. Mammalian taxa of the Walrus Ditch Local Fauna and their known ranges. Megalonyx leptostomus early to late Blancan Eremotherium eomigrans early late Blancan – early Irvingtonian Pachyarmatherium leiseyi early late Blancan – early Irvingtonian Holmesina floridanus early late Blancan – early Irvingtonian Dasypus bellus early late Blancan – Rancholabrean Sylvilagus webbi late Blancan – earliest Irvingtonian Castoridae indet. Peromyscus sp. Barstovian – Recent Allophaiomys pliocaenicus latest Blancan – middle Irvingtonian Erethizon ?bathygnathum late middle Blancan – late Blancan Phugatherium? sp. early late Blancan -- Rancholabrean Smilodon gracilis late Blancan – middle Irvingtonian Canis lepophagus middle Blancan – late Blancan Arctodus pristinus late Blancan – late Irvingtonian Procyon sp. late Hemphillian – Recent Nannippus sp. late Clarendonian – late Blancan Equus sp. Blancan – late Rancholabrean Tapirus haysii late Blancan – middle Irvingtonian Mylohyus sp. latest Hemphillian – Irvingtonian Hemiauchenia macrocephala late Blancan – Rancholabrean Odocoileus virginiana late Blancan – Recent Rhynchotherium falconeri late Hemphillian – late Blancan ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 145 insulatus, and Cormohipparion sp. (SC2012.10.2). (In addition to these Blancan species and to the dinosaurs, plesiosaurs, and mosasaurs recovered there, the Clapp Creek site has also yielded evi- dence of later Pleistocene mammals including a tooth fragment of Megalonyx sp., a lower molar of Tapirus veroensis [SC2012.10.1], several teeth and tooth fragments from Equus sp., a partial lower molar of Cervus sp., and several fragments of ivory and enamel from proboscidean teeth [mammutid and/or gomphotheriid, not Mammuthus]). The upper Waccamaw, which is present at Blackwelder’s (1979:A54) lectostratotype locality (“180 m downstream from Tilly Lake on the Wac- camaw River, Horry County”) and at sites along the intracoastal waterway near Myrtle Beach in Horry County, has a reversed magnetic polarity (Cronin et al., 1984; also see Liddicoat and Opdyke, 1981). Together with Bybell’s (1990) determination that these sediments were deposited during the lower part of nannoplankton zone NN19, these data support referral of the upper Waccamaw to chron C1r2r (Figs. 2C, 3C). The upper Waccamaw, there- fore must be younger than the 1.77 Ma top of the Olduvai normal chron, but older than the 1.34 Ma LAD of the nannofossil Helicosphaera sellii noted from this unit by Cronin et al. (1984) and by Bybell (1990). McCartan et al. (1982:339) noted a 1977 personal communication from M. L. Bender that “corals from the Waccamaw Formation have 1.1– 1.4 my dates by the He/U method.” Campbell and Campbell (1995:66), noting the dominance of lit- toral and shallow sublittoral species in the upper Waccamaw, considered it a direct correlate of the upper Caloosahatchee of southern Florida, “which unconformably underlies the 1.5 Ma Bermont beds.” We place the upper Waccamaw Formation at 1.6 Ma, coincident with marine isotope stage 55 (Fig. 3C). Mammals noted from the “upper bed” of the Waccamaw by Sanders (2002:131) include Neofiber cf. N. diluvianus (referred in this report to Ondatra idahoensis; see discussion in “Systematic Paleontology” section), Cuvieronius sp., Miraci- nonyx inexpectatus, Hydrochoerus holmesi, and Tapirus haysii. Another locality in Dorchester County that may occur in the upper Waccamaw Formation consists of a series of closely situated sites collec- tively referred to as the Austin Sand Pits. Located near Ridgeville, SC, the fossil mammals collected there constitute the Ridgeville Local Fauna of Boessenecker et al. (2018), who consider it some- what younger than the Walrus Ditch LF, i.e., lat- est Blancan to earliest Irvingtonian. A faunal list provided in Boessenecker et al. (2018) includes material referred to the cetotheriid mysticete Her- petocetus sp., the Gray Whale Eschrichtius sp., cf. E. robustus, the Bowhead whale Balaena sp., cf. B. mysticetus, the extinct mysticete Balaenula sp., the Humpback whale Megaptera sp., the large rorqual Balaenoptera, the delphinid Astadelphis, and the early Sperm whale Physeterula sp. M. Gib- son of The Charleston Museum (pers. commun. to LBA, March, 2017) notes that material of another early sperm whale, Scaldicetus sp., is also known from the Austin Pit. Boessenecker et al. (2018) also reported the presence of the walrus Ontocetus emmonsi in the Ridgeville Local Fauna, as well as the terrestrial mammals Equus sp., Tapirus sp., cf. Cuvieronius (more likely Rhynchotherium based on age; see discussion of Rhynchotherium in “Sys- tematic Paleontology” section), Castoroides sp. (perhaps Procastoroides based on age), Neochoe- rus pinckneyi, Eremotherium sp. (likely E. eomi- grans based on age), and cf. Holmesina (likely H. floridanus based on age). They concluded that the Austin Pit Site occurs in sediments stratigraphi- cally above those that yield the Walrus Ditch LF and suggest that it may lie within the upper Wac- camaw Formation. Penholoway Formation (Uppermost Lower Pleistocene). The term Penholoway, in the geologi- cal sense, was originally applied by Cooke (1925) as a name for a marine terrace, the type area of which is in Wayne County, Georgia. Later, Cooke (1936, 1943, 1945) applied the name Penholoway Formation for the deposits underlying the terrace. Huddlestun (1988) abandoned “the lithostrati- graphic context of the Penholoway,” retaining the name for its original intent, the terrace. Edwards et al. (2000) maintained lithostratigraphic use of the term for the “coastal complex of estuarine, lagoonal, 146 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) barrier island, and shallow marine shelf deposits” that crop out in Wayne County, and also for correl- ative strata that occur across the SC Coastal Plain (unit Q5 of McCartan et al., 1990). Bybell (1990) suggested a date for the unit of between 700 and 925 kyr based on calcareous nannoplankton and uranium series dating of corals (Szabo, 1985), and this appears to have been refined by Edwards et al. (2000), who reported an age spanning the interval from 730 to 900 kyr. Weems et al. (1997) noted a 1984 personal communication from L. Bybell who considered the age of the Penholoway Forma- tion as no older than the upper part of Quaternary Zone NN19. The upper part of NN19 falls within the uppermost reversed portion of the Matuyama chron (C1r1r), and supporting this correlation is the reversed magnetic polarity of the Penholoway (see discussion above regarding polarity of “Wac- camaw” beds in the Charleston area). Given, there- fore, that the unit must be between 0.78 and 0.99 Ma based on magnetostratigraphy, and that it was apparently deposited during a major transgression (Weems et al., 1997), we suggest that deposition occurred during the significant sea level high at marine isotope stage 21, or about 850 kyr (Fig. 3C). The molluscan fauna of the Penholoway For- mation appears to be correlative with that of the stratotype James City Formation of DuBar and Sol- liday (1963), which is located on the south bank of the Neuse River in southeastern North Carolina. The James City Formation is a later name for the older of two “very different and unrelated lithic units” earlier referred to the “Croatan beds” of Dall (1890, 1892; see discussion in Blackwelder, 1981:B2–B3, B11–B12, and in Ward et al., 1991:281). The Pen- holoway and James City formations both contain nearly identical taxa, including many extant forms, and both also include the exceptionally rare gastro- pod Calliostoma erosa Dall, 1892, a taxon known only from its unique Croatan holotype until a Pen- holoway specimen was discovered in SC. The stra- totype James City Formation is not the same as the “James City Formation” of Ward and Blackwelder (1987). Dated to approximately 2.4 Ma, the lat- ter strata are located at Aurora, NC (McCartan et al., 1982), and they contain a fauna that is largely extinct. Sanders (2002) noted the following mam- mals from the Penholoway Formation: Dasypus bellus (early late Blancan to Rancholabrean), Ere- motherium sp., Miracinonyx inexpectatus, Equus sp., and Cervus elaphus. More recent analysis of the Eremotherium material by Fields et al. (2012) provided an updated referral to E. laurillardi, thus extending the Rancholabrean range of this taxon into the middle Irvingtonian. The stratigraphic con- text of a specimen of E. laurillardi from the Pen- holoway Formation (ChM PV4748, see Fields et al., 2012:6) is provided in Figure 12. Ladson/Canepatch Formation (Middle Pleistocene). The Ladson Formation was named by Malde (1959) for deposits near Ladson (Summer- ville area) that represent a coastal complex similar to that noted above for the Penholoway Formation. Edwards et al. (2000) noted the possible correlation of the Ladson with the Canepatch Formation of the Myrtle Beach area, and the unit is also correla- tive with unit Q4 of McCartan et al. (1990). Lid- dicoat and Opdyke (1981) noted a normal polar- ity (i.e., C1n) for the Canepatch Formation and, as noted in Bybell (1990), Szabo (1985) reported uranium-series ages from corals for the unit that clustered around 460 kyr. Szabo (1985:403, 405) further noted that the Canepatch Formation repre- sented deposition during an interglacial high sea level stand, which he correlated to MIS 11 rather than 13 “because oxygen isotope values suggest that the interglacial represented by oxygen stage 11 was warmer and of longer duration than was stage 13.” This has since been confirmed, with estimates of sea level at ~6 to 13 meters above present (see Raymo and Mitrovica, 2012; Candy et al., 2014; and references within). The more recent astronomi- cally tuned correlations (e.g., Lisiecki and Raymo, 2005) place MIS 11 at 410 kyr, and it is MIS 13 that is closer to 460 kyr. With error bars of ±100,000 yr on the coral dates, MIS 11 remains our favored cor- relation due to its greater magnitude than MIS 13 (Fig. 3C). The late Irvingtonian aspect of the mammals known from the Ladson Formation provides addi- tional support for a correlation to MIS 11. Sanders ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 147 (2002) noted the following terrestrial vertebrates from this unit: Megalonyx jeffersonii (late Irvingto- nian to Rancholabrean), Eremotherium laurillardi, Arctodus pristinus (late Blancan to late Irvingto- nian), Equus sp., and Tapirus veroensis (late Irving- tonian to Rancholabrean). In addition, however, is the spectacular assemblage of beautifully pre- served fossil mammals that comprise the Camelot Local Fauna (Kohn et al., 2005; Beaty et al., 2007; Fields, 2010; Barbiarz, et al., 2018). Considered most similar to the latest Irvingtonian (0.3–0.4 Ma) Coleman 2A Local Fauna of Florida (Martin, 1974; Morgan, 2005), Kohn et al. (2005:649) noted that “the site is one of the most productive and diverse middle Pleistocene sites in eastern North America outside of Florida, and it contains some of the best preserved examples of specific taxa such as saber- toothed cats (Smilodon fatalis) outside of the La Brea tar pits, California.” The Camelot assem- blage was recovered from fluvial sediments resting unconformably above the Tupelo Bay Formation at the Giant Cement quarry in Dorchester County that are considered equivalent to the Ladson Formation. The fauna includes Megalonyx jeffersonii (or a form transitional between M. wheatleyi and M. jef- fersonii per Fields, 2010), Holmesina septentriona- lis, Dasypus bellus, Didelphis virginiana, Scalopus sp., Lepus sp., Sylvilagus palustris, Sciurus caroli- nensis, Thomomys sp., Sigmodon bakeri, Neofiber ?alleni, Neoochoerus sp., Procyon lotor, Urocyon cinereoargenteus, Smilodon fatalis, Miracinonyx inexpectatus, Canis armbrusteri, Mylohyus sp., Platygonus sp., Paleolama mirifica, Hemiauchenia macrocephala, Odocoileus virginianus, Tapirus veroensis, and Equus sp. Ten Mile Hill Formation (Upper Middle Pleistocene). Another complex of fluvial, estua- rine, and barrier island deposits, the “Ten Mile Hill beds” of Weems and Lemon (1984a, b), and named for Sloan’s (1908) “sands on Ten Mile Hill,” were originally included within the Ladson Formation of Malde (1959). But Weems and Lemon (1984a, b) Figure 12. Stratigraphic context of ChM PV4748, Eremotherium laurillardi (marked by “X”), recovered from the Penholoway Formation near Trailwood Trailer Park, North Charleston, Charleston County (see Fields et al., 2012:6 for further details). 148 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) separated these beds out as a different unit “because they record a different and younger transgression and regression of the sea than that recorded by the beds of the Ladson type section.” This was con- firmed by Szabo’s (1985) uranium series dating of corals from these beds, indicating deposition from around 202 to 230 kyr ago. Deposition over this interval results in a correlation, as noted by Szabo (1985), to MIS 7 (Fig. 3C). The Ten Mile Hill beds are also equivalent to unit Q3 of McCartan et al. (1990), who considered this unit correlative (in part) with the Socastee Formation of the Myrtle Beach area (see below). Sanders et al. (2009) elevated these beds to formational status and noted their particular impor- tance insomuch as they yielded the oldest well dated remains of Bison in the conterminous USA. The appearance of Bison in North America south of the 55th parallel defines the beginning of the Rancholabrean NALMA, and prior to the recov- ery of the specimen from the Ten Mile Hill For- mation there was little temporally well-constrained material that might provide an accurate date for the Irvingtonian/Rancholabrean boundary (see Bell et al., 2004). Thus, we consider the fauna derived from this formation as the oldest in North America (south of the 55th parallel) representative of the Rancholabrean. In addition to Bison, Sanders et al. (2009) listed the following taxa from the Ten Mile Hill Formation: Dasypus bellus, Holmesina septentri- onalis, Eremotherium laurillardi, Odobenus sp., Hydrochoerus holmesi, Tapirus haysii, Equus sp., and Cuvieronius sp. In this report we add Sylvilagus palustris and Castor canadensis. Last occurring in Florida in the middle Irvingtonian, the record of T. haysii from the Ten Mile Hill Formation provides an extension of this taxon into the earliest Rancho- labrean and a relatively short temporal interval of overlap with T. veroensis, which first appears in Florida in the late Irvingtonian (Hulbert, 1995). Socastee Formation (Upper Pleistocene). The Socastee Formation of DuBar (1971) and DuBar et al. (1974) is represented by back-barrier or estuarine deposits that, according to McCartan et al. (1982:351), comprise “the major surficial unit along the [Intracoastal] Waterway” in the Myrtle Beach area. As noted by Szabo (1985), however, McCartan et al’s. (1982) interpretation of this unit was much broader than that of DuBar et al. (1980). The latter considered only the narrow, uppermost deposit of sand overlying the Canepatch Forma- tion (= Ladson Formation) along the waterway near the Route 501 bridge as Socastee, designating all the underlying sediment at that location to the Canepatch Formation. In an attempt to resolve this, Szabo (1985) analyzed samples from the Route 501 bridge locality for uranium-series dating, but the equivocal results were disregarded as unreli- able. Weems and Lemon (1993) consider this unit correlative with the lower member of the Wando Formation. Sanders (2002) noted the recovery of the dis- tal end of a humerus of the phocid seal Erignathus barbatus from a coarse sand deposit that evidently is exposed intermittently along the south bank of the Intracoastal Waterway toward the northeastern lim- its of Myrtle Beach. He also reported a partial den- tary of Monachus tropicalis from “along the Intra- coastal Waterway at the Possum Trot Golf Course near the town of Crescent Beach, Horry County, South Carolina ...” (Sanders, 2002:76). Although both specimens were reported as being recovered from the Socastee Formation, whether they came from correlative sediments cannot be determined; nor is it unequivocal that they were recovered from the Socastee Formation given the varying interpretation of this unit noted above. However, supporting assignment of these vertebrate-bearing beds to the Socastee Formation, rather than to the older Canepatch (= Ladson) Formation, is the fact that both species are currently extant; i.e., taxa that range from the Rancholabrean to Recent are some- what more likely to occur in Socastee deposits than in Canepatch (late Irvingtonian) deposits. Sand- ers (2002) also noted the questionable recovery of Holmesina septentrionalis and Tremarctos florida- nus from the Socastee, but these specimens, having been found on Myrtle Beach, may also have origi- nated from an undetermined late Pleistocene unit, or units, that lie offshore. Wando Formation (Upper Pleistocene). As noted by Edwards et al. (2000), this unit was origi- ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 149 nally named by McCartan et al. (1980) for outcrops near the Wando River representative of “a coastal complex of fluvial, estuarine, lagoonal, barrier- island, and shallow-marine shelf deposits.” It also includes the beds of shelly sand, clayey sand, and clay mapped as unit Q2 by McCartan et al. (1984; also see McCartan, 1990, and McCartan et al., 1990) that underlie the Pamlico and Prince Anne terraces of Colquhoun (1974). Szabo (1985) recog- nized two depositional episodes within the Wando which he referred to as “early Wando deposits” and “late Wando deposits.” He reported an aver- age age for the lower Wando of 129,000 ± 10 kyr and 87,000 ± 4 kyr for the upper, with a conse- quent correlation to oxygen isotope stage 5. Sand- ers (2002), following Weems and Lemon (1993), further divided the Wando into upper, middle, and lower members, and it is the middle member that he determined to be the unit of origin for most of the Charleston area’s fossil vertebrates collected from the famous “Ashley River phosphate beds.” The principle collection of fossils from the Ashley River phosphate beds curated at The Charleston Museum includes nearly 200 speci- mens that were provided by Charles C. Pinckney, Jr. From 1869 until 1910, Pinckney owned the Magnolia Phosphate Mine in the vicinity of Run- nymede Plantation about 11 miles northwest of Charleston along the Ashley River (Sanders 2002). In addition to Domning’s (1989b) discussion of the Charleston phosphate beds, Sanders (2002) pro- vided a thorough discussion of the Pinckney col- lection with details of its provenance and acquisi- tion. Sanders (2002) listed the following taxa from the Wando Formation: Megalonyx jeffersonii, Neo- choerus pinckneyi, Hydrochoerus holmesi, Castor canadensis, Neofiber alleni, Canis dirus, Arctodus pristinus (but see discussion below), Ursus ameri- canus, Odobenus rosmarus, Hemiachenia sp. cf. H. macrocephala, Rangifer sp. cf. R. tarandus, Cervalces scotti, Cuvieronius sp., and Mammut americanum. To this can be added Equus sp., Tapi- rus veroensis, Bootherium bombifrons, and Mam- muthus columbi. It has long been recognized that several species known from the Ashley River phosphate beds are much older than the Rancholabrean age implied by Szabo’s (1985) dates on the Wando For- mation, evidently having been reworked into this unit from subjacent strata. This was noted as early as Leidy (1877:210) in which he stated: “Besides the phosphate nodules, the Ashley beds present a remarkable intermixture of the remains of marine and terrestrial animals, consisting of bones, teeth, coprolites, shells, etc., derived from the contiguous formations of various ages from the early tertiary [sic] to those of a comparatively recent period.” Domning (1989b) also noted the temporally (and ecologically) mixed nature of the fossils from the phosphate beds and their likely stratigraphic prov- enance. In fact, the Wando Formation rests uncon- formably on a variety of older units including the Oligocene Ashley, Chandler Bridge, and Edisto formations, the lower Pliocene Goose Creek Lime- stone, and the Pleistocene Penholoway Formation. Some of the most dramatic examples of this reworking are specimens of late to latest Oligo- cene age including the horse Anchippus texanus, the entelodont Daeodon mento, the dugongids Dio- plotherium manigaulti and Crenatosiren olseni, and the early odontocete cetaceans Agorophius pyg- maeus and Xenorophus sloanii. Daeodon, and pos- sibly Anchippus, may be reworked from the Edisto Formation, whereas the dugongids and cetaceans are known from the Ashley and Chandler Bridge formations. Another cetacean from the phosphate beds, the beaked whale Choneziphius trachops Leidy, 1876a, is also known from the lower Mio- cene (Burdigalian) Pungo River Formation at the Lee Creek Mine, NC (Whitmore and Kaltenbach, 2008). Its presence in the Wando, therefore, sug- gests reworking from what may have been the Marks Head Formation based on the correlation of Ward (2008:fig. 3). The same may hold true for several additional, highly worn, fossil beaked whale specimens (noted previously) reported from the phosphate beds by Leidy (1877) and Allen (1926), as well as the holotype of another ziphiid, Anoplanassa forcipata Cope, 1869 (also see True, 1907), recovered during dredging operations for phosphate rock in the Coosaw River of Beaufort County. On the other hand, these specimens may 150 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) have also been reworked from the Goose Creek Limestone or perhaps from another, now missing Pliocene unit, such as the Wabasso beds or a bed correlative with the Sunken Meadow Member of the Yorktown Formation. This supposition is based on the abundant fossil material of beaked whales from the Pliocene of Italy (Bianucci, 1997) that very closely resembles specimens from SC, as well as on the recovery of another species, Tusciziphius crispus Bianucci, 1997, originaly described from the Pliocene of Italy, but which is now known from what is thought to be the Goose Creek Limestone from the bottom of the Morgan River in Beaufort County (Post et al., 2008). Another example may be Leidy’s (1854) holotype of the bear Arctodus pristinus. That now lost specimen consisted of a single isolated molar that was recovered from the Ashley phosphate beds (= Wando Formation) at Bee’s Ferry on the west bank of the Ashley River northwest of Charleston (Sanders, 2002). Discussed in detail in the “System- atic Paleontology” section on this taxon below, it is now thought to have been reworked into the Wando Formation from a more age-appropriate subjacent unit, as A. pristinus is no longer considered to have existed during the Rancholabrean (Emslie, 1995; Schubert, 2008; Schubert et al., 2010). More recent collections from the Wando Formation include the highly diverse assemblage of vertebrates that comprise the Crowfield Local Fauna (Chandler and Knight, 2009; Fields et al., 2012). Recovered as a result of the excavation of a lake in the Crowfield subdivision between Goose Creek and Summerville “along the Dorchester- Berkeley county line,” the fauna includes 55 iden- tified mammalian taxa, 11 species of birds, and “a substantial herpetofauna” (Chandler and Knight, 2009:143). Although not yet studied in detail, the mammals of the Crowfield Local Fauna are listed in Table 4. “Silver Bluff beds.” Thin deposits of sand, mud, and clay are present beneath a coastal terrace about 2 m above modern sea level known as the Silver Bluff terrace (Puri and Vernon, 1964). Rec- Table 4. Mammals of the Rancholabrean-aged Crowfield Local Fauna. Didelphis virginiana Castoroides dilophidus Ursus americanus Megalonyx jeffersonii Castor canadensis Lontra canadensis Holmesina septentrionalis Oryzomys sp. Spilogale putorius Dasypus bellus Peromyscus sp. Mephitis mephitis Sorex longirostris Sigmodon hispidus Neovison vison Sorex sp. cf. S. arcticus Neotoma floridana Procyon lotor Sorex sp. cf. S. palustris Microtus pennsylvanicus Odobenus rosmarus Microsorex hoyi Microtus pinetorum Equus sp. Blarina brevicauda Microtus ochrogaster Tapirus veroensis Blarina sp. Synaptomys cooperi Mylohyus sp. Cryptotis parva Synaptomys australis Hemiauchenia macrocephala Scalopus aquaticus Neofiber alleni Palaeolama mirifica Condylura cristata Ondatra zibethicus Rangifer tarandus Sylvilagus floridanus Erethizon dorsatum Cervus elephus Sylvilagus palustris Lynx rufus Odocoileus virginianus Sciurus carolinensis Urocyon cinereoargenteus Bison sp. Ictidomys tridecemlineatus Canis sp. Mammut americanum Glaucomys volans Tremarctos floridanus Trichechus manatus ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 151 ognized and mapped in Florida, Georgia, and South Carolina as far north as the vicinity of the Santee River, deposits immediately beneath this terrace are typically between 1.5 and 3 m thick, but locally can be up to 10.5 m thick in filled buried channels (Weems and Lemon, 1993). These deposits previ- ously have been assigned either to the uppermost Sangamon stage (MIS 5A) or the middle Wiscon- sian stage (MIS 3) (Colquhoun, 1974), although the bulk of this unit is probably middle Wisconsian in age based on a 14C date of 33,070 ± 1830 years (Beta Analytic Laboratory Report B-20188, 1979). This date was obtained from surf-polished wood chips found near the base of the unit in a sand pit in the Fort Moultrie quadrangle near the Atlantic coast (between localities FM 24 and FM 17 in Weems and Lemon, 1993). However, other samples of woody material recovered by augering 1.5 m below the terrace surface in the Fort Moultrie quadrangle (at locality FM 23) yielded 14C ages of only 7,860 ± 80 years (USGS Radiocarbon Laboratory Report W-5038, 1982) and of 6,960 ± 200 years (USGS Radiocarbon Laboratory Report W-5322, 1984). This suggests that in some areas there are local deposits of early Holocene age also present within this terrace complex. The “Silver Bluff beds” have not yet yielded any vertebrate remains that can be unequivocally assigned to them, but it is possible that Wisconsian-age vertebrates found along Edisto and Myrtle beaches derive from offshore deposits of this unit which were scoured from the coastal sea floor and transported to the shoreline. Additional deposits younger than the Wando Formation. From upper Pleistocene sediments that filled previously noted solution cavities eroded into the top of the upper Eocene Tupelo Bay Formation and into the overlying Harleyville Formation at the Giant Cement quarry, Bentley et al. (1994) recov- ered the Ardis Local Fauna. 14C dates place the fauna between 18,530 and 18,940 years old, which coincides with the last glacial maximum, and the 43 mammalian species support this. According to Bentley et al. (1994:1) they reflect “a more equi- table climate, cooler summers and warmer win- ters, than that presently occurring in the region.” Occasionally, additional specimens turn up in these upper Pleistocene sands, including a virtually com- plete skull of Canis dirus (USNM 437648) that was collected and donated to the USNM by R. Ogilvie (Sanders, 2002), as well as a mammoth molar col- Table 5. Mammals of the late Rancholabrean Ardis Local Fauna. Didelphis virginiana Peromyscus sp. Spilogale putorius Megalonyx jeffersonii Neotoma floridana Mephitis mephitis Holmesina septentrionalis Microtus pennsylvanicus Conepatus robustus Dasypus bellus Microtus pinetorum Neovison vison Sorex sp. cf. S. longirostris Synaptomys cooperi Procyon lotor Blarina brevicauda Synaptomys australis Equus sp. cf. E. complicatus Scalopus aquaticus Neofiber alleni Tapirus veroensis Condylura cristata Ondatra zibethicus Mylohyus nasutus Sylvilagus palustris Hydrocheoridae Palaeolama mirifica Sylvilagus floridanus Smilodon fatalis Odocoileus virginianus Sciurus carolinensis Lynx rufus Bison antiquus Ictidomys tridecemlineatus Urocyon cinereoargenteus Mammut americanum Glaucomys volans Canis dirus Mammuthus columbi Castor canadensis Tremarctos floridanus Oryzomys palustris Lontra canadensis 152 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) lected by JLK after a bulldozer destroyed a partial skeleton. A faunal list of mammals from the Ardis Local Fauna is provided in Table 5. In addition to the unit that yielded the Ardis Local Fauna are the “undetermined offshore units” from which the late Rancholabrean remains found on Edisto and Myrtle beaches originate (Table 6). As Sanders (2002:7) noted, these units were obvi- ously exposed “during the periods of extremely low sea level that accompanied Wisconsinan gla- ciation” (~18 to 60 kyrs), but to date there has been no underwater investigation seaward of these beaches that might provide more detailed infor- mation about these units. Roth and Laerm (1980) provided an account of the fossil mammals and turtles from Edisto Beach, with additions by Sand- ers (2002), but there has been no formal treatment of those from Myrtle Beach. Similarly aged verte- brate remains also wash ashore onto Fernandina, Jacksonville, and Ponte Vedra beaches in northern Florida, but as in SC, the unit of origin has not been determined, and there has been no formal treatment of the northern Florida Rancholabrean fauna. To date there are 36 known taxa in the Edisto Rancholabrean fauna including a new record of Miracinonyx trumani (discussed below). Five addi- tional taxa from Myrtle Beach include Holmesina septentrionalis, the first report of Canis latrans from SC, Tremarctos floridanus, Mylohyus fossi- lis, and Bootherium bombifrons. Most assuredly, remains of additional Rancholabrean taxa have been recovered from Myrtle Beach (and nearby beaches in Horry County), but these have not been reported or, as far as we know, curated into acces- sible collections. A similar situation exists regard- ing fossils from Edisto Beach, including a number of unstudied specimens held in private collections. Fossils representing several taxa of marine mammals are also recorded from Edisto Beach, and these include Tursiops truncatus (Bottlenose Dol- phin), Pseudorca crassidens (False Killer Whale), Physeter catodon (Sperm Whale), Monachus tropi- calis (Monk Seal), and Halichoerus grypus (Gray Seal) (Ray et al., 1968). However, these taxa seem more likely to have been eroded from Holocene sediments found at the top of an augered section drilled by AES and REW at Edisto Beach, rather than from offshore Rancholabrean deposits. The uppermost stratum consisted of 13 feet of shelly sand deposited during the Holocene, probably at the time of the last transgression to its present stand no more than 7600 years ago (Cronin et al., 2007). All are from extant taxa known to have occurred in western Atlantic waters by the Holocene. Table 6. Terrestrial mammals represented from fossils collected on Edisto and Myrtle beaches. Megalonyx leptostomus Miracinonyx trumani Monachus tropicalis Megalonyx jeffersonii Smilodon fatalis Odobenus rosmarus Eremotherium laurillardi Panthera atrox Equus sp. Paramylodon harlani Felis onca augusta Tapirus veroensis Holmesina septentrionalis Puma concolor Mylohyus fossilis Glyptotherium floridanum Lynx rufus Palaeolama mirifica Dasypus bellus Urocyon cinereoargenteus Cervus elephus Sylvilagus sp. Canis dirus Odocoileus virginianus Castoroides sp. Canis latrans Bison antiquus Castor canadensis Tremarctos floridanus Bootherium bombifrons Erethizon dorsatum Ursus americanus Mammuthus columbi Neochoerus pinckneyi Procyon lotor Mammut americanum Miracinonyx inexpectatus Halichoerus grypus Trichechus manatus ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 153 In addition to a marine component to the fauna from Edisto Beach and an apparent Holo- cene component, there are also remains of ter- restrial taxa older than those Rancholabrean taxa that comprise the bulk of the fauna (e.g., Megal- onyx leptostomus; see discussion below). Thus, the “Edisto Beach fauna” actually consists of a tempo- rally mixed assemblage. ADDITIONS TO THE CENOZOIC MAMMA- LIAN FAUNA OF SOUTH CAROLINA Locality information more detailed than that pro- vided below can be found on file in the institutions where the specimens are curated. The use of the question mark (“?”) in association with a taxo- nomic name follows Kornicker (1979). SYSTEMATIC PALEONTOLOGY MAMMALIA Linnaeus, 1758 MARSUPIALIA (Illiger, 1811) DIDELPHIDAE Gray, 1821 DIDELPHIS Linnaeus, 1758 DIDELPHIS VIRGINIANA Kerr, 1792 Figure 13A Referred Specimens.—SC75.31.70, left dentary with p3–m4; SC2003.75.293, right max- illary fragment with P3–M1; SC2003.75.294, left dentary fragment with p2–3, m2–3, partial m4; SC2003.75.295, left dentary fragment with m4; SC2003.75.296, left dentary fragment with p2-3, m1–4; SC2003.75.297, right dentary fragment with p1, p3; SC2003.75.298, left dentary frag- ment with p3, m4; SC2004.1.239, right dentary with c, p2–3, m1–4; SC2004.1.240, right dentary with p3, partial m1 and m3, m4; SC2003.75.693, left humerus; SC2003.75.694, partial left humerus; ChM PV7692, atlas vertebra. Locality, Stratigraphic Horizon, and Age.— SC75.31.70 from Cooper River, ?Wando Forma- tion, late Pleistocene, Rancholabrean; all other SCSM specimens from Camelot locality, Dorches- ter County, Ladson Formation, middle Pleistocene, late Irvingtonian; ChM PV7692 from “near Bee’s Ferry Rd.,” Charleston County, Wando Formation, late Pleistocene, Rancholabrean. Discussion.—Bentley et al. (1994) first reported fossil remains of the Virginia opossum in SC from the very late Rancholabrean Ardis Local Fauna (18–19 kyrs), and Morgan (2002) noted that the earliest records of D. virginiana in North Amer- ica are from two late Irvingtonian faunas in Flor- ida. Although not identified to species, older speci- mens of the genus were recorded from the middle Irvingtonian Fyllan Cave LF, Texas, by Winkler and Gose (2003). These late Irvingtonian records from Florida are now matched in South Carolina by the presence of D. virginiana in the similarly aged Camelot Local Fauna. ChM PV7692 was col- lected by G. Pettus in 2006 and is considered to have originated from the Wando Formation on the basis of the proximity of Bee’s Ferry Road to the old Charleston phosphate mining district. PLACENTALIA Owen, 1837 XENARTHRA Cope, 1889 PILOSA Flower, 1883 MEGALONYCHIDAE Gervais, 1855 MEGALONYX Harlan, 1825 MEGALONYX LEPTOSTOMUS Cope, 1893 Figure 13B Referred Specimens.—See Fields et al. (2012). Locality, Stratigraphic Horizon, and Age.— See Fields et al. (2012). Discussion.—In a recent paper on all the ground sloth material known from South Carolina, Fields et al. (2012) noted the occurrence of Meg- alonyx leptostomus from two localities, including several elements from the late Blancan Walrus Ditch LF of Dorchester County. A fragment of humerus from the bottom of the Cooper River in Berkeley County is thought to have originated from the ~3.8 myr old Goose Creek Limestone based on the nature of the matrix adhering to the specimen. The oldest occurrence of Megalonyx in the USA is in the early late Hemphillian (Hh3, 6–7 Ma) ZX Bar Local Fauna, Nebraska (Hirschfeld and Webb, 1968; Morgan, 2005). In the southeast- ern USA, Megalonyx is first recorded in the latest Hemphillian (Hh4, 4.7–5 Ma) Palmetto Fauna of Florida as M. curvidens (Morgan, 2005; Webb et al., 2008). The oldest and youngest records of M. lep- 154 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 155 tostomus occur, respectively, in the Hagerman Local Fauna, Idaho, at a level considered approxi- mately 3.98 Ma by D. Ruez (pers. comm. to LBA, 2008), and in the Inglis 1A site, Florida (1.8 Ma), apparently confining this taxon to the Blancan NALMA (Bell et al., 2004; Morgan, 2005). Thus, as Fields et al. (2012) noted, the partial humerus (ChM PV7681) purportedly from the Goose Creek Limestone represents the oldest record of a ground sloth in the state, and for the first time confirms the presence of this species in the Southeastern USA during the early Blancan. A fragment of cheek tooth (SC2008.8.16) from Edisto Beach is similar in size and morphology to those of M. leptostomus, and JLK has observed additional teeth from Edisto Beach the size of M. leptostomus in a private collection. However, given that the “Edisto Beach fauna” per se is of Rancho- labrean age, the possibility exists that these teeth belonged to a juvenile or subadult species of M. jeffersonii. The Edisto Beach “fauna” comprises a temporally mixed assemblage based on the recov- ery of both terrestrial and marine fossils, and on taxa of apparent Holocene age, as well. CINGULATA Illinger, 1811 GLYPTODONTIDAE Gray, 1869 GLYPTOTHERIUM Osborn, 1903 GLYPTOTHERIUM TEXANUM Osborn, 1903 Figure 13C Referred Specimen.—SC 90.121.1, carapa- cial osteoderm. Locality, Stratigraphic Horizon, and Age.— Collected by R. Ogilvie in a spoil pile that con- sisted of a “bed of white sand” near the town of Florence, Florence County, about 0.34 km north of the intersection of US Route 52/301 and County Road 107 (= Alligator Road). Formation and age unknown, but possibly from the Duplin Formation. Description and Discussion.—Following the revised taxonomy of Gillette et al. (2016), Glypto- therium texanum is the valid species name for Blan- can and Irvingtonian glyptodonts in the southeast- ern USA, superseding the usage of Glyptotherium arizonae advocated by Gillette and Ray (1981). SC 90.121.1 is a large carapacial osteoderm, measur- ing about 61 mm in diameter by 20.6 mm thick. It matches osteoderms referred to G. arizonae by Gillette and Ray (1981:13) in its large size, flat to weakly convex external surface, and in the central figure occupying greater than half the scute diam- eter (about 32 mm; distinctively broader than the relatively narrow peripherals). In G. floridanum the central figure of carapacial osteoderms is “approxi- mately equal in size to peripherals, usually slightly raised and weakly concave” (Gillette and Ray, 1981:15). SC 90.121.1 was found on a spoil pile of sed- iments described as a white sand by the collector. These sediments were originally thought to match those representative of the Duplin Formation from a site near Darlington, SC, (~16 km WNW of the Florence site) where a specimen of Phugath- erium dichroplax (discussed below) was found. This would have resulted in the oldest record of Glyptotherium in the USA. Further inquiry regard- ing the provenance of this specimen, however, revealed that the sediment from the two sites was Figure 13. A, Didelphis virginiana, SC2004.1.239, right dentary with c, p2–3, m1–4, from Camelot locality, Ladson Formation; B, Megalonyx leptostomus, ChM PV7681, partial left humerus from Cooper River, Goose Creek Limestone; C, Glyptotherium texanum, SC90.121.1, carapacial osteoderm, from ?Duplin Formation; D, Pachyarmatherium leiseyi, SC2006.1.123, carapacial osteoderm, from Walrus Ditch locality, Waccamaw Formation; E, Holmesina floridanus, SC2006.1.19, partial carapacial osteoderm, from Walrus Ditch locality, Waccamaw Formation; F, Holmesina septentrionalis, SC2003.75.129, movable osteoderm from Camelot locality, Ladson Formation; G, Sylvilagus palustris, ChM PV7675, labial view of left dentary with incisor plus p3–m3, from Ten Mile Hill Formation; H, same specimen, occlusal view; I, Castoroides dilophidus, SC2017.10.25, distal right humerus from Broad River, Beaufort County; J, Castoroides dilophidus, SC2015.53.5, right metatarsal III from Broad River, Beaufort County, in (left to right) anterior, lateral, posterior, and medial view. 156 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) not similar, leaving the geologic unit from which the Glyptotherium scute originated unresolved. Interestingly, the Glyptotherium locality is also the site from which Late Cretaceous (Maastrichtian) reptile material was recovered (Schwimmer et al., 2015), the “white sands” evidently being the unit capping the upper Cretaceous Peedee Formation. The oldest USA records of Glyptotherium are those from early late Blancan sites in the South- west, such as Cita Canyon, Texas, and 111 Ranch, Arizona, which occur magnetostratigraphically just below the Gauss-Matuyama boundary at about 2.6–2.7 Ma (Bell et al., 2004; Morgan and White, 2005; Morgan, 2005, 2008). Even older records are known from the Guanajuato region of central Mex- ico (Carranza-Castenada and Miller, 1988) where Flynn et al. (2005:304) dated the strata yielding these specimens to about 3.6 Ma. Younger records from the Southeast include specimens from late Blancan sites in Florida such as the De Soto Shell Pit and the Santa Fe River 1 faunas, where Morgan (2005:292) noted that G. arizonae co-occurs with Nannippus, and also from the early Irvingtonian Leisey site. GLYPTOTHERIUM FLORIDANUM (Simpson, 1929) Referred Specimens.—ChM PV2415, post- glenoid cranial fragment; ChM PV2417, 2418, 2090, isolated carapace osteoderms (Gillette and Ray, 1981); ChM PV4879, osteoderm; ChM PV4880, osteoderm; SC94.57.8b, osteoderm. Locality, Stratigraphic Horizon, and Age.— ChM PV2415, 2417, 2418, 2090, 4879, and SC94.57.8b from Edisto Beach, Colleton County, undetermined offshore unit; ChM PV4880 from Garden City Beach, Horry County, undetermined offshore unit, late Pleistocene, Rancholabrean. Discussion.—These elements are referred to G. floridanum on the basis of their much smaller size relative to those of G. texanum, and on the diameter of the central figure (about 17 mm), which is less than half the total scute diameter (about 43 mm). They are noted here to complement those specimens previously reported by Ray (1965) and Roth and Laerm (1980). Taxonomy of Glyptoth- erium species here follows Gillette et al. (2016); some favor synonymy of G. floridanum with a spe- cies named from Mexico (Ramírez-Cruz and Mon- tellano-Ballesteros, 2014; Zurita et al., 2018). PACHYARMATHERIIDAE Fernincola et al., 2018 PACHYARMATHERIUM Downing and White, 1995 PACHYARMATHERIUM LEISEYI Downing and White, 1995 Figure 13D Referred Specimens.—SC2006.1.123, osteo- derm (plus several others with SC2006.1 prefix). Locality, Stratigraphic Horizon, and Age.— Walrus Ditch locality, Dorchester County, lower Waccamaw Formation, early Pleistocene, late Blancan. Discussion—The specimens noted here from the Walrus Ditch locality were mentioned, but not discussed in detail, in Downing’s and White’s (1995) original description of Pachyarmatherium leiseyi from the early Irvingtonian Leisey Shell Pit Local Fauna of Florida. Although specimens are also known from Costa Rica (Laurito et al., 2005), the Walrus Ditch material provides the first USA record of this taxon outside of Florida. A more detailed account of the SC material will be pre- sented upon completion of the study on the Walrus Ditch LF. PAMPATHERIIDAE Paula Couto, 1954 HOLMESINA Simpson, 1930 HOLMESINA FLORIDANUS (Robertson, 1976) Figure 13E Kraglievichia floridanus Robertson, 1976. Holmesina floridanus (Robertson). Edmond, 1987; Hulbert and Morgan, 1993; Downing and White, 1995. Referred Specimens.—ChM PV7596, tooth 4; SC2006.1.19, partial imbricating osteoderm. Locality, Stratigraphic Horizon, and Age.— ChM PV7596 from Clapp Creek locality, King- stree, Williamsburg County, ?lower Waccamaw Formation; SC2006.1.19 from Walrus Ditch local- ity, Dorchester County, lower Waccamaw Forma- tion, early Pleistocene, late Blancan. Discussion.—These specimens represent the first records of H. floridanus from SC, a taxon known previously from late Blancan to early Irvingtonian sites in Florida (Hulbert and Morgan, 1993; Morgan, 2005). The small size of the osteo- ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 157 derm (21.2 mm wide by 6.6 mm thick; length inde- terminable due to breakage), together with other elements of the Walrus Ditch Local Fauna indica- tive of a late Blancan age, supports referral to H. floridanus rather than the much larger and later occurring H. septentrionalis (Fig. 13F). The latter is known from the late Irvingtonian Camelot Local Fauna, in addition to several unnamed sites in the Summerville area. The tooth from the Clapp Creek locality is 12 mm long by about 5 mm wide, which is slightly smaller than material described from the early Irvingtonian Leisey Shell Pit Local Fauna, Florida (Downing and White, 1995; Hulbert and Morgan, 1993; Morgan and Hulbert, 1995). As noted in Downing and White (1995), Edmund (1987) quantified a trend in Holmesina from small- est size in Blancan species to increasingly larger size through the Irvingtonian and Rancholabrean. The slightly smaller size of ChM PV7596 relative to the Leisey specimens, together with the recov- ery of Ondatra idahoensis and Nannippus penin- sulatus (discussed below), supports our late Blan- can assignment for many of the mammalian fossils from the Clapp Creek site. LAGOMORPHA Brandt, 1855 LEPORIDAE Gray, 1821 SYLVILAGUS Gray, 1867 SYLVILAGUS PALUSTRIS Chapman and Willner, 1981 Figure 13G–H Referred Specimens.—ChM PV7675, left dentary with incisor plus p3–m3; ChM PV7733, right dentary with incisor plus p3–m2. Locality, Stratigraphic Horizon, and Age.— ChM PV7675 from Tall Pines subdivision, Berke- ley County, Ten Mile Hill Formation, late middle Pleistocene, early Rancholabrean; ChM PV7733 from Trailwood Trailer Park, Charleston County, Penholoway Formation, late early Pleistocene, middle Irvingtonian. Description.—Descriptive terminology of leporid dentition follows White (1991). In leporids the p3 is the most diagnostic tooth with which to determine species. In PV7675, the p3 measures 3.9 mm AP by 3.02 mm TR. The anterior surface of the tooth has multiple anterior reentrants, the antero- external reentrant (AER) is shallow and heavily cemented, and the posteroexternal reentrant (PER) extends across the entire occlusal surface. The thick enamel of the anterior wall of the PER has no cren- ulations, whereas the thin enamel of the posterior wall is highly crenulated, as in p4–m2. Measuring along the occlusal surfaces, the length of p3–m3 = 14.8 mm. PV7733 is smaller than PV7675, but of similar morphology with the exception that the anterior surface of p3 is not as highly crenulated. The AP length of p3 measures 3.3 mm and the length of p3–m2 = 11.5 mm. Discussion.—The extant marsh rabbit Syl- vilagus palustris differs from the extinct species S. hibbardi, known from Blancan and Irvingtonian sites in western North America, in its larger size, greater number of anterior reentrants, and exten- sion of the PER to the lingual border. It closely resembles the extinct S. webbi, from the Blancan and very early Irvingtonian of Florida, but differs in having a PER that extends across the tooth to the lingual border (White, 1991). White (1991) sug- gested that S. palustris may have evolved from S. webbi. Bentley et al. (1994) also noted this species from the late Rancholabrean Ardis Local Fauna. RODENTIA Bowdich, 1821 CASTORIDAE Hemprich, 1820 CASTOROIDES Foster, 1838 CASTOROIDES DILOPHIDUS (Martin, 1969) Figure 13I–J Castoroides ohioensis dilophidus Martin, 1969; Martin, 1975. Castoroides leiseyorum Morgan and White, 1995; Parmalee and Graham, 2002. Castoroides dilophidus (Martin, 1969). Hulbert, Kerner, and Morgan, 2014. Referred Specimens.—SC75.33.1, nearly complete cranium; SC2016.1.24, fragment of upper incisor; SC2015.17.6, right p4; USNM PAL 530187, two cheek teeth; SC2017.10.25, distal end of right humerus; SC2015.53.5, right metatarsal III. Locality, Stratigraphic Horizon, and Age.— SC2016.1.24, and SC2017.10.25 from Broad River, seaward of Edward B. Rogers Bridge (SC Highway 170) and SC2015.53.5 from Whale Branch region of Broad River inland from Edward B. Rogers 158 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) Bridge, Beaufort County, likely derived from a unit equivalent to the Wando Formation, late Pleisto- cene, Rancholabrean; SC75.33.1 and SC2015.17.6 from Cooper River, Berkeley County, thought to be derived from the Wando Formation, late Pleis- tocene, Rancholabrean; USNM PAL 530187 from Myrtle Beach, Horry County, from undetermined offshore upper Pleistocene unit, Rancholabrean. Discussion.—Parmalee and Graham (2002) provided a full description, plus illustrations, of the skull, SC75.33.1, referring it to Castoroides leisey- orum. Recently, Hulbert et al. (2014) determined that C. leiseyorum is the junior synonym of C. dilophidus (Martin, 1969). This, in turn, resulted in their referral of the other known specimens of Cas- toroides from Florida, Georgia, and SC (the South- eastern Coastal Plain) to C. dilophidus, noting several characters that distinguish it from C. ohio- ensis, now putatively restricted to the northeastern and mid-continental USA and Canada. Although SC2015.17.6, the right p4 from the Cooper River, does not have the “dilophid” pattern in which the second anterior lophid of the p4 is divided as do many of the specimens from Florida, Hulbert et al. (2014) do not consider this morphology as neces- sarily diagnostic or characteristic of C. dilophidus because of individual variation of this feature. It is interesting to note that SC75.33.1 exhib- its nearly identical preservation to that of three other specimens recovered from the same general area of the Cooper River: the cranial material of Neochoe- rus pinckneyi (ChM PV2796) noted in Sanders (2002) and discussed further below, the well pre- served skull of Tapirus veroensis (ChM PV4257) described by Ray and Sanders (1984), and the den- tary of Canis dirus (SC83.118.1) also noted below. All four specimens are brown in color, not strongly permineralized, and essentially pristine in preser- vation. Although their exact stratigraphic origin is unknown, Sanders (2002) provided information based on USGS mapping in the region to support a Wando Formation provenance. However, also noteworthy is the discussion by Sanders (2002:13) regarding the preservation of vertebrate material in the late Pleistocene deposits of the Charleston area: There are four Pleistocene units within a ten mile radius of Runnymede Plantation, viz, the early Pleistocene Penholoway Formation, the Ladson Formation and the Ten Mile Hill Beds, both of middle Pleistocene age, and the late Pleistocene Wando Formation. Fossil bones found in place in the Penholoway Formation are medium to dark brown in color, while those from the Ladson and Ten Mile Hill Beds are usually of light brown or buff color, sometimes with light orange or black iron stains. As noted in the Meg- alonyx account in the present paper, the only specimens yet found in place in the Wando Formation are well mineralized and almost uniformly black, particularly those from the lag deposit at the base. Although difficult to confirm, the above statement suggests a unit of origin for these four specimens other than the Wando Formation. The color of the specimens suggests a Penholoway origin, and the early Irvingtonian to late Rancho- labrean range of Castoroides dilophidus does not preclude this suggestion. Nor do the ranges of Tapirus veroensis and Canis dirus (late Irving- tonian to Rancholabrean) or Neochoerus pinck- neyi (Blancan to Rancholabrean). Their excellent preservation indicates (1) that they had not been long on the floor of the river since eroding from their entrapping matrix, (2) that they certainly had not been transported very far from their point of origin, and (3) that they did not originate from a basal lag deposit. Plans are currently underway to resume scuba diving efforts at the localities where these specimens were recovered in hopes of shed- ding new light on the accuracy of our hypotheses regarding their geological provenance. CASTOR Linnaeus, 1758 CASTOR CANADENIS Kuhl, 1820 Referred Specimen.—ChM PV5027, incisor; ChM PV9635, distal end of right humerus. Locality, Stratigraphic Horizon, and Age.— ChM PV5027, collected by V. McCollum from a ditch adjacent to County Road 199, Dorches- ter County, Ten Mile Hill Formation, late middle Pleistocene, early Rancholabrean; ChM PV9635, ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 159 Myrtle Beach, Horry County, from undetermined offshore upper Pleistocene unit, Rancholabrean. Discussion.—Whereas Castor canadensis is relatively common in the fossil record of South Carolina, we note these specimens simply as addi- tions to the known material from the state. ChM PV5027 represents the first occurrence of Castor from the Ten Mile Hill Formation. A label with ChM PV9635 identified that specimen as belong- ing to Castoroides, but the specimen is not nearly the size of the giant beaver; it is similar in size to that of Castor, and, as in the latter taxon, the ele- ment lacks the entepicondylar foramen. ARVICOLIDAE Gray, 1821 ONDATRINAE Repenning, 1982 ONDATRA Link, 1795 ONDATRA IDAHOENSIS Wilson, 1933 Figures 14A–I Ondatra idahoensis Wilson, 1933. Ondatra idahoensis Wilson. Hibbard, 1959; Martin, 1972; Eshelman, 1975; Repenning et al., 1995; Albright, 1999a. Ondatra annectens (Brown). Repenning and Grady, 1988. Neofiber cf. N. diluvianus (Cope). Sanders, 2002. Referred Specimens.—ChM PV7579, left m1; SC87.158 (accession number only), unassoci- ated right and left m3. Locality, Stratigraphic Horizon, and Age.— ChM PV7579 and SC87.158 from Clapp Creek site in Kingstree, Williamsburg County, ?lower Wacca- maw Formation, early Pleistocene, late Blancan. Description.—The m1 from Clapp Creek is nearly identical to those from the latest Blancan Ing- lis 1A Local Fauna, Florida, and with an AP mea- surement of 4.3 mm, it falls within the 4.2–4.8 mm size range of the Inglis sample (Morgan and White, 1995:439). The tooth has a posterior loop followed by five alternating triangles and an anteroconid (or anterior cap). The anteroconid has a well-devel- oped lingual reentrant (the fifth lingual reentrant) that isolates a seventh triangle, but it is not nearly as deep as the four lingual reentrants posterior to it. The opposing sixth triangle, on the labial side of the anteroconid, is not as distinct as the second and fourth triangles posterior to it due to the absence of an anterolabial (fourth labial) reentrant. Thus there are three (prominent) labial reentrants and five lin- gual reentrants, the fifth somewhat less developed than the others. All dentinal commissures are very narrow and nearly equal in width. The tooth has well developed roots, enamel of equal width every- where except on the anterolabial surface, and little if any cement in the reentrants. Both labial and lingual dentine tracts are well developed, although the former are more so than the latter. As in many arvicolids, including several of the Inglis 1A speci- mens, the only dentine that extends up the crown to the occlusal surface in these teeth is that at the labial termination of the posterior loop. The m2 (based on specimens from North Carolina; see discussion below) has a posterior loop followed by three alternating triangles, with a smaller fourth anterior triangle. Thus there are two lingual and labial reentrants. As in the m1, the reentrants are very narrow and the dental commis- sures between triangles are nearly closed except for a very narrow commissure between the third and fourth triangles. The enamel is of equal width everywhere, and there is minimal cement in the reentrants. The m3 has a posterior loop followed by three alternating triangles, the third being the ante- rior triangle; there is only a single labial reentrant. The labial dentine tracts extend up to the occlusal surface. The lingual dentine tract of triangle two extends approximately half way up the crown; that of triangle three extends about one-third the way up the crown. There is another dentine tract on the anterior face of the third triangle that extends about two-thirds the way up the crown. The left m3 asso- ciated with SC87.158 measures about 2.9 mm AP with a crown height of about 5.3 mm. Discussion.—ChM PV7579 and SC87.158 represent the first records of Ondatra idahoensis from South Carolina. Germane to this discussion, however, are two specimens from North Carolina: ChM PV5070, a partial right dentary with i1, m1, m2, and root of m3 and ChM PV5398, a partial right dentary with the base of i1, m1, and m2. These specimens were collected from the approxi- mately 1.6 Ma upper Waccamaw Formation (mid- dle early Pleistocene, latest Blancan) at Marsh Harbor Marina on the Intracoastal Waterway near Calabash, Brunswick County, North Carolina (see Sanders, 2002:85). Although Sanders (2002) origi- 160 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 161 nally referred these specimens to Neofiber cf. N. diluvianus, more recent analysis has determined that they differ considerably from the poorly known middle to late Irvingtonian N. diluvianus, the mid- dle to late Irvingtonian (1.0–0.6 Ma) N. leonardi, and the latest Irvingtonian (~0.4 Ma) to Recent N. alleni. All the aforementioned round-tailed musk- rats have ever-growing teeth, ample cement in the reentrants, and well-developed dentine tracts. The Waccamaw and Clapp Creek specimens show none of these derived features; close inspection of the NC specimens has determined that they have rooted teeth (contra Sanders, 2002) and only very limited cement in the reentrants, if any. For this reason we herein revise the identification of ChM PV5070 and PV5398 from Neofiber cf. N. diluvia- nus to Ondatra idahoensis, an assignment much more temporally and taxonomically consistent with their morphology and their recovery from the Wac- camaw Formation. Antero-posterior measurements of the m1s of ChM PV5070 and 5398 are 4.8 mm and 4.7 mm, respectively; thus, they too fall within the 4.2–4.8 mm size range of the Inglis 1A sample (Morgan and White, 1995:439). It is interesting that Sanders (2002:90) com- mented on the inconsistency of finding an arvico- line rodent, his Neofiber cf. N. diluvianus, with what he thought were rootless teeth in sediments of such early age “inasmuch as the latter specimens appear to be as old or slightly older than the pro- posed ancestor of Neofiber” (i.e., Proneofiber from the 1.4–1.6 Ma Gilliland Local Fauna of Texas [Hibbard and Dalquest, 1973]). Our new findings explain this temporal/morphological inconsistency insomuch as the Waccamaw specimens are not members of either Neofiber or Proneofiber. Like so many of the terrestrial Miocene and Pliocene vertebrate fossils from SC found in lag deposits, the exact formation from which the Clapp Creek specimens originated cannot be determined. The oldest occurrence of O. idahoensis is in the Tusker (111 Ranch) LF, Arizona, and the Cita Can- yon LF, Texas, where it occurs in normally mag- netized strata just below the Gauss-Matuyama boundary at about 2.6 Ma, which is also the base of Blancan V within which O. idahoensis is a char- acterizing taxon (Bell et al., 2004). Its latest occur- rence is in the 1.3–1.4 Ma El Casco Local Fauna, California (Albright, 1999a). However, additional taxa from the Clapp Creek locality, such as the dwarf horse Nannippus, confirm the late Blancan age of some material from this site, in turn suggest- ing derivation from the Waccamaw Formation. ONDATRA ZIBETHICUS (Linnaeus, 1766) Referred Specimen.—ChM 55.103.117, right dentary with partial i1 plus m1–m3, and humerus. Locality, Stratigraphic Horizon, and Age.— The only information accompanying this specimen is the following: “W. B. Cormark. Loc. Santee & Cooper rivers. June, 1941. Identified by Gilmore, U.S.N. Museum.” ?Wando Formation; late Pleisto- cene, Rancholabrean. Description and Discussion.—The m1 of ChM 55.103.117 has a posterior loop followed by seven alternating triangles and an anteroconid. The anteroconid has a weakly-developed lingual reen- trant and no anterolabial reentrant. There are five lingual reentrants (excluding the weak one of the anteroconid) and four labial reentrants. All trian- gles are closed (i.e., there are no dentinal commis- sures between them) and enamel thickness on their Figure 14. A–F, Ondatra idahoensis from Clapp Creek site, Kingstree, SC. A, B, C, ChM PV7579, left m1, in (left to right) occlusal, labial, and lingual views, respectively; D, SC87.158, right m3, in occlusal view; E, F, SC87.158, left m3, in (left to right) lingual and labial views, respectively. G, H, I, Ondatra idahoensis from Waccamaw Formation, Brunswick County, NC. ChM PV5070, right dentary, in occlusal (top), labial (middle), and lingual (lower) views, respectively; J, Allophaiomys pliocaenicus, SC2001.51.4, right m1, from Walrus Ditch site, Waccamaw Formation; K, Erethizon ?bathygnathum, SC2006.1.105, right M1, from Walrus Ditch site; L, Phugatherium dichroplax, USNM 181573, left M3, occlusal view, from Duplin Formation; M, Neochoerus pinckneyi, left M2, from Cooper River; N, Phugatherium dichroplax, USNM 181640, right metacarpal III, from Duplin Formation. 162 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) posterior and anterior surfaces is equal. This tooth, as well as the m2 and m3, has no cement in the reentrants, no dentine tracts that reach the occlu- sal surface, and were ever-growing. The length of the m1 is 7.6 mm. The m2 has a posterior loop followed by four alternating triangles, with two lingual and two labial reentrants; it measures 4.0 mm in length. The m3 is similar to the m2 except smaller, measuring 3.5 mm. A right humerus cata- logued together with this specimen under the same number, although similar in color, appears to be that of a recent individual. The extant muskrat, Ondatra zibethicus, was noted by Bentley et al. (1994) to be the most com- mon mammal of the latest Rancholabrean Ardis Local Fauna (in which they also noted the first record from SC of the round-tailed muskrat Neo- fiber alleni). The vague information accompany- ing ChM 55.103.117 precludes an accurate deter- mination of the unit of origin, and therefore age, of this specimen, but its size is indicative of very late Pleistocene populations (Martin et al., 2009); hence, the tentative referral of this specimen from the Wando Formation. ARVICOLINAE Bonaparte, 1837 ARVICOLINI Kretzoi, 1954 ALLOPHAIOMYS Kormos, 1932 ALLOPHAIOMYS PLIOCAENICUS Kormos, 1932 Figure 14J Referred Specimen.—SC2001.51.4, right m1. Locality, Stratigraphic Horizon, and Age.— Walrus Ditch locality, Dorchester County, lower Waccamaw Formation, early Pleistocene, late Blancan. Description.—In his revised diagnosis of the Tribe Arvicolini, Repenning (1992:24) noted “a first lower molar with posterior loop preceded successively by three substantially closed and alternating triangles and terminated by an Antero- conid Complex with confluent, lingual and buccal Primary Wings and an uncomplicated, more or less globular Cap.” He further noted that Allophaio- mys was one of only three rootless genera in the Tribe Arvicolini, and that enamel-thickness dif- ferentiation was typically intermediate (i.e., equal thickness on the anterior and posterior edges of the triangles) vs. “Mimomys” or “Microtus” like. This diagnosis exactly describes the morphology of SC2001.51.4. The tooth measures 3.1 mm AP by 1.4 mm TR. Although cement is present in the reentrants, the tooth would not be considered heav- ily cemented. (Note: Storer [2003] subscribes to a different interpretation of Allophaiomys in which included specimens retain rooted cheek teeth). Discussion.—According to Repenning and Grady (1988), Allophaiomys “is known from per- haps 2.5 million years ago in Asia.” But in North America the oldest occurrence is based on a speci- men (not necessarily diagnostic of the genus; see Bell et al., 2004:279) recovered from a core drilled at the foot of Hansen Bluff, Colorado, in the east- ern part of the Rocky Mountains that Repenning (1992:32) placed “either during the oldest part of the Olduvai event or just before it,” indicating a date for that specimen of approximately 1.95 to 2.0 Ma. More diagnostic specimens of Allophaiomys occur in the Nash Fauna of Kansas, which may be slightly older (see Bell et al., 2004), although Mar- tin et al. (2008:202) considered the appearance of “Microtus cf pliocaenicus” in the Short Haul and Aries A local faunas, also in Kansas, at “about 2.0 Ma” as the oldest records. At the time of Repen- ning’s (1992) publication, Allophaiomys was only known from east of the Rocky Mountains. It has since been noted from the Irvingtonian of Nevada (Bell et al., 2004:280). Martin et al. (2008) recommended “the Micro- tus immigration event,” constrained in the Meade Basin of southwestern Kansas to between 1.95 and 2.06 Ma, as a replacement for the first appearance of Mammuthus to define the Blancan/Irvingtonian NALMA boundary. The record of Allophaiomys in the late Blancan Walrus Ditch Local Fauna, dated to between 2.3 and 2.4 Ma, presents an interesting record, as it represents an even older appearance (the oldest in North America) and a geographic extension resulting in its eastern-most and south- ern-most occurrence. ERETHIZONTIDAE Thomas, 1897 ERETHIZON Cuvier, 1822 ERETHIZON ?BATHYGNATHUM Wilson, 1935 Figure 14K ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 163 Erethizon bathygnathum Wilson, 1935. ?Coendou brachignathum [sic] (Wilson). White, 1968. Coendou stirtoni White, 1968. Coendou brachygnathum [sic] (Wilson). White, 1970. Erethizon bathygnathum Wilson. Shotwell, 1970; Albright, 1999a; Sussman et al., 2016. Coendou stirtoni White, 1968. Harrison, 1978. Referred Specimens.—SC2006.1.105, right M1; SC2001.51.5, left m1. Locality, Stratigraphic Horizon, and Age.— Walrus Ditch locality, Dorchester County, lower Waccamaw Formation, early Pleistocene, late Blancan. Discussion.—Erethizontids are members of the first major suite of Neotropical mammals that dispersed through Central America into North America during the Great American Biotic Inter- change upon development of the Panamanian Isthmus (GABI 1 of Woodburne, 2010), although the earliest immigrants were megalonychid and mylodontid sloths, which arrived during the early Hemphillian (Lindsay et al., 1984). The oldest occurrence of erethizontids in the USA is based on a specimen of “Coendou stirtoni” from the Wolf Ranch Local Fauna of Arizona (Harrison, 1978; later synonymized with E. bathygnathum by Fra- zier, 1981) dated magnetostratigraphically to about 2.6 Ma (just below the Gauss/Matuyama boundary; early late Blancan). Additional records of this spe- cies are found elsewhere across the American West in late middle to late Blancan sites (Frazier, 1981; Albright, 1999a). Prior to the recovery of the Walrus Ditch specimens, the oldest known records of porcupine in the southeastern USA were E. poyeri Hulbert, 1997, from the middle late Blancan Haile 7C local- ity (1.9–2.2 Ma) of Florida, and the much smaller E. kleini Frazier, 1981, from the slightly younger latest Blancan Inglis 1A locality (1.6–1.9 Ma) (Hulbert, 1997, 2010; Morgan, 2005, 2008). The extant North American porcupine E. dorsatum (Linnaeus, 1758) is first recorded from Florida’s earliest Irvingtonian Haile 16A Local Fauna after the demise of E. poyeri and E. kleini (Morgan and White, 1995; Hulbert, 1997; Morgan, 2005, 2008). This species is also recorded from the primarily Rancholabrean Edisto Beach fauna (SC83.17.1; Sanders, 2002). Prompted by the study of new erethizontid material from the El Golfo region of northern-most Mexico (~1.0 Ma), Sussman et al. (2016) reviewed the taxonomy of the North American species. They concluded that all Florida taxa older than Ran- cholabrean age (i.e., those noted above) represent Coendou, rather than Erethizon; thus, C. poyeri, C. kleini, and for the pre-Rancholabrean material previously referred to E. dorsatum, Coendou spp. They referred the El Golfo material to C. cf. C. kle- ini. The evidence for their conclusions focused on morphological changes that took place in the ere- thizontid lineage as the tropical taxon, Coendou, adapted to colder climatic conditions as it dispersed into (northern and western) North America, result- ing in Erethizon. Their hypothesis (supported with morphological data) postulated that these pre-Ran- cholabrean Florida taxa constituted a southeastern population of Coendou that, living in the subtropi- cal part of North America, were not subjected to the selection pressures of inclement climatic condi- tions that led to the evolution of Erethizon. The relatively large size of the Walrus Ditch M1 (7.64 mm AP by 7.98 mm TR) compared with those of E. kleini, E. poyeri, and E. dorsa- tum from Florida suggests referral to E. bathyg- nathum, although this species has never before been recorded beyond western North America (hence, the questionable specific assignment). The early late Blancan age (2.2–2.6 Ma; Hulbert, 2010) of the Walrus Ditch specimens (~2.3 Ma), however, which predates all other southeastern records of Erethizon, is consistent with that of E. bathygnathum at its western localities. The size of the unerupted m1 crown, at 6.39 mm AP by 6.96 mm TR, is similar in size to those of E. poyeri and also falls within the range of E. dorsatum (Hulbert, 1997:table 1; Sussman et al., 2016, also provided extensive tables of dental measurements for both extant and extinct North American species of por- cupines). This m1 differs from those of E. poyeri and E. dorsatum in being slightly wider trans- versely than it is long anteroposteriorly, but this may be a variable trait. As noted above, however, E. dorsatum is not known from the Southeast (i.e., 164 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) Florida) until the early Irvingtonian (Morgan and White, 1995; Hulbert, 1997). If the Walrus Ditch specimens are correctly identified, and if our estimation of the age of the Walrus Ditch LF is accurate (~2.3 Ma), then this occurrence not only provides the oldest record of an erethizontid in the southeastern USA, it also provides the first record of Erethizon (i.e., E. bathygnathum) beyond the western USA. This record, in turn, directly contradicts the findings of Sussman et al. (2016) that Erethizon was confined to western and northern regions of North America until the Rancholabrean. As an explanation for the late Blancan appearance of a porcupine (E. poyeri) in Florida, Hulbert (1997; also see Morgan and Emslie, 2010) noted the influx of xeric adapted western mammals during this time. Erethizon bathygnathum could well have been among that group, appearing earlier in South Carolina than in Florida due to high earlier Blancan sea levels which inundated peninsula Florida. Given this sce- nario, however, E. bathygnathum might then be expected to appear in Florida by the late Blancan, in turn hinting that the Walrus Ditch material might better be referred to a large individual of E. poyeri. Only more diagnostic material from the Walrus Ditch site will resolve these hypotheses, including whether all erethizontid material from the pre-Ran- cholabrean Southeast should be assigned to Coen- dou per Sussman et al. (2016). CAVIIDAE Fischer von Waldheim, 1817 HYDROCHOERINAE Gill, 1872 The relatively recent consensus reached by nearly all those who study fossil and extant capy- baras is that the taxonomy of, particularly, North American species is in a state of confusion and in much need of revision (e.g., Mones, 1984; Mones and Ojasti, 1986; Morgan and White, 1995; Mor- gan, 2008; Pérez et al., 2017; Vucetich et al., 2015). An attempt to resolve these issues is currently underway and being led primarily by the Argentine paleontologists C. M. Deschamps, M. E. Pérez, and M. G. Vucetich (e.g., Deschamps et al., 2007, 2013; Vucetich et al., 2015; Pérez et al., 2017, and references therein). The main reason for such con- fusion is that capybaras have ever-growing cheek teeth that change in size and morphology through- out life. The result, therefore, has been an inter- pretation of higher taxonomic diversity in the fos- sil record than actually exists. Fossils of different size and morphology at a single locality are now known to represent different ontogenetic stages of a single taxon, not multiple taxa (e.g., Deschamps et al., 2007; Vucetich et al., 2015). Adding to this problem is the referral of North American Pliocene species to genera that do not appear in the fossil record of South America until the Pleistocene. Spe- cifically, species referred to Neochoerus Hay, 1926, and Hydrochoerus Brisson, 1762, appear in North America (Arizona, Texas, Florida, South Caro- lina) before their ancestors are known from South America (see Vucetich et al., 2015). Also problem- atic is the questionable provenance of many speci- mens. For example, several of the specimens from South Carolina (referred to two different species) were recovered as a result of phosphate dredging and mining operations within and near the Ashley River in the earliest 1900s (Sanders, 2002). In an attempt to stem this tide of confusion, we herein adopt the more parsimonious hypoth- esis of North American capybara taxonomy as advocated by Vucetich et al. (2015). They referred all North American species recovered from Blan- can aged sites previously assigned to Neochoerus dichroplax (from Arizona and Florida) and Neo- choerus cordobai (from the Guanajuato region of central Mexico [Carranza-Castenada and Miller, 1988]) to the South American genus Phugatherium Ameghino, 1887; hence Phugatherium dichroplax. In addition to the lineage represented by Phugath- erium, Vucetich et al. (2015:331) recognized a sec- ond “Neochoerus-Hydrochoerus” lineage that they noted “was undoubtedly represented in the Pleisto- cene with N. aesopi, N. pinckneyi, and H. holmesi.” (Note: Sanders (2002) regarded N. aesopi a junior synonym of N. pinckneyi). They added, however, that “the Pliocene records of the two latter [spe- cies] must be revised both taxonomically and strati- graphically.” Although we make an attempt below to align our evaluation of SC capybaras with these recent findings, a revision such as that suggested by Vucetich et al. (2015) is beyond the scope of this report. ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 165 As noted by Morgan and White (1995) and Morgan (2008), Mones (1984) and Mones and Ojasti (1986) argued that Hydrochoerus, the genus to which extant capybaras belong, does not occur in North America; i.e., that specimens referred to this taxon (e.g., Hydrochoerus holmesi Simpson, 1928) have been done so in error and that they are instead representative of Neochoerus pinckneyi (Hay, 1923), considered by Ahearn and Lance (1980) to be the only species of Neochoerus in North Amer- ica (also see discussion in Sanders, 2002:96). Until further research provides new information on the relationships between Phugatherium, Neochoe- rus, and Hydrochoerus, we provisionally follow the conclusions of Mones (1984) and Mones and Ojasti (1986) that specimens from the USA previ- ously assigned to Hydrochoerus are more likely representatives of either of the other two taxa. This results in the reassignment of the nine specimens that Sanders (2002) referred to Hydrochoerus holmesi to Neochoerus pinckneyi and in the refer- ral of all the capybara material found in the Ash- ley River phosphate mining region to one species rather than two. Moreover, we maintain that all the capybara material currently known from SC, with the exception of limited material discussed below but including the numerous specimens currently curated at the SCSM, is referable only to N. pinck- neyi. PHUGATHERIUM Ameghino, 1887 PHUGATHERIUM DICHROPLAX (Ahearn and Lance, 1980) Figure 14L–N Referred Specimens.—USNM 181573, left M3; USNM 181640, right metacarpal III. Locality, Stratigraphic Horizon, and Age.— “Middle Swamp, approximately 100 yards south- east of SC Rte. 340 on Myrtle Branch, 8.5 mi. SW Darlington,” (approximate location 34º 11’ 36” N, 79º 55’ 57”), Darlington County; collected in 1971 by A. Langley of Darlington, SC, from cor- respondence archived at the USNM; see discussion below); Duplin Formation, early late Pliocene, middle Blancan. Description.—The small size of USNM 181573 (~ 32 mm AP by ~ 8 mm TR) implies that it represents a juvenile (or subadult) individual; it is not indicative of which species is represented. The tooth has an anterior prism (the first lamina, following terminology of Ahearn and Lance, 1980) followed by 13 posterior lamina prisms. None of the laminae show the distinct, prominently bifur- cated morphology on the labial edge that character- izes Phugatherium dichroplax, although this may be a variable character (see below). The metacar- pal, USNM 181640, was found at the same locality as the M3, but represents an adult, as the epiphy- ses are fused. It has a total length of 83.9 mm; the width of the proximal end = 16.9 mm; the width of the distal end = 18.9 mm. Discussion.—Although USNM 181573 and 181640 are labeled “Neochoerus cf. dichro- plax” in the collections at the USNM, the former was referred to Hydrochoerus holmesi by Ahearn (1981:59), and this identification was seconded by Sanders (2002:98). But as noted above, Mones (1984) and Mones and Ojasti (1986) do not con- sider Hydrochoerus a North American taxon. Nor are North American specimens referred to Hydro- choerus recorded from the Blancan (contra Mor- gan, 2008, who assigned the late Blancan Haile 1A material to that genus). Following these criteria, therefore, it may be more appropriate (and accu- rate) to assign USNM 181573 and USNM 181640 to either Phugatherium dichroplax or to Neochoe- rus pinckneyi. Although the morphology of USNM 181573 does not fit the description of P. dichro- plax in that it lacks the prominent bifurcations on the anterior few laminae that Vucetich et al. (2015) considered a synapomorphy of the genus, Morgan (2008) noted that the specimens of N. cordobai from central Mexico, which Vucetich et al. (2015) synonymized with Phugatherium dichroplax, also lack this character. If that is indeed the case, then it will be difficult to distinguish P. phugatherium from N. pinckneyi if only teeth are available. Dif- ferentiation of these two taxa is based primarily on characters of the skull (Vucetich et al., 2015). We provisionally refer this material to Phugath- erium dichroplax to maintain consistency with the currently associated USNM label, fully realizing, however, that this is certainly subject to change 166 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) upon future study of the Neochoerus-Hydrochoe- rus lineage. The oldest records of capybara in the USA from well-calibrated sites are those of “Neochoe- rus” dichroplax (i.e., Phugatherium dichroplax) from the early late Blancan 111 Ranch fauna, Arizona. These date to about 2.6–2.7 Ma, or just below the Gauss-Matuyama magnetochron bound- ary (Ahearn and Lance, 1980; Galusha et al., 1984; Bell et al., 2004; White and Morgan, 2005; Morgan and White, 2005; Morgan, 2005, 2008). The same species is also known from late Blancan faunas of Florida, such as Sommer’s Pit, Mule Pen Quarry, Macasphalt Shell Pit, Kissimmee River, and Ing- lis 1A (Morgan and White, 1995; Morgan, 2008; R. Hulbert, pers. comm. to LBA, May, 2019). The specimens noted above from Mexico which were originally assigned to Neochoerus cordobai by Carranza-Castenada and Miller (1988), but more recently synonymized with Phugatherium dichro- plax by Vucetich et al. (2015), were recovered from even older, early Blancan sites from the Rancho Viejo area of Guanajuato. Flynn et al. (2005:304) dated the strata yielding these specimens (including Glyptotherium) as “at least as old as 3.1–3.5 Ma ... or older ... correlative with levels slightly above a 3.9 Ma FT [fission track] date in La Pantera II ....” In summary, Phugatherium dichroplax is recorded in North America from the early middle Blancan (approx. 3.6 Ma) in Mexico, then again in the early late Blancan of Arizona (approx. 2.7 Ma), and finally in the early late to latest Blancan of Florida (approx. 1.8 Ma). Bearing on the age of USNM 181573 and 181640 is a series of correspondence archived at the USNM and dated from March, 1971, to August, 1972, between the collector of the material, a Mr. A. A. Langley of Darlington, SC, and Drs. T. R. Waller, who at the time was the Associate Cura- tor of Invertebrate Paleontology, and C. E. Ray, who was the Curator of Vertebrate Paleontology. Included in this correspondence is a list of mol- luscs collected by Mr. Langley together with the capybara material. The list includes Oliva caroli- nensis, Siphocypraecea carolinensis, Glycymeris subovata, “Ostrea” sculpturata, “Ostrea” dispari- lis, Mercenaria sp., Argopecten eboreus, and Nas- sarius vibex. Upon seeing this list, L. D. Campbell (pers. comm. to LBA, May, 2019) noted that sev- eral of these were long-ranged taxa, but that one in particular, Siphocypraecea carolinensis, was quite age diagnostic. Occurring only in the Duplin For- mation’s Natural Well, NC, stratotype5 and in South Carolina sites considered equivalent (and also in the slightly older Stokes quarry located 9.65 km directly NE from the capybara site and closer to the town of Darlington; see discussion of Duplin For- mation in section above on the fossil bearing beds of SC), S. carolinensis is confined to about 3.0–3.6 Ma, i.e., middle Blancan. Matrix extracted from between the enamel laminae of USNM 181573 was examined under a microscope by AES and found to be identical to that collected at the nearby Stokes quarry site in grain size, color, condition, and in the presence of tiny shell fragments. L. D. Campbell considers the Stokes quarry to be 3.4 to 3.6 myr old. This range of dates, 3.0–3.6 Ma, places a capybara (regardless of its identification) in the southeastern USA at about the same time, perhaps slightly more recently, than “Neochoerus” cordobai (= Phugath- erium dichroplax) is recorded in Mexico. The SC specimens would therefore represent some of the oldest records of capybara in the USA. Even older, however, may be several speci- mens from the Ashley River phosphate beds that Sanders (2002) referred to Neochoerus pinckneyi. This older age was based on his conclusion that these specimens, including the holotype (ChM PV2506), apparently originated from the Goose Creek Limestone as determined from the pres- ence of its distinctive matrix adhering to the teeth. The Goose Creek Limestone underlies the Duplin Formation (Figs. 2C, 3C) and was considered no younger than calcareous nannofossil zone NN15 by Weems et al. (1982). GTS2012 places NN15 in the late Zanclean (middle Pliocene) at about 3.8 Ma, and Campbell and Campbell (1995) suggested an age for the Goose Creek Limestone of between about 3.55 and 3.75 Ma on the basis of their mol- lusc studies. Both of these estimates fall within 5L. D. Campbell notes that there are about 360 species of molluscs from the Duplin Formation’s Natural Well, NC, stratotype curated in the Florida Museum of Natural His- tory’s Locklin Collection (L. D. Campbell, pers. comm. to AES, October 2007). ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 167 magnetochron C2Ar, and recent paleomagnetic analysis by LBA suggesting reversed polarity for the Goose Creek Limestone lends additional sup- port. These teeth therefore, appear to represent the oldest records of this Neotropical rodent in what is now the USA. If correctly identified, Neochoerus must have entered what is now the southern USA at a time correlative with, or even earlier, than the P. dichroplax occurrence in central Mexico. It appears, therefore, that capybaras not only reached North America much earlier than what has typically been considered the primary pulse of the GABI beginning about 2.7 myr ago, but that two lineages, Phugatherium and the Neochoerus- Hydrochoerus lineage, apparently reached the southeastern USA at about the same time as their arrival in the Guanajuato region of Mexico. As noted above in the discussion of Glyptotherium texanum, the apparent absence of capybara in the early to middle Blancan of Florida, in contrast to their early and middle Blancan records in SC, is likely attributable to the absence in Florida of ter- restrial faunas of this age due to high sea levels during that time. NEOCHOERUS Hay, 1926 NEOCHOERUS PINCKNEYI (Hay, 1923) Figure 14M Referred Specimens.—Additions to ChM PV2796: right I1, left M2; SC2017.10.24, left lower incisor; SC2015.17.7, sacrum; SC2015.17.8, distal right tibia; SC2015.17.9, metacarpal III. Locality, Stratigraphic Horizon, and Age.— ChM PV2796, SC2017.10.24, SC2015.17.7, and SC2015.17.8 from west branch of Cooper River, Berkeley County, thought to be derived from Wando Formation (see Sanders, 2002:102–104); SC2017.10.24 from Broad River, seaward of Edward B. Rogers Bridge (SC Highway 170), Beaufort County, likely derived from a unit equiv- alent to Wando Formation; late Pleistocene, Ran- cholabrean. Discussion.—Sanders (2002:102) noted cra- nial elements and teeth collected by LBA in 1977 from the west branch of the Cooper River cata- logued as ChM PV2796. In the Spring of 2015, Ms. Susan Wallace, who was LBA’s scuba diving partner throughout the 1970s (and the collector of the beautiful Tapirus skull described in Ray and Sanders, 1984), came across a box of fossil speci- mens while preparing for a move from her Isle of Palms, SC, residence. Contacting LBA to look over the long-packed specimens, two capybara teeth were noted that identically matched those of ChM PV2796 in color, degree of wear, and in the matrix remaining in the reentrants of the teeth. In conversation it became apparent that the teeth “re- discovered” by Ms. Wallace were almost assuredly collected on the same dive, at the same location, as those specimens collected by LBA, resulting in the consequent conclusion that they likely belonged to the same individual. Upon comparison by LBA, this hunch was confirmed, and Ms. Wallace agreed to donate the newly found specimens to The Charleston Museum where they are now included with the material earlier assigned to ChM PV2796. Confirmation was supported by the perfect fit of the “re-discovered” right incisor into the empty right alveolus seen in figure 42B of Sanders (2002). The sacrum, distal tibia, and metacarpal III were also among the specimens in the box of Ms. Wallace’s fossils. Although they too were collected in the Cooper River, they were not recovered from the identical location as ChM PV2796, nor does their preservation match that of the same individual. EULIPOTYPHLA Haeckel, 1866 TALPIDAE Fischer von Waldheim, 1817 CONDYLURA Illinger, 1811 CONDYLURA CRISTATA (Linnaeus, 1758) Referred Specimen.—ChM PV9636, right humerus. Locality, Stratigraphic Horizon, and Age.— Bank of Eagle Creek near junction with Ashley River, Dorchester County (Fig. 10), in proximity to outcrops of Chandler Bridge Formation, but likely derived from overlying Wando Formation; late Pleistocene, Rancholabrean. Discussion.—This specimen is an addition to the known fossil talpid material from South Caro- lina. It was originally thought to have originated from the upper Oligocene Chandler Bridge Forma- tion due to the occurrence of the latter unit in the vicinity of the site where the specimen was col- 168 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) lected, but identification of the element as belong- ing to the extant star-nosed mole, Condylura cris- tata, resulted in the realization that the specimen had to have originated in much younger sediments overlying the Chandler Bridge. In South Carolina Condylura cristata is also recorded from the Ran- cholabrean aged Crowfield and Ardis local faunas (Bentley et al., 1994; Chandler and Knight, 2009). North American fossil talpids that span the interval of time represented by the Chandler Bridge Formation (late Arikareean) include Mystipterus Hall, 1930, Proscalops Matthew, 1901, Mesosca- lops Reed, 1960, Quadrodens Macdonald, 1970, and Scalopoides Wilson, 1960 (Gunnell et al., 2008). The humeri of Mystipterus, Proscalops, Mesoscalops, and Quadrodens are entirely unlike that of ChM PV9636. Mystipterus has a narrow, non-expanded humerus with a rounded head, indic- ative of an ambulatory adaptive mode (Hutchison, 1976), whereas Proscalops and Mesoscalops have the derived, greatly broadened humeri characteris- tic of a fully fossorial adaptive niche (Hutchison, 1972; Barnosky, 1981, 1982). Quadrodens is a poorly known form from the early Arikareean aged Sharps Formation of South Dakota, the humerus of which Gunnell et al. (2008:102) described as having a “long, medially directed distal pectoral process.” Although ChM PV9636 resembles the humerus of Scalopoides more so than that of any other Oligocene talpid, comparison with humeri of Condylura cristata leaves little doubt regarding its taxonomic assignment. In contrast to Scalopoides, ChM PV9636 has the humeral head angled later- ally to the long axis vs. parallel in Scalopoides, a clavicular articular facet that is nearly parallel to the long axis, vs. strongly angled in Scalopoides, and a teres tubercle that is angled medially relative to the long axis, vs. parallel in Scalopoides. CARNIVORA Bowdich, 1821 VIVERRAVIDAE Wortman and Matthew, 1899 DIDYMICTIS Cope, 1875 DIDYMICTIS PROTEUS Simpson, 1937 Figure 15 Referred Specimen.—ChM PV7687, trigo- nid of right m1. Locality, Stratigraphic Horizon, and Age.— Collected by B. Palmer from a spoil pile of sedi- ment obtained from stratigraphically below the San- tee Limestone at the Martin Marietta Aggregates Jamestown Quarry, Berkeley County, “Jamestown beds” (Chicora Member of the Williamsburg For- mation); late Paleocene, calcareous nannoplankton zone NP9a, late Thanetian, Clarkforkian. Description and Discussion.—Polly (1997) concluded that there were seven species of viver- ravid carnivorans from the Bighorn and Clarks Fork basins of Wyoming, represented by two gen- era, Viverravus and Didymictis. ChM PV7687, which consists only of the trigonid, is referred to Didymictis on the basis of its much more closely appressed paraconid and metaconid relative to Viverravus, in which the more anterior position of the paraconid gives the trigonid a more “open” morphology (Polly, 1997). Of the three recognized species of Didymictis, ChM PV7687 most closely resembles that of Didymictis proteus, a late Tif- fanian (Ti5) through earliest Wasatchian (Wa0) species, although the SC tooth is slightly smaller than specimens from Wyoming. Didymictis lepto- mylus Cope, 1880, and D. protenus (Cope, 1874) are larger and occur later in the Wasatchian (Polly, 1997). Measurements of this trigonid are 3.6 mm AP by 4.3 mm TR. Prior to the discovery of latest Paleocene/ earliest Eocene fossil-bearing sediments at the Jamestown quarry (calcareous nannoplankton zone NP9a; Cicimurri et al., 2016), only three sites along the Eastern Coastal Plain had yielded terrestrial fos- sils of this age: (1) land mammals from the Chicora Member of the Williamsburg Formation exposed in the St. Stephen pit, Berkeley County, referred to as the Black Mingo Fauna, are considered to be late Tiffanian (Ti5) in age (Schoch, 1985, 1998; Lof- gren et al., 2004); (2) from the Fisher/Sullivan site near Fredericksburg, Virginia, Rose (1999, 2010) identified eight mammalian taxa of Wasatchian age from the Potapaco Member of the Nanjemoy For- mation (Weems and Grimsley, 1999); and (3) from the Aquia Formation in Maryland, Rose (2000) described three Clarkforkian aged mammals. None of these faunas include taxa representative of the Viverravidae, thus making ChM PV7687 the first ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 169 Figure 15. CT scanned images of Didymictis proteus, ChM PV7687, partial right m1 (trigonid only), from “Jamestown beds.” Left column: labial, anterior, and lingual views (top to bottom). Right column: posterior and occlusal views. 170 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) known occurrence of a viverravid from the East- ern Coastal Plain of the USA. The nearest known occurrence of viverravids to SC are those described by Beard and Dawson (2009) from the early Wasat- chian Red Hot Local Fauna from the uppermost Tuscahoma Formation of Mississippi along the Gulf Coastal Plain. FELIDAE Gray, 1821 LEOPARDUS Gray, 1842 LEOPARDUS AMNICOLA (Gillette, 1976) Figure 16A–B Felis amnicola Gillette, 1976. Leopardus wiedii amnicola Werdelin, 1985. Felis amnicola Gillette. Spearing, 2006. Leopardus amnicola (Gillette). Hulbert and Pratt, 1998. Referred Specimen.—ChM PV2636, partial left dentary with p3–4. Locality, Stratigraphic Horizon, and Age.— “Ashley River Phosphate beds,” Magnolia Phos- phate Mine, Runnymede Plantation, Charleston County, Wando Formation, late Pleistocene, Ran- cholabrean. Discussion.—ChM PV2636 represents the first record of Leopardus amnicola from SC. In size and morphology it appears closest to specimens found primarily in the Aucilla River of the Florida panhandle originally described by Gillette (1976) as Felis amnicola. Considered a subspecies of the margay, L. wiedii (L. wiedii amnicola), by Werde- lin (1985), this taxon was maintained as a sepa- rate species, F. amnicola, by Spearing (2006). The Charleston specimen plots out at what would be the largest specimen of the species (p3: 7.27 mm AP x 3.47 max TR; p4: 9.05 AP x 4.01 max TR; depth of dentary below p3: 14.5 mm), but not so large as to warrant referral to a new species. According to Werdelin (1985:195), L. amnicola differs from L. yagouaroundi in having “the principal cusps of the cheek teeth aligned nearly in parallel” and a p4 that is transversely narrower for its length. In L. yagouaroundi, Leopardus sp. from coastal Georgia (Hulbert and Pratt, 1998), and even more so in the larger Lynx rufus, the p4 is relatively shorter and broader. MIRACINONYX Adams, 1979 MIRACINONYX ?TRUMANI (Orr, 1969) Figure 16C Referred Specimen.—USNM 533994, par- tial right tibia collected by E. A. Crawford. Locality, Stratigraphic Horizon, and Age.— Edisto Beach, Colleton County, from undetermined offshore upper Pleistocene unit; late Pleistocene, Rancholabrean. Discussion.—Identification of this partial tibia as belonging to the North American cheetah- like cat Miracinonyx was based on comparisons with material referred to M. inexpectatus (Cope, 1895) from Hamilton Cave, West Virginia, by F. Grady (USNM). That species is known from late Blancan and Irvingtonian localities, including sites in SC (Sanders, 2002), with a last known occur- rence of about 0.6 Ma (Van Valkenburgh et al., 1990). More recently, material referred to this taxon has been recovered from the approximately 0.41 Ma Camelot Local Fauna, thus extending its range into the latter part of the Irvingtonian. Though not yet studied in detail, the material from Camelot includes a radius, the distal part of a humerus, the proximal part of a femur, and two phalanges (SC2003.75.148–151, 153). A partial mandible with m1 (SC2003.75.147) is provision- ally assigned to M. inexpectatus, but detailed com- parisons with other cats have not yet confirmed this Figure 16. Leopardus amnicola, ChM PV2636, partial left dentary with p3–4, in A, labial and B, lingual views, from “Ashley River phosphate beds”; C, Miracinonyx ?trumani, ChM PV553994, partial right tibia, in (left to right) anterior, lateral, and medial views, from Edisto Beach; D, Borophagus hilli, cast of AMNH 144657, left p4, in (left to right, top) labial and lingual view, and (left to right, bottom) anterior and posterior views, from Martin Marietta Orangeburg Quarry, Raysor Formation; E, Canis lepophagus, SC2006.1.15, right P4, in (left to right) labial, lingual, and occlusal views, from Walrus Ditch site, Waccamaw Formation; F, lingual and G, labial views of Canis armbrusteri, SC2004.1.2, right dentary with c, p2–m2, from Camelot locality, Ladson Formation. ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 171 172 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) identification. A second species, M. trumani, is apparently confined to the Rancholabrean, with an earliest known occurrence in Crypt Cave, Nevada, which has been dated to 19,750 ± 650 years BP (Van Valkenburgh et al., 1990). Miracinonyx trumani was originally recorded from Natural Trap Cave in the Bighorn Mountains of Wyoming, the oldest stratum in which was radiocarbon dated to 12,770 ± 900 years (Martin et al., 1977). From a biochrono- logic perspective, therefore, USNM 533994 seems referable to M. trumani rather than M. inexpecta- tus, as the fauna from Edisto Beach is primarily Rancholabrean. On the other hand, F. Grady (pers. comm. to LBA, May, 2019) noted that the size of the Edisto tibia was similar to M. inexpectatus. Further study of this specimen is warranted before a definitive identification to species can be made. CANIDAE Fischer von Waldheim, 1817 BOROPHAGINAE Simpson, 1945 BOROPHAGUS Cope, 1892 BOROPHAGUS HILLI (Johnston, 1939) Figure 16D See Wang et al. (1999:296) for synonymy. Referred Specimen.—AMNH 144657, left p4. Locality, Stratigraphic Horizon, and Age.— Martin Marietta Aggregates Cross Quarry, Orange- burg County, Raysor Formation, early Pliocene, early Blancan. Description.—Because detailed description of this tooth was recently provided by Tseng and Geisler (2016), only a brief summary is provided here. Referral of AMNH 144657 to Borophagus hilli, rather than to the other Blancan borophagine, B. diversidens Cope, 1892, is based on the pres- ence of a posterior accessory cusplet (absent in B. diversidens) and on the absence of the strongly posteriorly-sloped crown noted for the latter (Wang et al., 1999; Tseng and Geisler, 2016). Borophagus dudleyi (White, 1941), from the latest Hemphillian Palmetto Fauna of Florida (Bone Valley Forma- tion), is known only from an edentulous skull, therefore precluding comparison with the SC tooth, although the latter matches very closely those p4s from the same fauna but referred to B. hilli (see discussion in Wang et al., 1999:301). Borophagus dudleyi was also reported from the late Hemp- hillian Lee Creek Local Fauna of North Carolina (Eshelman and Whitmore, 2008), but the absence of a p4 again precludes comparison with the SC specimen. Measurements provided by Tseng and Geisler (2016) place the tooth at 15.4 mm long by 13.1 mm wide, slightly shorter but comparable in width to the smallest specimen of B. hilli noted in Appendix III of Wang et al. (1999). Discussion.—In May 2006, Ms. D. Young, searching for fossils with Mr. B. Palmer, found an isolated carnivoran p4 in spoil material on the floor of the Martin Marietta Aggregates Orange- burg quarry west of Cross, SC. In the part of the quarry from which the tooth was found, when it was found, excavation had yet to cut into the Eocene units which are the source of the limestone being mined. Unconformably overlying the Eocene units is the Pliocene age Raysor Formation, and it is from this unit the tooth is considered to have origi- nated, although Tseng and Geisler (2016) suggested the Goose Creek Limestone as a possible source. Examination by LBA of the area of the quarry from which the tooth was recovered revealed exposures of the Raysor Formation, but not of the Goose Creek Limestone. AMNH 144657 represents the only known occurrence of a borophagine canid from SC. CANINAE Fischer de Waldheim, 1817 CANIS Linnaeus, 1758 CANIS LEPOPHAGUS Johnston, 1938 Figure 16E Canis lepophagus Johnston, 1938. Canis latrans lepophagus Giles, 1960. Canis lepophagus Johnston. Tedford et al., 2009. Referred Specimen.—SC 2006.1.15, right P4. Locality, Stratigraphic Horizon, and Age.— Walrus Ditch locality, Dorchester County, lower Waccamaw Formation, early Pleistocene, late Blancan. Description and Discussion.—This tooth represents the first reported occurrence of C. lep- ophagus in SC. It measures 22.6 mm AP by 11.7 mm TR at the paracone by 10.5 mm TR at the pro- tocone. In Florida, C. lepophagus is known from ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 173 the late Blancan Santa Fe River and Withlacoochee River 1A faunas. CANIS ARMBRUSTERI Gidley, 1913 Figure 16F–G See Tedford et al. (2009) for synonymy. Referred Specimens.—SC 2004.1.1, right dentary with broken canine, p2, p4–m2; SC 2004.1.2, right dentary with canine, p2–m2; SC2003.75.152, left Mt II; SC2003.75.692, left Mt V. Locality, Stratigraphic Horizon, and Age.— Camelot locality, near Harleyville, Dorchester County, from fluvial channel sediments likely cor- relative with the Ladson Formation; middle Pleis- tocene, late Irvingtonian. Discussion.—Known from the early Irving- tonian of Arizona and Florida to the early Rancho- labrean of Florida, Canis armbrusteri is broadly distributed across the USA (Tedford et al., 2009). Taxonomically the Camelot Local Fauna appears similar to the late Irvingtonian Coleman 2A Local Fauna of Florida (Martin, 1974), and this is sup- ported by the 400–450 kyr dates on the Ladson Formation from which the Camelot LF is thought to be derived. The two above noted rami were uti- lized by Kohn et al. (2005) for isotopic analysis of enamel in an attempt to infer trophic levels, habitat, etc. Additional undescribed material of this taxon is currently catalogued and under study by DJC, JLK, and associates at the SCSM. CANIS DIRUS Leidy, 1858 Figure 17A–D See Tedford et al. (2009) for synonymy. Referred Specimens.—ChM PV2637, partial right mandibular dentary with p4; SC 83.118.1, left dentary with p2, partial p3, p4–m2; ChM PV7697, right medial phalanx, pes digit 3. Locality, Stratigraphic Horizon, and Age.— ChM PV2637 from “Ashley River phosphate beds” (Wando Formation) near Runnymede Plan- tation, about 15 km northwest of Charleston; SC 83.118.1 from Cooper River, Berkeley County, likely derived from Wando Formation; ChM PV7697 from “near Summerville,” exact locality unknown, likely derived from Wando Formation; late Pleistocene, Rancholabrean. Discussion.—The partial right dentary, ChM PV2637, was originally mentioned and figured by Hay (1923:365-366) who noted its similarity to “C. occidentalis,” but he did not assign it to a specific taxon. Comparisons with material at the SCSM and the FLMNH, and with measurements in Nowak (1979, 2002) and Kurtén (1984), result in a confident referral of the specimen to the dire wolf. The p4 measures about 18.5 mm AP by 9.5 mm TR (max). Measurements for SC 83.118.1 are as follows: p2, 14.6 mm; p3, 16.9 mm; p4, 19.9 mm; m1, 34.6 mm; m2, 14.2 mm; p2-m2, 102.9 mm. See the above discussion on Castoroides regarding the unit of origin for this specimen. ChM PV7697, the medial phalanx, measures about 2.8 cm long. CANIS LATRANS Say, 1823 Figure 17E–F Referred Specimens.—SC2016.20.1, right dentary with p1–4, m1–2; SC2016.21.1, edentu- lous dentary showing alveoli for p3–m3; USNM 533995, right m1. Locality, Stratigraphic Horizon, and Age.— SC2016.20.1 collected by J. Thompson from Edisto River, south of Highway 17, Colleton County, ?Wando Formation, Rancholabrean; SC2016.21.1 collected by B. Orr from the Cooper River, Berke- ley County, ?Wando Formation, Rancholabrean; USNM 533995 from Myrtle Beach, Horry County, undetermined offshore upper Pleistocene unit, Rancholabrean. Discussion.—These specimens represent the first reported fossil occurrences of C. latrans in SC. Measurements of the teeth are provided in Table 7, and they fall within the range of variation for this species as presented in figure 4 of Nowak (2002:106). SC2016.21.1, the edentulous dentary, shows no diastemata whatsoever between any of the alveoli. SC2016.20.1, on the other hand, shows prominent diastemata between all the teeth. Exami- nation of several coyote jaws in the collections of the SCSM shows this feature to be variable. The m1s have a robust hypoconulid, diminutive ento- conid, and weak posterior cingulum. As Nowak (2002:118) noted, fossil material of eastern coyotes is rare, although late Rancho- 174 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 175 offshore Pleistocene unit. Discussion.—In his discussion of A. pristi- nus, Sanders (2002:40) noted the provenance of two specimens, ChM PV5472 and ChM PV2536, as the Wando Formation, and also concluded that Leidy’s lost holotype was derived from that unit, as well. He also cited Kurten’s and Anderson’s (1980) range for this taxon as “only from a few Irving- tonian and Rancholabrean localities in the eastern United States.” More recent studies of A. pristinus, however, from a variety of localities across the eastern USA, especially Florida and Pennsylva- nia where several much more complete specimens have been recovered, have determined that there are no confirmed Rancholabrean occurrences, and that this taxon is diagnostic of the late Blancan to late Irvingtonian (Emslie, 1995; Schubert, 2008; Schubert et al., 2010). The Rancholabrean spe- cies, A. simus, is much larger than A. pristinus and apparently extremely rare in southeastern North America, currently known only from two locali- ties in Florida (Schubert et al., 2010). These find- ings, therefore, prompt a reevaluation of the age and provenance of specimens discussed by Sanders (2002) as having been recovered from the Wando Formation. First is Leidy’s (1854) holotype. That now lost specimen consisted of a single isolated m2 that was recovered from the Ashley phosphate beds at Bee’s Ferry on the west bank of the Ashley River about 17 km northwest of Charleston, and Sanders (2002) convincingly argued that it was collected labrean records have been reported from Indiana, Pennsylvania, West Virginia, Mississippi, Ala- bama, and Florida. This rarity of coyotes in the east during the late Rancholabrean, Nowak concluded, is most likely due to the influx of the small wolf, C. rufus, at that time. URSIDAE Gray, 1825 ARCTODUS Leidy, 1854 ARCTODUS PRISTINUS Leidy, 1854 Referred Specimen.—USNM PAL 530189, left m2. Locality, Stratigraphic Horizon, and Age.— Myrtle Beach, Horry County, from undetermined Figure 17. Canis dirus, ChM PV2637, partial right mandibular dentary with p4, in A, labial, and B, lingual view, from the “Ashley River phosphate beds”; Canis dirus, SC 83.118.1, left dentary with p2, partial p3, p4–m2, in C, lingual, and D, labial view, from the Cooper River; E, Canis latrans, SC2016.20.1, lingual view of right dentary with c–m2 from Edisto River, ?Wando Formation; F, USNM 533995, stereo view of right m1, from Myrtle Beach; G, Tremarctos floridanus, ChM PV3463, right M2, from Edisto Beach; H, Ursus americanus, SC83.99.3, left M2, from Horry County; I, Ontocetus emmonsi, ChM PV6949, right astragalus, from Cross quarry, Goose Creek Limestone or Raysor Formation; J, Phocanella pumila, SC2015.18.4, right femur, from Ashepoo River, Colleton County; K, Phocanella pumila, SC98.60.6 (cast of USNM 181649), right femur, from Yorktown Formation, Lee Creek Mine, NC; L, Phocanella pumila, SC2009.3.1, left femur, from Morgan River, Beaufort County, ?Wabasso beds or ?Goose Creek Limestone; M, Callophoca obscura, SC2012.16.1, left humerus, from Broad River, Beaufort County, ?Wabasso beds or ?Goose Creek Limestone; N, Monatherium sp., ChM PV7688, distal part of humerus, in dorsal (top) and ventral (bottom) view, from Summerville area, ?Ebenezer Formation. Table 7. Measurements (mm) of lower teeth in selected specimens of Canis latrans. Specimen/Tooth AP TR SC2016.20.1 canine 7.0 6.4 p1 4.7 3.2 p2 8.0 4.4 p3 10.3 5.0 p4 11.2 6.2 m1 22.2 9.0 m2 9.3 7.0 USNM 533995 m1 19.5 8.5 176 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) from the Wando Formation. To the extent, how- ever, that the Wando Formation is correlative with the Rancholabrean NALMA within which A. pris- tinus apparently does not occur, the Leidy speci- men more likely originated from the Penholoway or Ladson formations from which middle and late Irvingtonian taxa, respectively, are known, fol- lowed by reworking into the Wando at a later time. Determining that the tooth was found in the Wando Formation, Sanders (2002:42) suggested that it per- haps had been “transported [to the Wando Forma- tion in the vicinity of Bee’s Ferry] by water from an inland source.” That is likely correct, the “inland source,” however, being the older, more age-appro- priate formations noted above. Further supporting this is a statement by Sanders (2002:61) regarding the provenance of another specimen collected from the Ashely phosphate beds in the same region. Referring to MCZ 16512, an edentulous dentary of Miracinonyx inexpectatus, Sanders stated, “since most of the land mining operations in the Charles- ton area were located along the Ashley River this specimen probably came from one of the mines on the north side of the river, where the early Pleis- tocene (Middle Irvingtonian) Penholoway Forma- tion underlies the Wando Formation ….” The same might be said about Leidy’s holotype. ChM PV2536, a lower left canine, was found in the same area as Leidy’s holotype (the Ashley phosphate beds on the west bank of the Ashley River), and like the latter tooth was also likely reworked into the Wando. Supporting these conclusions is the presence of A. pristinus in the late Blancan Walrus Ditch LF (lower Waccamaw Formation) and another m2 from the middle Pleis- tocene aged Ladson Formation (ChM PV5146: Sanders, 2002), units whose age is consistent with that of this species. Another specimen noted from the Wando Formation is ChM PV5472, a partial dentary with m1-m2. This specimen was actually found on the bottom of Tail Race Canal (see Sanders, 2002:40 and fig. 16), which refers to the upper-most west branch of the Cooper River, not in situ. Pleistocene fossils from the Cooper River are always found as isolated specimens in lag deposits that have accu- mulated on the bottom of the channel – never in situ (LBA, pers. observ.). Thus, based on the bio- chronological findings of Schubert et al. (2010) noted above, it is more likely that this specimen originated from the Waccamaw, Penholoway, or Ladson formations from which late Blancan to late Irvingtonian taxa, respectively, are known, rather than from the Wando Formation as reported. USNM PAL 530189, from deposits off of Myrtle Beach, is noted here simply as an addi- tion to the known Arctodus material from SC. This specimen, too, was likely reworked into the off- shore upper Pleistocene deposits from a subjacent, older unit. TREMARCTOS Gervais, 1855 TREMARCTOS FLORIDANUS (Gidley, 1928) Figure 17G Arctodus floridanus Gidley, 1928. Tremarctos mexicanus Stock, 1950. Tremarctos floridanus (Gidley). Stock, 1950; Kurtén, 1966. Referred Specimen.—SC81.140.1, right M2. Locality, Stratigraphic Horizon, and Age.— Edisto Beach, Colleton County, undetermined off- shore upper Pleistocene unit; late Pleistocene, Ran- cholabrean. Discussion.—The photograph of the M2 of Tremarctos floridanus in Figure 16G is of ChM PV3463 (also from Edisto Beach) rather than SC81.140.1. The two specimens are nearly identi- cal, however, and we note the latter as an addition to material previously referred to this taxon from SC, such as specimens from the Ardis LF (Bentley et al., 1994; Sanders, 2002). SC81.140.1 measures 29.9 mm AP x 15.0 mm TR; ChM PV3463 mea- sures 29.83 mm AP x 15.7 mm TR. URSUS Linnaeus, 1758 URSUS AMERICANUS Pallas, 1780 Figure 17H Referred Specimen.—SC83.99.3, left M2. Locality, Stratigraphic Horizon, and Age.— SC83.99.3 from Horry County, undetermined off- shore upper Pleistocene unit; late Pleistocene, Ran- cholabrean. Discussion.—Measuring 27.02 mm AP by 16.16 mm TR, SC83.99.3 is considerably smaller than the late Blancan to late Irvingtonian lesser short-faced bear, Arctodus pristinus, and the even ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 177 larger late Rancholabrean greater short-faced bear, A. simus. SC83.99.3 also lacks the dramatic reduc- tion of the posterior portion of the tooth that is char- acteristic of Arctodus. Although the late Pleisto- cene Ursus americanus did not overlap temporally with A. pristinus, it did co-occur with Tremarctos floridanus (Schubert et al., 2010). According to B. Schubert (pers. comm. to LBA, November, 2016), the M2 of U. americanus has a cingulum and the lingual side of the occlusal surface is concave, whereas Tremarctos tends to lack a cingulum ridge and the lingual side of the tooth forms a straight line. We note this specimen simply as an addition to the known Ursus material from SC. PROCYONIDAE Gray, 1825 PROCYON Storr, 1780 PROCYON LOTOR (Linnaeus, 1758) Referred Specimen.—ChM PV7696, axis vertebra. Locality, Stratigraphic Horizon, and Age.— Near Summerville (exact locality unknown), Berkeley County, ?Wando Formation; late Pleisto- cene, late Rancholabrean. Discussion.—Identification of this well pre- served specimen is based on the nearly identical morphology it shares with the axis vertebra of the extant raccoon. Fossil remains of raccoon are also known from Edisto Beach (Roth and Laerm, 1980), the Ardis Local Fauna (Bentley et al., 1994), the Crowfield Local Fauna, plus the older, late Irving- tonian aged Camelot Local Fauna. ODOBENIDAE Allen, 1880 ONTOCETUS Leidy, 1859 ONTOCETUS EMMONSI Leidy, 1859 Figure 17I See Kohno and Ray, 2008, for synonymy. Referred Specimens (in part from Kohno and Ray, 2008:77; also see Boessenecker et al., 2018).—USNM 437544, 437545, 475484, upper canines, from Little River, Horry County; USNM 475463, upper canine, from Martin Marietta Cross quarry, Berkeley County; USNM 521227, upper canine, Walrus Ditch locality, Dorchester County; SC2017.14.2 (original number SC98.51.1), right tusk, Austin Sand Pit, Ridgeville, Dorchester County; CCNHM 1144, left tusk, Austin Sand Pit, Ridgeville, Dorchester County; SC76.15.9, 77.14.1, 77.14.2, 77.14.3, 79.38.199, and 84.37.1, upper canines, from Cooper River, Berkeley County; AMNH 104788, 104790, upper canines, from “Ashley River phosphate beds” (Sanders, 2002:66); ChM PV1028, 1029, upper canines, from “Charleston”; ChM PV13296, upper canine, from Coosaw River, Beaufort County; SC2017.10.6, proximal left tusk, from Broad River, Beaufort County; SC2016.1.23, distal right tusk, from Broad River, Beaufort County; ChM PV13497, upper canine, from “?Charleston”; USNM 481879, upper canine, from Beaufort County; USNM 475482 left dentary with i3, c, p1–3 (cast ChM PV7213), from Martin Marietta Cross quarry, Berkeley County; ChM PV6949, right astragalus, from Martin Mari- etta Cross quarry, Berkeley County. Stratigraphic Horizon and Age.—Speci- mens from Cross quarry from Raysor Formation or Goose Creek Limestone, late early Pliocene, late Zanclean, early Blancan; specimens from Cooper River probably from Goose Creek Limestone, late early Pliocene, late Zanclean, early Blancan; speci- mens from Ashley River phosphate beds probably from Goose Creek Limestone, late early Pliocene, late Zanclean, early Blancan; specimen from Wal- rus Ditch locality from lower Waccamaw Forma- tion, early Pleistocene, late Blancan; specimen from Ridgeville LF (= Austin pit locality) from ?upper Waccamaw Formation, early Pleistocene, latest Blancan-earliest Irvingtonian. Discussion.—Kohno and Ray (2008) reviewed all Pliocene walruses from North Amer- ica, Europe, and northern Africa, which constituted several named genera and species, and concluded that they all belonged to a single taxon, Ontocetus emmonsi (also see Deméré, 1994, and Deméré et al., 2003). Found in Pliocene coastal deposits of the eastern and western North Atlantic (and the early Pliocene of the North Pacific), O. emmonsi is best represented in North America from the Yorktown Formation at the Lee Creek Mine, NC. Several specimens have also been recovered from equivalent strata in South Carolina, primarily the Goose Creek Limestone and/or Raysor Forma- 178 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) tion (listed above from Kohno and Ray, 2008:77). Although Kohno and Ray (2008) noted the ques- tionable occurrence of USNM specimens 437544, 437545, 475484 (upper canines) from the Cane- patch (= Ladson) Formation of Horry County, it is highly unlikely that they are from that unit given its middle Pleistocene age; O. emmonsi is not known to have survived along the Eastern Coastal Plain of North America later than the early late Pliocene (Kohno and Ray, 2008:63; but note the following). Similarly, they noted the questionable occurrence of the Walrus Ditch specimen (USNM 521227) from the Goose Creek Limestone. This, too, is untenable, as the Walrus Ditch LF is not derived from the Goose Creek Limestone; it includes a rich late Blancan terrestrial mammal fauna from what is considered to be the lower Waccamaw Forma- tion (approx. 2.4 Ma). The Walrus Ditch specimen, in addition to the specimen noted in Boessenecker et al. (2018) from the Ridgeville LF, therefore, extends the range of O. emmonsi into what is now considered the early Pleistocene based on the recent extension of the Pleistocene downward to the base of the Gelasian Stage (Gibbard et al., 2010; Pillans and Gibbard, 2012). Kohno and Ray (2008:77) also referred two upper canines from the Ashley River phosphate beds near Charleston to O. emmonsi (AMNH 104788 and 104790), whereas Sanders (2002) assigned these specimens to Odobenus rosmarus. The morphology of the specimens indicates refer- ral to the former, i.e., laterally compressed with fluting along the length of the tusk. Based on O. emmonsi’s termination by (what is now considered) the early Pleistocene, and on the non-overlapping ranges of O. emmonsi and Odobenus rosmarus, these specimens evidently were reworked into the Ashley phosphate beds (within the Wando Forma- tion) from an older unit, presumably the Goose Creek Limestone upon which the Wando uncon- formably rests in many places. PHOCIDAE Gray, 1825 PHOCINAE Gill, 1866 PHOCANELLA Van Beneden, 1877 PHOCANELLA PUMILA Van Beneden, 1877 Figure 17J–L See Koretsky and Ray (2008) for synonymy. Referred Specimens.—SC2016.1.16, partial left innominate; SC2016.1.17, partial right innomi- nate; SC2015.18.4, right femur; SC2009.3.1, left femur; SC2017.14.1, proximal left femur. (Note: SC2016.1.16, SC2016.1.17, and SC2015.18.4 may be associated). Locality, Stratigraphic Horizon, and Age.— SC2016.1.16, SC2016.1.17, and SC2015.18.4 col- lected by M. Swilp from the Ashepoo River, Colle- ton County (exact locality unknown); SC2017.14.1 collected by M. Swilp from Whale Branch region of Broad River, Beaufort County; SC2009.3.1 col- lected by scuba divers from the Morgan River, Beaufort County; questionably from Wabasso beds, although matrix in pores suggests possible derivation from Goose Creek Limestone; early Pliocene, late Zanclean, early Blancan (see further discussion below). Discussion.—Originally described by Van Beneden (1877) from the Zanclean of Belgium, Phocanella pumila is known from the USA Atlan- tic Coastal Plain from the Yorktown Formation, NC (Ray, 1976a; Koretsky and Barnes, 2008; Koretsky and Ray, 2008), and from the latest Hemphillian Palmetto Local Fauna of the upper Bone Valley Formation, central Florida (Morgan, 1994). To our knowledge, this is the first report of this taxon from SC. Referral of the SC femora to P. pumila is based on the nearly identical morphology they share with the Lee Creek specimen (USNM 181649) described by Koretsky and Ray (2008) as determined from direct comparisons with a cast of the same (SC98.60.6; Fig. 17K). The larger size of SC2009.3.1 may be a function of ontogenetic stage or sexual dimorphism (Table 8). USNM 181649 was stated to be from the “lower beds of the Yorktown Formation” in the Lee Creek Mine (Koretsky and Ray, 2008:121), which indicates its likely derivation from the Sunken Meadow Member. Terrestrial mammals from this unit, the Lee Creek Local Fauna (Eshelman and Whitmore, 2008), are indicative of a latest Hemp- hillian age – the same age as the Palmetto Local Fauna. In addition to the mutual occurrence of ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 179 Phocanella pumila in these two faunas, the pinni- peds Callophoca obscura and Ontocetus emmonsi are also shared, as well as the cetacean Ninoziphius platyrostris (Morgan, 1994). This suggests that the specimens from the Broad and Morgan rivers near Beaufort, SC, may have originated from a Sunken Meadow Member equivalent, perhaps the Wabasso beds, which, as noted previously, apparently occur in the shallow subsurface near Beaufort (Huddles- tun, 1988:98). This is further supported by the quality and, at one time, common and abundant presence of Megaselachus megalodon teeth from the Morgan River, given that this taxon last occurs during Sunken Meadow (= Wabasso) time (Ward, 2008). On the other hand, there is a cream-colored matrix within pore spaces of SC2009.3.1 that resembles Goose Creek Limestone. To the extent that Phocanella has not been found in strata more recent in age than Zanclean, the latest Zanclean age of the Goose Creek Limestone does not preclude that unit as the specimen’s source. MONACHINAE Gray, 1869 CALLOPHOCA Van Beneden, 1877 CALLOPHOCA OBSCURA Van Beneden, 1877 Figure 17M See Koretsky and Ray (2008) for synonymy. Referred Specimen.—SC2012.16.1, partial left humerus, collected by M. Swilp from Whale Branch region of Broad River, Beaufort County. Locality, Stratigraphic Horizon, and Age.— Questionably from Wabasso beds (or possibly Goose Creek Limestone); early Pliocene, late Zan- clean, early Blancan. Discussion.—Referral of SC2012.16.1 is based primarily on its similarity in size and mor- phology to those specimens assigned to this taxon from the Lee Creek Mine by Koretsky and Ray (2008). This humerus differs significantly from that of Phocanella pumila in being relatively shorter with a more stocky appearance and in the much larger size of the capitulum relative to the total length, which is 131 mm. This length, together with its somewhat gracile morphology, suggests that the element belonged to a female (see measurements in Koretsky and Ray, 2008). SC2012.16.1 represents the first record of Callophoca obscura in SC. Like Phocanella pum- ila discussed above, C. obscura was also originally described from the Zanclean of Belgium (Van Beneden, 1877), but it too occurs in the Lee Creek and Palmetto faunas (Ray, 1976a; Morgan, 1994; Koretsky and Barnes, 2008; Koretsky and Ray, 2008). Its presence in South Carolina, therefore, from deposits that also yielded Phocanella is not particularly unexpected. MONATHERIUM Van Beneden, 1877 MONATHERIUM sp. Figure 17N Referred Specimen.—ChM PV7688, distal part of humerus. Locality, Stratigraphic Horizon, and Age.— Near Summerville, Berkeley County (exact local- ity unknown), ?Ebenezer Formation; late Miocene, late Tortonian, late Barstovian. Discussion.—Identification of this specimen as Monatherium was provided by I. Koretsky (pers. comm. to AES, 2010). First described from Europe, this taxon ranges in age from the middle to late Miocene (Langhian to Messinian) and is known in North America from the Calvert, St. Mary’s, and Eastover formations of Virginia and Maryland Table 8. Measurements (mm) of femora of Phocanella pumila; (brkn = broken, sl worn = slightly worn). Specimen Maximum length Max. Width (proximal end) Max. Width (distal end) SC 98.60.6 (cast of USNM 181649) 124.3 60.2 58.8 SC 2009.3.1 125.2 (sl worn) 70.4 59.2 (brkn) SC 2015.18.4 120.3 (worn) 61.8 (worn) 59.6 180 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) (Ray, 1976a, b; Deméré et al., 2003); it is not known from the Yorktown or other Pliocene formations of the Atlantic Coastal Plain. In South Carolina the only middle to upper Miocene units are the Coo- sawhatchie and Ebenezer formations, respectively. According to Weems and Lewis (2002:29), the Ebenezer Formation is only known in the Charles- ton/Summerville area from “two small patches of shelly shelf sand, informally named the Rudd Branch beds …” They further noted that these beds represent “the only occurrences of any strata repre- senting the late Miocene” in that region. The mid- dle Miocene (lower Serravallian) Coosawhatchie Formation does not occur in the Charleston/Sum- merville area (Weems and Lewis, 2002). As with many of South Carolina’s vertebrate fossil occur- rences, however, and given the extremely limited distribution of these two “small patches,” it is likely that this specimen was reworked from the Ebene- zer Formation into a younger formation of greater extent in the Summerville area, such as the Goose Creek Limestone (see Weems and Lewis, 2002:fig. 5), if Koretsky’s identification of this fragmentary specimen is correct. PERISSODACTYLA Owen, 1848 EQUIDAE Gray, 1821 ANCHIPPUS Leidy, 1868b ANCHIPPUS TEXANUS Leidy, 1868b Figure 18A Anchippus texanus Leidy, 1868b. Anchippus texanus Leidy. Leidy, 1869; Albright, 1999b. Parahippus texanus (Leidy). Gidley, 1907; Osborn, 1918; Stirton, 1940. Parahippus cf. P. texanus (Leidy). Forsten, 1975, in part. Referred Specimen.—ChM PV13526, par- tial left maxillary with P4–M3. Locality, Stratigraphic Horizon, and Age.— “Ashley River phosphate beds” near Runnymede Plantation, about 15 km northwest of Charleston. Likely reworked into the upper Pleistocene Wando Formation from either the Chandler Bridge or Edisto formations; late Oligocene, late Chattian, Arikareean 3. Discussion.—ChM PV13526 is here referred to Anchippus texanus on the basis of the nearly identical morphology that the teeth share with the type specimen and with those referred to this taxon from the late Arikareean Toledo Bend Local Fauna of easternmost Texas (Albright, 1991, 1999b). Just as in A. texanus from Toledo Bend, the Charleston teeth, although slightly smaller (M1 = 17.5 AP x 20.0 TR, M2 = 16.8 AP x 19.3 TR, M3 = 15.3 AP x 18.3 TR; M1–M3 = 48.4 mm; P4 broken ante- riorly), are low-crowned, they lack cement, they have slightly crenulated enamel, the protocone and protoconule, as well as the metacone and metaco- nule, are distinctly separated and remain so at least through medium wear, and there is a single crochet extending anteriorly from the metaloph toward the division between the protocone and protoconule. The teeth also closely resemble those of the late Arikareean Parahippus nebrascensis and P. wyomingensis from the Great Plains. This simi- Figure 18. A, Anchippus texanus, ChM PV13526, partial left maxillary with P4–M3, from “Ashley River phosphate beds”; B, Nannippus sp., SC2006.1.24, and C, SC2006.1.23, upper left cheek teeth, from the Walrus Ditch locality, Waccamaw Formation; D, Subhyracodon mitis, right P2 (SCTC400; the palate shown, from which the P2 is derived, with right P2–4 and left P1–4, M1 or M2, and M3 is held in a private collection [see text for discussion]), from the Argos Cement quarry, Harleyville Formation; E, Aphelops ?malacorhinus, photograph of original left M2 from Wright River, Jasper County, from which cast SC2010.12.1 was made; F, ?Teleoceras gymonense, left p3–4, occlusal view, from Morgan River, Beaufort County; G, H, same specimen in lateral and lingual views, respectively; I, Perchoerus sp., ChM PV5025, right m2, from Chandler Bridge Formation; J–P, Bootherium bombifrons: J, USNM 533996, left M3, occlusal view, Surfside Beach, Horry County; K, labial view of same; L, SC75.31.204, right metacarpal III–IV, Edisto Beach; M, ChM PV40.183.17, right metacarpal III–IV, Edisto Beach; N, SC83.168.1, left metacarpal III–IV missing distal end, Edisto Beach; O, SC77.14.4, right metatarsal III– IV, dorsal view, Edisto Beach; P, ventral view of same. ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 181 182 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) larity is further emphasized through an unpub- lished manuscript prepared by O. P. Hay sometime between 1915 and 1923, and currently archived in The Charleston Museum, in which he described the Charleston specimen as the new species “Parahip- pus carolinus.” Hay also noted in this manuscript with “great interest ... that at least one species of the genus extended to the Atlantic coast” (O. P. Hay, unpublished ms, p. 4). ChM PV13526 is thought to have been col- lected, like so many other fossils from the Charles- ton area, from the “Ashley River phosphate beds,” which, as noted previously, occur in the middle member of the upper Pleistocene Wando Formation according to Sanders (2002). However, this speci- men is particularly intriguing because there is no doubt that it belongs to a taxon of late Arikareean (latest Oligocene to early Miocene) age based on its early parahippine grade. Thus, this specimen could not have originated from the Wando Formation, but must have been reworked into that unit from subjacent strata. Dinoflagellates found in matrix taken from the specimen do not include particu- larly restrictive species, although one is indicative of the Oligocene to middle Miocene (L. Edwards, pers. comm. to LBA, 2007). This limits the units from which the specimen could have originated to the upper Chattian Chandler Bridge Formation, the uppermost Chattian Tiger Leap or Edisto for- mations, or the upper Aquitanian Parachucla For- mation, all of which occur in the subsurface of the Charleston area (see Weems and Lewis, 2002:fig.5). The morphological grade of PV13526 appears too primitive to be considered as having originated from the middle Burdigalian Marks Head Forma- tion, which falls within the interval of time repre- sented by the Hemingfordian NALMA. Although a recently reported 87Sr/86Sr date of 23.4 Ma for the Edisto Formation (a date that falls within late Ari- kareean 3 of Albright et al., 2008) supports origina- tion of the specimen from this unit (Weems et al., 2006), the matrix found within the specimen, and from which samples were collected for microfossil analysis, most closely resembles the lithology of Bed 2 of the Chandler Bridge Formation. Another Arikareean taxon recovered from the Ashley phos- phate beds is the giant entelodont, Daeodon mento. But, as is discussed in more detail below, this spec- imen (MCZ 17015) has matrix adhering to it that is entirely unlike that associated with PV13526, and more indicative of an Edisto Formation origin. As Voorhies (1990) observed, many of the species currently referred to Parahippus are done so in error. Anchippus, therefore, may well be considered a valid genus, and the one to which many horses of this grade previously assigned to Parahippus should be referred, particularly P. nebrascensis and P. wyomingensis. MacFadden’s (1998:546) approach was to place the low-crowned “primitive parahippines,” including Anchippus, in the genus Desmatippus Scott, 1893. But, Anchip- pus has priority, having been erected by Leidy in 1868(b). In addition, there is now an abundance of material of A. texanus, rather than the single, dam- aged type specimen, based on the correlation of the Hutchen’s Well type specimen locality in Washing- ton County, Texas, to the Cedar Run Local Fauna locality, also in Washington County, and to the Toledo Bend Local Fauna in Newton County. Due to (1) the uncertain state of the genus Parahippus at this time, (2) the fact that the Charleston spe- cies lacks the derived features that typify Parahip- pus sensu stricto, and (3) because the Charleston specimen so closely resembles the type specimen and those specimens from Toledo Bend referred to A. texanus, ChM PV13526 is here referred to that taxon. NANNIPPUS Matthew, 1926 NANNIPPUS PENINSULATUS (Cope, 1885) See MacFadden (1984) and Hulbert (1993) for synonymy. Referred Specimens.—ChM PV7563, left M1 or M2; ChM PV7576, right M3; ChM PV7564, left M3; ChM PV7575, left M3; ChM PV7568, right p3 or p4; ChM PV7577, right ?m2; ChM PV7569, partial left m3. Locality, Stratigraphic Horizon, and Age.— Clapp Creek locality, Kingstree, Williamsburg County, ?lower Waccamaw Formation, early Pleis- tocene, late Blancan. Description.—ChM PV7563 is a worn M1 or M2, with only about 22 mm of tooth crown remain- ing. The tooth measures about 15 mm long (AP) by 16 mm wide (TR) and the oval protocone is about ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 183 5.7 mm long. There is no pli caballin or hypoconal groove, and fossette borders are relatively simple (weakly plicated). Although not collected in SC, another tooth here referred to N. peninsulatus is ChM PV7567 (55.103.41), a right M2. According to Charleston Museum archives, this tooth was collected by Ivan Tomkins in 1938 from spoil piles near Brunswick, Georgia (exact locality unknown). It thus represents the first record (although unpublished) of Nannip- pus from that state. This tooth differs from ChM PV7563 in the presence of a hypoconal groove and a more transversely compressed protocone. It has a crown height of about 33 mm and measures about 15 mm AP by 14 mm TR. Additional teeth of Nan- nippus from a Hemphillian site in Georgia were noted by Voorhies (1974). ChM PV7564, an M3, is relatively well worn with a crown height of about 30 mm. It measures about 16 mm AP by 13 mm TR, and has an elon- gate, transversely compressed protocone, a weak pli caballin, weakly plicated fossette borders, and a hypoconal groove. ChM PV7575 has a similar mor- phology to PV7564, although it lacks a pli caballin; it has a crown height of about 50 mm and it mea- sures about 15.5 mm AP by 10 mm TR. The broken base of the crown of another M3, ChM PV7576, precludes an estimate of its crown height, but its occlusal morphology resembles that of PV7575 in the absence of a pli caballin; it measures 16 mm AP by 12 mm TR. ChM PV7568, the p3 or p4, has a crown height of about 42 mm and it measures about 19 mm AP by 11 mm TR; the right ?m2, ChM PV7577, has a crown height of about 48 mm and it mea- sures about 16 mm AP by 10 mm TR. Both teeth are similar in having a moderately deep ectoflexid, no protostylid, and no pli caballinid. Discussion.—MacFadden (1984) recog- nized four valid species of Nannippus: N. minor, N. ingenuus, N. peninsulatus (to which he syn- onymized N. phlegon), and N. beckensis. Hulbert (1990) assigned populations previously referred to N. minor to N. aztecus, and then later (1993) estab- lished a fifth species, N. westoni, for late Clarendo- nian and earliest Hemphillian specimens previously referred to the latter. Hulbert (1993) also referred N. ingenuus of MacFadden (1984) to N. lenticu- laris, and he described a sixth species, N. morgani, from the late early Hemphillian of Florida. The teeth from the Clapp Creek locality are larger than those of N. morgani, thus far known only from late early Hemphillian sites in Florida; and they also differ in the absence of features that characterize the latter, such as well-developed pli caballins and pli caballinids, a shallow ectoflexid, and moderately complex fossette borders. The late early to late Hemphillian Nannippus aztecus (= N. minor) is also smaller than the Clapp Creek spe- cies. The largest species of Nannippus, the early to late Hemphillian N. lenticularis (= N. ingenuus of MacFadden, 1984), is larger than the Clapp Creek species, although its morphology is similar in hav- ing simple to moderate fossette plications, weak to absent pli caballins and pli caballinids, and moder- ately deep ectoflexids. Nannippus westoni, a latest Clarendonian through early Hemphillian species, is of similar size to that from Clapp Creek, but it differs in having a weak metastyle and prominent pli caballins and pli caballinids. The two species most similar to that from Clapp Creek are N. beckensis and N. peninsulatus. The former is known only from the early Blancan of Texas, whereas the latter is widely distributed and known best from late Blancan faunas. It is to the latter species that we refer the Clapp Creek mate- rial. The Clapp Creek species shares with N. pen- insulatus a prominent mesostyle with a restricted neck, a reduced hypocone that is labially retracted relative to the protocone (also seen in N. beckensis), a protocone that is divided nearly in half if a line is drawn transversely across the occlusal surface from the mesostyle (Hulbert, 1993:358), relatively simple fossette borders, a moderately deep ecto- flexid, and no protostylid. Apparently limited to the Blancan (Bell et al., 2004), N. peninsulatus was originally described from the “Loop Fork” shales of Tehuichila, Mexico (see MacFadden, 1984:138), but is well represented from Mt. Blanco, Texas, and is also known from several late Blancan sites in Florida (MacFadden and Waldrop, 1980; Mac- Fadden, 1984; Hulbert, 1993; Morgan and Hulbert, 1995; Morgan, 2005). As noted in Bell et al. (2004), one of the youngest occurrences is in the Macas- 184 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) phalt Shell Pit fauna at about 2.2 Ma (Morgan, 2005), although R. Hulbert (pers. comm. to LBA May, 2019) noted what may be the youngest occur- rence in North America at Florida’s Withlacoochee River 1A site dated biochronologically to approxi- mately 2.0 Ma. The disappearance of Nannippus in the southwest (New Mexico) nearly coincides with the first arrival of South American immigrants near the Gauss-Matuyama boundary at approximately 2.6 Ma (Tedford, 1981). Additionally supporting our referral of the Clapp Creek material to N. pen- insulatus is the co-occurrence there with Ondatra idahoensis (see above), a late Blancan rodent. The mutual presence of these late Blancan taxa implies a likely origin from either the Cypresshead or the lower Waccamaw formations. Efforts to locate the tooth described by Leidy (1859) as Hipparion venustum from the “Ashley River phosphate beds,” also mentioned by Voor- hies (1974) and MacFadden (1984:126), and illus- trated by Osborn (1918:200), were unsuccessful. NANNIPPUS sp. Figure 18B–C Referred Specimens.—SC2006.1.23, right P4 or M1; SC2006.1.24, right upper cheek tooth; SC2006.1.26, fragment of upper cheek tooth; SC2006.1.22, partial upper cheek tooth; SC2006.1.27, partial left upper cheek tooth; SC2006.1.25, labial half of upper cheek tooth; SC89.245.11, fragment of lower cheek tooth; SC89.245.12, fragment of lower cheek tooth; SC2006.1.21, left m3. Locality, Stratigraphic Horizon, and Age.— Walrus Ditch locality, Dorchester County, lower Waccamaw Formation; early Pleistocene, late Blancan. Description.—SC2006.1.23, a P4 or M1, measures 16.4 mm AP by 15.3 mm TR. It has a “necked” metastyle, a hypoconal groove, a single pli caballin, an oval protocone that is somewhat flattened medially, and the posterior pre-fossette and anterior post-fossette borders are well crenu- lated. SC2006.1.24, another upper cheek tooth, also has a “necked” metastyle, a hypoconal groove, well crenulated opposing fossette borders but no pli caballin, and an oval protocone although more elongate than in SC2006.1.23. SC2006.1.24 mea- sures 16.3 mm AP by 16.0 mm TR. Discussion.—In the collections at the SCSM are several teeth from the Walrus Ditch locality that represent a different species than that from Clapp Creek. They are larger than the Clapp Creek speci- mens, and the upper cheek teeth have highly plicated opposing pre- and post fossette borders. The occlu- sal pattern of the upper teeth most closely resem- bles that of N. aztecus from the latest Hemphillian Palmetto Fauna of Florida (Webb et al., 2008:fig. 8A) and Nannippus sp. A from gravel pits along the Nueces River, Texas (Baskin, 1991), but the Walrus Ditch teeth are distinctly larger. Although listed as Nannippus peninsulatus by Fields et al. (2012:15) in their discussion of the Walrus Ditch assemblage, we consider these teeth referable to a different, yet-to-be determined, species. RHINOCEROTIDAE Owen, 1845 SUBHYRACODON Brandt, 1878 SUBHYRACODON MITIS (Cope, 1875) Figure 18D See Prothero (2005:46) for synonymy. Referred Specimen.—Right P2; SCTC400. Locality, Stratigraphic Horizon, and Age.— Argos Cement quarry, near Harleyville, Dorchester County, Harleyville Formation; lower calcareous nannoplankton zone NP21, late Eocene, latest Pri- abonian, late Chadronian. Discussion.—As noted in the discussion of the Harleyville Formation earlier in this report, a rhinoceros specimen was collected from the Argos Cement quarry near Harleyville, SC, by Mr. J. Metts, an avocational fossil collector who allowed the first author to examine it. The specimen consists of a palate with teeth of the primarily Chadronian to Whitneyan taxon, Subhyracodon. It includes the right P3–4 and the left P1–4, M1 or M2, and M3. The identification of the M1 or M2 is equivocal because the palate is evidently pathologic – only a single molar is emplaced between the P4 and M3 – but the tooth is most likely M2 based on size and morphology. Unfortunately, the palate is currently retained in Mr. Metts’ private collection, and there- fore not immediately available for further study. Fortunately, however, a fragmentary P2 belonging to the same individual as the palate is curated in the ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 185 SCSM collection, thus providing the basis for this most important record. According to Prothero (2005:47), Subhyra- codon mitis is “the common late Chadronian rhino.” Although S. occidentalis is also known from the late Chadronian, referral of the SC specimen to S. mitis is based on the primitive, non-molariform morphology of the P3 and P4 and on its smaller size. As Prothero (2005) further noted, P3 and P4 of S. mitis have strongly posteriorly projecting pro- tolophs and unconnected metalophs, exactly the condition of the SC specimen, whereas in S. occi- dentalis the protoloph shows only a trace of poste- rior projection with a strong and merged metaloph. The specimen was collected from one of the many shallow solution pits filled with Har- leyville Formation that commonly occur across the upper surface of the Pregnall Member of the Tupelo Bay Formation. Dinocysts (e.g., Batiacas- phaera baculata, B. compta, Cordosphaeridium funiculatum, Homotryblium plectilum, Samlandia chlamydophora, and Trigonopyxidia fiscellata) and calcareous nannoplankton (e.g., Ismolithus recurvus, Ericsonia formosa, and Reticulofenes- tra umbilicus) analyzed from matrix associated with the palate indicate assignment to lower nan- noplankton zone NP21 (USGS Paleobotanical sample number R6747, L. Edwards and J. Self- Trail, pers. comm. to LBA, 2012). Although NP21 spans the Priabonian-Rupelian boundary (Eocene- Oligocene boundary), which closely approximates the Chadronian-Orellan boundary, we consider this specimen to be latest Chadronian (Ch3) in age. The Chadronian rather than Orellan age assignment is based on the additional discovery of a brontothere upper molar from the same horizon at the nearby Giant Cement quarry (also retained by the same owner as the Subhyracodon palate); brontotheres are not known to have extended into the Oligocene in North America (Mihlbachler, 2008). Although the bulk of this specimen is cur- rently held in a private collection, the curated P2 from the same individual provides the first and only record of Subhyracodon east of Mississippi (Man- ning, 1997; Albright et al., 2016), as all other records are from California, the northern and central Great Plains, and Oregon (Prothero, 1998, 2005). APHELOPS Owen, 1845 APHELOPS ?MALACORHINUS Cope, 1878 Figure 18E Referred Specimens.—SC2010.12.1, cast of left M2 (labeled “Teleoceras”); SC2015.18.3, ecto- loph of right M2. Locality, Stratigraphic Horizon, and Age.— SC2010.12.1, Wright River, Jasper County; SC2015.18.3, Chechessee River, Beaufort County; stratigraphic unit undetermined, but possibly Coo- sawhatchie or Ebenezer formations; middle to late Miocene, late Barstovian or early Hemphillian, respectively. Discussion.—Like the brontothere tooth and Subhyracodon palate noted above, the origi- nal specimen from which SC2010.12.1 was cast resides in a private collection. Fortunately, the first author was allowed to observe, measure, and pho- tograph the original specimen. It is in medium wear stage with a crown height of about 38 mm, a maxi- mum AP length of about 60 mm, and a TR width of about 53 mm. Aphelops and Teleoceras are represented by several species that collectively range from the late Hemingfordian to the latest Hemphillian, and species of both genera have been recorded from Florida (Hulbert, 2001; Prothero, 2005). Although the stage of wear of SC2010.12.1 makes it difficult to discern which taxon may be represented, it does not appear to be Teleoceras. Upper molars of Tele- oceras typically have a prominent antecrochet that extends posteriorly from the protoloph, lingual to the crochet - a feature lacking in both Aphelops and in the SC tooth. The tooth does have what might be considered a small antecrochet, but this is not the same structure as in Teleoceras; the structure in SC2010.12.1 is labial to the crochet (i.e., it does not originate from the protoloph). Regardless of which taxon is represented, these specimens pro- vide an exceptionally rare record of a Miocene land mammal from SC. TELEOCERAS Hatcher, 1894 TELEOCERAS ?GUYMONENSE Prothero, 2005 Figure 18F–H Referred Specimens.—SC2017.14.4a and b, 186 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) cast of associated left p3–4 in fragment of man- dible (fragment with p3 = SC2017.14.4a; fragment with p4 = SC2017.14.4b). Locality, Stratigraphic Horizon, and Age.— Morgan River, Beaufort County; stratigraphic unit undetermined, but possibly Ebenezer Formation or Wabasso beds; late Miocene to earliest Pliocene, early to latest Hemphillian. Description and Discussion.—This speci- men, too, currently resides in a private collection, but the collector, M. Swilp, graciously allowed the first author to observe, measure, and photograph it, and to have it cast by the SCSM. A digital file of this specimen was also created by scanning the speci- men with a CT scanner at the Johnson and Johnson Laboratory within the Department of Engineering at the University of North Florida. Measurements of the teeth are as follows: p3, 28 mm AP x 19.7 mm TR; p4, 32 mm AP x 22.2 mm. Both the p3 and p4 have a small, but distinct cingulum that wraps around the anterior surface of the tooth from the antero-labial corner to the antero- lingual corner. There is also evidence of a posterior cingulum, but it is worn down due to abutment of the teeth. Approximately 1 cm above the base of the crown on the labial surface is a weak, nearly indistinguishable rugosely textured cingulum (which does not appear to be due to water wear), and there is a weak cingular segment protecting the posterior lingual reentrant of the p4, as well. The anterior lingual reentrant of both teeth is very shal- low. The teeth are in medium wear stage and they show no cement. There is a wear facet on the ante- rior surface of p3 indicating the presence of a p2. Rhinoceroses in North America are known to have gone extinct in the latest Hemphillian, the last known taxa including Teleoceras hicksi Cook, 1927, from Florida’s Palmetto Fauna; the dwarf species T. guymonense Prothero, 2005, known only from Oklahoma, Texas, Kansas, and New Mexico; and T. aepysoma Short et al., 2019, from the Gray Fossil Site of eastern Tennessee. The species from SC is much smaller than T. hicksi, T. aepysoma, and the above noted Aphelops malacorhinus, but is similar in size to T. guymonense. It differs from the latter in having a wear facet on the anterior surface of p3, thus indicating the presence of p2. Although the p2 is typically lost in later species of the genus, Prothero and Manning (1987) noted that the p2 is occasionally retained, and this was recently exem- plified by specimens of T. aepysoma described by Short et al. (2019) that ranged from having no p2s, to having vestigial p2s, to specimens that retained the p2s. Aphelops malacorhinus also retains p2, but this species is considerably larger than the Morgan River taxon. Measurements of the small Barstovian rhino Peraceras hessei Prothero and Manning, 1987, indicate a size similar to the SC species, and P. hessei retains the p2, but the lower teeth of the latter differ from those from SC in the presence of prominent labial cingula. Another small, but pri- marily Barstovian species is T. meridianum (Leidy, 1865), known from the Texas Coastal Plain. It too shows variation in the presence/absence of the p2 and, like the Morgan River species, it lacks promi- nent cingula (Prothero and Manning, 1987). In size and morphology, therefore, the Morgan River species most closely resembles T. meridianum and T. guymonense. Our referral of these small rhino teeth from SC to the latter spe- cies rather than the former is based more on their purported age than on any particulars of morphol- ogy – the mainly Barstovian age of T. meridianum vs. the late Hemphillian age of T. guymonense. Although the SC specimen was found out of strati- graphic context on the bottom of a river, support for its latest Hemphillian age is provided by the recovery of specimens of Phocanella pumila and gomphothere teeth likely representative of Rhyn- chotherium from the same locality – two taxa also known from Florida’s latest Hemphillian Palmetto Fauna (and from which T. hicksi is also known). The recovery of upper molars from the Morgan River locality would help refine our identification because, according to Prothero (2005:124), those of T. guymonense show a unique character in the form of a “short but distinct rib on the lingual face of the crochet, which produces a short lingual spur on the wear surface.” Rhinocerotidae, Genus and species indeterminate Referred Specimen.—MCZ 17134, “a large flake from the outer side of a molar,” from the Rob- ert Wilson collection (Allen, 1926:454). ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 187 Locality, Stratigraphic Horizon, and Age.— from the “phosphate beds of Ashley River, South Carolina” near Charleston (Allen, 1926:447); age not determinable. Discussion.—In his paper on fossil mam- mals from South Carolina, Allen (1926:454) noted a fragment “from the outer side” of an upper molar of a rhinoceros that he referred to ?Teleoceras proterus (Leidy).” In the same discussion, he also mentioned the note of rhinoceros remains from the Ashley River beds by Packard (1871). Because the Ashley phosphate beds include a temporally mixed assemblage, rhinoceros remains from this unit could be referable to a number of species includ- ing Diceratherium, Menoceras, Teleoceras, or Aphelops. The MCZ fragment is non-diagnostic. ARTIODACTYLA Owen, 1848 ENTELODONTIDAE Lydecker, 1883 DAEODON Cope, 1878 DAEODON MENTO (Allen, 1926) Dinohyus (?) mento Allen, 1926. Referred Specimen.—MCZ 17015, symphy- seal region of mandible, from the Robert Wilson Collection (Allen, 1926). Locality, Stratigraphic Horizon, and Age.— “Ashley River phosphate deposits,” considered to have been reworked into the Wando Formation from the uppermost Chattian Edisto Formation; late Arikareean (Ar3). Discussion.—This specimen still represents the only record of an entelodont from SC (although rumors suggest that additional material may exist in private collections), and is noted again here because of its record as a second Arikareean land mammal taxon from the “Ashley River phosphate beds” together with the above noted Anchippus texanus. MCZ 17015 was evidently reworked into the Wando Formation from the Edisto Formation, which has been determined to be of latest Chattian age (Weems and Harris, 2008; Weems et al., 2006; Weems et al., 2016). Other possible units of ori- gin include the upper Rupelian Ashley Formation, the upper Chattian Chandler Bridge Formation, the lower Miocene (Aquitanian) Parachucla Forma- tion, and the lower Miocene (Burdigalian) Marks Head Formation, as all span, respectively, the early Arikareean to early Hemingfordian – the interval of time over which the giant entelodonts are known in North America. These formations, however, are excluded as the units of origin on the basis of infor- mation provided below. Although Effinger (1998) listed “?D. mento” from the Hawthorn Formation, it should be noted that this formation is no longer recognized in SC. The Hawthorn Formation was once represented in SC by the Marks Head Member, but this unit has more recently been elevated to formational status (Huddlestun, 1988). The Marks Head For- mation, described in the Dorchester County area by Edwards et al. (2000) as a “clayey phosphatic quartz sand,” is excluded as the unit of origin for MCZ 17015 on the basis of its dissimilar lithology relative to the matrix still adhering to the speci- men. Another option is that the entelodont speci- men originated from the upper Rupelian (earliest Arikareean) Ashley Formation, often referred to as the “Ashley marl.” As Allen (1926) noted for MCZ 17015, “the fact that the posterior tooth-sockets are filled with indurated marl may indicate that the fragment was deposited in the earlier Tertiary marl beds while they were forming.” However, it is our opinion that this specimen originated from the Edisto Formation, rather than the Ashley Formation. First, Edwards et al. (2000) described the Edisto Formation as a “quartz and phosphate sandy, calcarenite.” This description closely matches the lithology of the matrix adher- ing to MCZ 17015, particularly considering that the matrix also effervesces when hydrochloric acid is applied (R. Weems, pers. observ., 2009). Further precluding origination from the Ashley Formation, and additionally supporting derivation from the Edisto Formation, are two lines of circumstantial evidence: (1) the Edisto Formation is a more inshore and shallow water unit, and therefore more amena- ble to harboring fossils of terrestrial mammals, than the mid-shelf depositional environment of the Ash- ley Formation, and (2) the reported 87Sr/86Sr date of 23.4 Ma for the Edisto Formation is a more likely age for Daeodon, based on its temporal span, than is the 28-29 myr age of the Ashley Formation. 188 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) TAYASSUIDAE Palmer, 1897 PERCHOERUS Leidy, 1869 PERCHOERUS sp. Figure 18I Referred Specimen.—ChM PV5025, right m2. Locality, Stratigraphic Horizon, and Age.— Bank of drainage ditch approximately 0.52 km south of County Road 996, Goose Creek, Charles- ton County, Chandler Bridge Formation; late Oli- gocene, Chattian, Arikareean 3. Discussion.—The serendipitous discovery of this isolated peccary tooth (16.4 mm AP, 14.9 mm TR, crown height 10 mm) during excavation of fossil whale material from the Chandler Bridge Formation represents the only terrestrial mammal material found in situ from the Oligocene of South Carolina. Material of other latest Oligocene/earli- est Miocene land mammals (Anchippus and Dae- odon) were found as specimens reworked into the “Ashley River phosphate beds.” According to Wright (1998:fig. 26.5) and Prothero (2009), there are only two valid North American genera of Oligocene tayassuids, Thino- hyus and Perchoerus. The Chandler Bridge tooth is similar in size to those of both taxa. Although its morphology more closely resembles that of Thino- hyus, peccary teeth are highly variable and noto- riously unsuitable for alpha level taxonomy. With only a single tooth available, we tentatively assign this specimen to Perchoerus based on the broader geographic distribution this taxon had than that of Thinohyus, which is known only from the John Day Formation, Oregon (Prothero, 2009). MYLOHYUS Cope, 1889 MYLOHYUS FOSSILIS (Leidy, 1860) Referred Specimens.—ChM PV2499, right m1; ChM PV4986, right m2; USNM 530190, left m3 (additional specimens from SC noted in Roth and Laerm [1980] and Bentley et al. [1994]). Locality, Stratigraphic Horizon, and Age.— ChM PV2499 and ChM PV4986 from Edisto Beach, Colleton County, undetermined offshore unit; late Pleistocene, late Rancholabrean. USNM 530190 from Myrtle Beach, Horry County, unde- termined offshore unit; late Pleistocene, late Ran- cholabrean. Discussion.—The m1 is represented by what appears to be an unerupted crown, and it measures 15.5 mm AP by 12.5 mm TR; the m2 is some- what worn and measures 16 mm AP by 13 mm TR; and the m3 measures 18.5 mm AP by 14 mm TR. Referral to M. fossilis rather than M. nasutus (Leidy, 1869) follows Lundelius (1960:34), who concluded that the former was the east coast spe- cies, whereas the latter was “more western in its distribution,” and Wright (1995, 1998) rather than Kurtén and Anderson (1980). Wright (1998) also noted that M. fossilis spans the late Blancan through Rancholabrean. In the collections of the SCSM are additional specimens currently under study from the late Irvingtonian Camelot Local Fauna. CERVIDAE Gray, 1821 CERVUS Linnaeus, 1758 CERVUS ELAPHUS Linnaeus, 1758 Referred Specimens.—ChM PV7595, partial right m3; McK67.41.169, unerupted M2. Locality, Stratigraphic Horizon, and Age.— ChM PV7595 from Clapp Creek locality, Kingstree, Williamsburg County, unit of origin unknown; McK67.41.169 from Eddingsville Beach, Colleton County, undetermined offshore unit, late Pleisto- cene, Rancholabrean. Discussion.—Sanders (2002) reported the first remains of fossil Cervus in SC from Edisto Beach and from a site northwest of Charleston in the Penholoway Formation, providing a middle Irvingtonian through Rancholabrean age range. Although the majority of mammal fossils recov- ered from the Clapp Creek locality are of middle- to-late Blancan age, the nature of the deposit there (a lag deposit resting upon the Maastrichtian Steel Creek Formation; see above) results in a mixed fauna that includes taxa from the late Blancan, late Rancholabrean, and even the late Cretaceous and early Paleocene. (Note: Ludt et al. [2004] consid- ered the species of Cervus that dispersed from NE Asia into North America to be Cervus canadensis rather than C. elaphus.) RANGIFER Hamilton-Smith, 1827 RANGIFER TARANDUS (Linnaeus, 1758) Referred Specimen.—SC2010.13.151, right ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 189 dP3. Locality, Stratigraphic Horizon, and Age.— Crowfield locality, Dorchester County, Wando For- mation; late Pleistocene, late Rancholabrean. Discussion.—Although records of caribou in SC were noted previously by McDonald et al. (1996, 2000) and Sanders (2002), the importance of such boreal taxa occurring at what is likely the southern extreme of their Pleistocene range in the eastern USA warrants mention of this isolated spec- imen, as well. The tooth measures about 15 mm AP by 11 mm TR, and is from the Rancholabrean-aged Crowfield locality currently under study by JLK, DJC, and F. Grady (USNM). The southern-most extent of Pleistocene Rangifer appears to be based on a specimen from Bartow County, Georgia (Mar- tin and Sneed, 1989). BOVIDAE Gray, 1821 BOOTHERIUM Leidy, 1852 BOOTHERIUM BOMBIFRONS (Harlan, 1825) Figure 18J–P Referred Specimens.—USNM 533996, left M3, collected “near pier at Surfside Beach,” Horry County; USNM PAL 530183, partial cheek tooth, Myrtle Beach, Horry County; SC92.121.1, cast of partial m3, from St. Helena Sound, Beau- fort County; SC77.8.17, radius-ulna, from Edisto Beach, Colleton County; ChM PV40.183.17, right Mc III–IV from Edisto Beach; SC75.31.204, right Mc III–IV, from Edisto Beach; SC83.168.1, left Mc III–IV missing distal end, from Edisto Beach; SC2015.17.1, left Mc III–IV, from west branch of Cooper River; SC77.14.4, right Mt III–IV, from Edisto Beach; ChM PV7199, left calcaneum, from the Holcim (formerly Holnam) Cement Company quarry, Harleyville, Dorchester County. Stratigraphic Horizon and Age.—Speci- mens from Myrtle, Surfside, and Edisto beaches, undetermined offshore unit, late Pleistocene, late Rancholabrean; SC2015.17.1 from the Cooper River, ?Wando Formation, late Pleistocene, Ran- cholabrean; ChM PV7199, unnamed Pleistocene deposits overlying Harleyville Formation; late Rancholabrean. Description.—USNM 553996, the well pre- served M3, measures about 4.1 cm long by 2.9 cm wide and it has a crown height of about 4.5 cm. Additional elements of Bootherium from SC include a partial radius-ulna and fused metacar- pals and metatarsals, measurements of which are provided in Table 9. Referral to Bootherium rather than to Bison is based primarily on the significantly more robust morphology of matching elements for the latter. The calcaneum was determined to repre- sent that of Bootherium on the basis of its similar- ity to that of Ovibos moschatus, the extant Musk Ox. It is of similar size to calcanea of Hemiauche- nia and Palaeolama (from Leisey), but differs in the following features: (1) it has relatively paral- lel dorsal and ventral surfaces when viewed from a lateral perspective vs. distally diverging dorsal and ventral surfaces in the camels (resulting in sig- nificantly more bone ventral to the sustentacular facet); (2) the dorsal surface of the calcaneal heel is pinched and forms a ridge vs. the broad dorsal surface in camels; (3) the dorsal surface of the cal- caneal heel has a concavity immediately before the calcaneal tuberosity. Discussion.—One previous record of the woodland musk ox, Bootherium bombifrons, from SC was reported by McDonald et al. (2000) in which they noted an m3 from Saint Helena Sound, Beaufort County – the southern-most occurrence of musk ox along the Atlantic Coastal Plain. That record, together with those noted herein, indicates that Bootherium was broadly distributed along the coastal plain of SC during the Wisconsian, if not before (Fig. 19). The presence of both Bootherium and Rangifer in SC supports the existence of what McDonald et al. (2000:132) referred to as a “cohort of large-bodied boreal ungulates ... in the southeast- ern extremes of their ranges.” To date, Bootherium is not known from Georgia or Florida. The calcaneum, ChM PV7199, was collected by B. Palmer in October, 2001, from Pleistocene sediments overlying the Harleyville Formation in the Holcim Cement plant quarry near Harleyville, Dorchester County. Based on his familiarity with the stratigraphy and paleontology of the Harleyville area, Palmer concluded that the sediments from which the specimen originated were very likely equivalent to those from which the nearby Ardis 190 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) Local Fauna was recovered. In light of the fact that Palmer did not know to which taxon the specimen belonged, his conclusion was an astute one, as the Ardis LF was recovered from beds 14C dated to between 18,530 and 18,940 years old (Bentley et al., 1994). This is exactly the interval of time over which Bootherium would be expected to be as far south as South Carolina, i.e., during the last glacial maximum. SIRENIA Illiger, 1811 PROTOSIRENIDAE Sickenberg, 1934 PROTOSIREN Abel, 1907 PROTOSIREN sp. indet. Figure 20A Referred Specimens.—GSM 1333, partial left inominate; SC2015.65.1, parietal-supraoccip- ital skull cap. Locality, Stratigraphic Horizon, and Age.— GSM 1333, Martin Marietta Orangeburg Quarry, Orangeburg County, lower part of Cross Member of Tupelo Bay Formation “within 30 cm of the Santee Limestone/Tupelo Bay Formation contact” (Beatty and Geisler, 2010:2); NP17, late Eocene, Bartonian, late Uintan. SC2015.65.1, Giant Cement quarry, Dorchester County, Tupelo Bay Formation; NP17 or 18, late Eocene, Bartonian or Priabonian, Table 9. Measurements (mm) of selected postcranial elements of Bootherium bombifrons from SC. Specimen Maximum Length Width (prox. end) Width (distal end) Metacarpals III–IV ChM PV40.183.17 230 SC75.31.204 253 81.5 82.1 SC83.168.1 -- 81.3 missing Metatarsal III–IV SC77.14.4 284 62.5 74.2 Radius/ulna SC77.8.17 410 87.8 92.6 Calcaneum ChM PV7199 121 33.8 -- Figure 19. Atlantic Coastal Plain of USA showing localities where specimens of fossil musk ox (O = Ovibus; B = Bootherium) have been recovered. Base map, localities 1–7, and cave localities after McDonald and Ray (1993:fig. 1); SC localities 8–12: 8, Surfside and Myrtle beaches, Horry County; 9, Cooper River, Berkeley County; 10, Harleyville, Dorchester County; 11, Edisto Beach, Colleton County; 12, St. Helena Sound, Beaufort County. ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 191 Duchesnean to early Chadronian. Discussion.—GSM 1333 was collected by B. Palmer from the same stratigraphic horizon as the holotype of Carolinacetus gingerichi Geisler et al., 2005. Noting the “striking similarity” of this specimen to that of Protosiren sattaensis from the Bartonian of Pakistan, Beatty and Geisler (2010:2) conservatively referred it to Protosiren sp. Another specimen provisionally referred to Protosiren, SC2015.65.1, was collected by V. McCollum from a spoil pile at the bottom of the Giant Cement plant quarry. DUGONGIDAE Gray, 1821 EOTHEROIDES Palmer, 1899 EOTHEROIDES sp. Referred Specimens.—SC2006.30.1–15, partial skull with limb material (15 elements num- bered individually); SC2013.32.1, partial skull with M1–2; USNM 537206, partial skull and par- tial skeleton; ChM PV7639, fragment of skull cap. Locality, Stratigraphic Horizon, and Age.— SC2006.30.1–15 and USNM 537206 from Giant Cement quarry, Dorchester County, Tupelo Bay Formation, Pregnall Member, late Eocene, NP18, early Priabonian, early Chadronian; SC2013.32.1 from Argos Cement quarry, Dorchester County, Tupelo Bay Formation, Pregnall Member, late Eocene, NP18, early Priabonian, early Chadro- nian; ChM PV7639 from Giant Cement quarry, Harleyville Formation-filled pits eroded into upper surface of Pregnall Member of Tupelo Bay Forma- tion, late Eocene, NP21, late Priabonian, Chadro- nian (possibly reworked into Harleyville Forma- tion from underlying Pregnall Member). Discussion.—Because these specimens are currently being described as potential new species of Eotheroides by D. Domning and I. Zalmout, we mention them only as records of this genus from SC. The two SCSM specimens were collected by V. McCollum and the USNM skeleton was col- lected by B. Palmer. METAXYTHERIUM de Christol, 1840 ?METAXYTHERIUM ALBIFONTANUM Vélez- Juarbe and Domning, 2014b Figure 20B–C Referred Specimen.—ChM PV9480, partial left m3. Locality, Stratigraphic Horizon, and Age.— Drainage ditch south to southwest, and between 0.6 and 1.7 km, of the Charleston Airport Terminal, Charleston County, Ashley Formation; early Oli- gocene, late Rupelian, Arikareean 1. Description and Discussion.—ChM PV9480 is missing the anterior portions of the protoconid and metaconid, as well as the lingual surface of the metaconid, entoconid, and the lingual hypoconu- lid lophule (terminology follows Vélez-Juarbe and Domning, 2014b:fig.8, p. 453). A prominent cris- tid obliqua blocks the transverse valley, and the hypoconulid lophule consists of two cusps, the lin- gual of which is crescentic in shape. In its broken state it measures about 18 mm AP by 14 mm TR. Both measurements would be slightly greater if the tooth was complete. The morphology of the tooth closely resembles that of the middle to late Mio- cene Metaxytherium floridanum Hay, 1922, but ChM PV9480 is smaller. Furthermore, the tooth is thought to have originated from the middle Oligo- cene (upper Rupelian) Ashley Formation based on the nature of the matrix adhering to it; thus it is much older than M. floridanum. According to Vélez-Juarbe et al. (2012) and Vélez-Juarbe and Domning (2014a), three species of late Oligocene dugongids were sympatric across at least part of their ranges in the western Atlantic region, and remains of each of these, Metaxythe- rium albifontanum, Crenatosiren olseni (Reinhart, 1976), and Dioplotherium manigaulti, are known from the Chandler Bridge Formation. Recently, Vélez-Juarbe and Domning (2015) described a fourth late Oligocene species from this region (Puerto Rico), Callistosiren boriquensis. Early Oligocene taxa in this region include Priscosiren atlantica Vélez-Juarbe and Domning (2014a) from the San Sebastian Formation of Puerto Rico, of similar age to the Ashley Formation, as well as the above noted Crenatosiren olseni, which has also been recorded from the Ashley (Domn- ing, 1997). The m3s of both of these taxa are sig- nificantly smaller and different in morphology than ChM PV9480 (LBA, pers. observ.). Although the larger Chandler Bridge taxa, Metaxytherium albifontanum and Dioplotherium 192 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) Figure 20. A, Protosiren sp., SC2015.65.1, parietal-supraoccipital skull cap, from the Giant Cement quarry, Dorchester County, in dorsal (left) and ventral (right) views; B, ?Metaxytherium albifontanum, ChM PV9480, partial left m3, stereo occlusal view and C, labial view, Ashley Formation, Charleston County; D, Rhynchotherium falconeri, SC2017.6.1, left m3. ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 193 manigaulti, are not yet known from the Ashley Formation, their teeth (no m3s are known for the latter) are similar in size to ChM PV9480. So too are the M3s of Callistosiren boriquensis, but this taxon is not yet recorded from the Oligocene of South Carolina (nor are m3s known yet). This sug- gests that ChM PV9480 may belong to one of the former two species if either of them are ever deter- mined to have ranged down into the Ashley For- mation. Metaxytherium albifontanum specimens SC89.255.1 and SC89.255.2 include only upper teeth, but another specimen, SC89.115, includes an m3, which measures 22.2 mm AP x 13.66 mm TR. The partial ChM PV9480 would be similar in size if it was complete. If our identification to M. albi- fontanum is correct, then its range, too, is herein adjusted down into the upper Rupelian from the Chattian. PROBOSCIDEA Illiger, 1811 GOMPHOTHERIIDAE Hay, 1922 RHYNCHOTHERIUM Falconer, 1868 RHYNCHOTHERIUM FALCONERI Osborn, 1923 Figure 20D Referred Specimens.—SC2017.6.1, left m3; SC2017.6.2, partial right M3; SC2006.1 (accession number only), partial molar. Locality, Stratigraphic Horizon and Age.— SC2017.6.1 and SC2017.6.2 from a site in Dorchester County with a Walrus Ditch-equivalent assemblage (detailed locality information on file at SCSM); SC2006.1 from Walrus Ditch locality, Dorchester County, Waccamaw Formation, late Pliocene, late Blancan. Discussion.—Based on evidence from Flor- ida, there are three different proboscidean families that occurred in the southeast: the Mammutidae, the Gomphotheriidae, and the Elephantidae (Hulbert, 2001). The various species recorded from Florida include the mammutids Zygolophodon tapiroides, Mammut sellardsi, and Mammut americanum; the gomphotheriids Gomphotherium calvertense, G. simplicidens, Amebelodon floridanus, A. britti, Platybelodon sp., Rhynchotherium falconeri (see below), and Cuvieronius hyodon; and the elephan- tid Mammuthus columbi (Hulbert, 2001; Lister and Sher, 2015; Morgan and Harris, 2015; Morgan et al., 2016). These are the taxa, therefore, that might be expected to be found in the fossil proboscidean record of South Carolina. Excluding the abundance of material, both dental and skeletal, confidently referred to the American Mastodon, Mammut americanum, and to the Columbian Mammoth, Mammuthus columbi, the remainder of fossil pro- boscidean material from the state is almost exclu- sively comprised of teeth, fragments of teeth, and fragments of jaws. Unfortunately, the highly vari- able condition of the gomphothere dentition makes it nearly impossible to determine which taxon is represented when teeth are the only record avail- able (see Lucas and Morgan, 2008:74). As Hulbert (2001:315) noted, “Relatively complete skulls and jaws with the mandibular symphysis and lower tusks are needed for secure identifications.” The temporal range of these taxa can be used with some utility as an aid to identification, but only cautiously, as several taxa overlap in age. For example, Zygolophodon, Gomphotherium, Amebe- lodon, and Platybelodon are all present in the late Miocene (although the former two also occur in the middle Miocene), which due to the paucity of fossil-bearing exposures in SC of that age would not be expected to be found in the state, at least not commonly. On the other hand, the few, rare, fos- sil rhinoceros specimens reported herein provide tantalizing glimpses into this interval of time and portend the possibility that late Miocene probos- cidians may eventually be found, if they haven’t been already. That leaves Rhynchotherium and Cuviero- nius. In a detailed and much needed study on the taxonomy of Rhynchotherium, Lucas and Morgan (2008) concluded that only one of several named species was valid, Rhynchotherium tlascalae Osborn, 1918, although they considered the holo- type lower jaw of this species as representative of the genus Gomphotherium, not Rhynchotherium. Therefore, they established as a neotype for R. tlas- calae the type specimen lower jaw of R. browni Osborn, 1936, AMNH 15550. Synonymizing R. browni, R. edense, R. falconeri, and R. simpsoni with R. tlascalae, they also concluded that Rhyn- chotherium sensu stricto was solely a North Amer- 194 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) ican taxon. Subsequently, however, Lucas (2010) applied for and obtained a ruling from the ICZN to recognize R. falconeri as the type species based on Osborne’s (1923) holotype lower jaw (AMNH 8532) from the late Blancan Mt. Blanco LF of Texas, which “shows the key diagnostic features of Rhynchotherium sensu stricto” (Lucas and Mor- gan, 2008:78). Typically considered a late Hemphillian through late Blancan taxon with a broad North American distribution, R. falconeri is known from the Palmetto (latest Hemphillian) and Macasphalt Shell Pit (late Blancan) faunas of Florida. The late Hemphillian Rhynchotherium from Florida was originally named R. simpsoni by Olsen (1957), then referred to R. edense by Webb et al. (2008), but synonymized with R. falconeri by Lucas and Morgan (2008; also see Morgan and Harris, 2015). Cuvieronius, however, is considered to have an early Irvingtonian through Rancholabrean range, and extensive research on these two taxa by G. Morgan, S. Lucas, and others has resulted in the conclusion that they do not temporally overlap; that Rhynchotherium went extinct around 2 my ago and that Cuvieronius first appeared about 1.5 my ago (Morgan et al., 2016). This, therefore, provides for our assignment of the specimens from the Walrus Ditch LF to R. falconeri. It should be noted, how- ever, that there are several gomphothere teeth in the collections of the SCSM, the Charleston Museum, and the Mace Brown Museum of Natural History, as well as in known private collections, for which detailed identification is not available due to their highly variable morphology and lack of detailed provenance data. SUMMARY AND CONCLUSIONS Approximately 107 species of terrestrial fossil mammalian taxa and approximately 56 marine mammalian taxa (not including several undescribed species) are known from at least 18 of 41 named formations (and at least two unnamed units) under- lying the SC Coastal Plain. Paleocene and Eocene terrestrial vertebrates are exceptionally rare, as they are only recovered from quarries or construction sites where deep excavation reached strata of that age. One of these, the viverravid Didymictis proteus from the approximately 57 my old (late Clarkfork- ian/earliest Wasatchian) Williamsburg Formation, provides a new first occurrence for the Atlantic Coastal Plain. Two others, a brontothere molar and rhinoceros palate with teeth (Subhyracodon mitis) from the approximately 34 my old Harleyville For- mation (latest Chadronian) also deserve mention because they too represent new Eocene records for SC, but sadly they are currently in a private collec- tion and inaccessible to study. Fortunately, a partial tooth from the rhinoceros specimen is housed at the SCSM and is formally recognized in this report. A third specimen that deserves mention is a metatar- sal of the hyena Chasmaporthetes. Although the first author was allowed to examine, measure, and photograph the specimen (a left Mt III), it is not included in the “Systematic Paleontology” section of this work because it too resides in the same pri- vate collection. Collected from the Ashepoo River by a scuba diver, the exact location and unit of ori- gin are unknown. This specimen represents the first eastern occurrence of this taxon outside of Florida. Additional first occurrences for SC reported herein include Ondatra idahoensis, Canis lepophagus, Phocanella pumila, Callophoca obscura, Anchip- pus texanus, and possibly Teleoceras guymonense. Other notable occurrences include what may be the oldest records of capybara and Allophaiomys in the USA and the oldest record of Erethizon in the southeastern USA. The latter two specimens were recovered from the approximately 2.3 my old, late Blancan, Walrus Ditch LF. Particularly impressive is the stunning diver- sity of fossil cetaceans from the state, which helps to document their evolution from rear-legged pro- tocetids with the nasal opening still far forward on the snout, through their cladogenesis into primi- tive odontocetes and bizarre toothed mysticetes, to the earliest baleen-bearing mysticetes and echo- locating odontocetes. Indeed, SC’s fossil cetacean record rivals the very best on the planet, including those of western Pakistan and the Fayum of Egypt. With its wealth of paleontological resources, in particular those from the famous “Ashley River phosphate beds,” the SC Coastal Plain played a seminal role in the development of vertebrate pale- ontology as a scientific discipline in the United ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 195 States. Were it not for phosphate and limestone mining operations, however, this potential may not have been realized as early as it was, considering that natural exposures of fossil-bearing strata are exceptionally rare in the Carolina “Low Country.” But now, many decades later, explosive escala- tion in development and construction, particularly in the Charleston-Berkeley-Dorchester tri-county area, has resulted in at least ephemeral exposure of fossil-bearing beds that lie in some cases just below the surface. The consequent dramatic increase in avocational fossil collecting has often led to impor- tant specimens remaining in private hands, but for- tunately many are now housed within museum col- lections where they can be accessed in perpetuity for study. The discovery in the late 1960s and early 1970s that the bottoms of many South Carolina coastal rivers were littered with fossils eventually led to the inclusion of fossils in the SC Underwa- ter Antiquities Act (1991). This Act identifies fos- sils occurring in coastal waterways as State prop- erty, but also allows for avocational collectors to recover and in most cases keep the material they find. Such specimens that have been donated to and placed within museum collections, many of which are included in this report, have provided an even greater understanding of the state’s record of pre- historic life. With progressively increasing study of these new finds over the last few decades came the real- ization that the weak link in a thorough understand- ing of the state’s vertebrate paleontology was the lack of a refined, modern, temporal framework. Our attempt in this report has been to provide, in a single body of literature, (1) a statewide review of all the vertebrate fossil-bearing geologic units, (2) an update and refinement of their temporal place- ment with the consequent development of a refined chronostratigraphic context for the state’s verte- brate fossil record, (3) a comprehensive review of that record, including several additions previously unknown from the state (or in some cases from the Southeastern Atlantic Coastal Plain), plus (4) revised taxonomic assessment of some previously reported species. We have provided this interpretation in the hope that our initial attempt will aid other research- ers interested in the vertebrate paleontology of South Carolina, and inspire at least occasional updates as the GPTS and understanding of the state’s stratigraphy further refines our knowledge of the age of the state’s vertebrate fossil-bearing units. As less than half of the recognized units within the South Carolina Coastal Plain are known to contain vertebrate fossils, we fully recognize that our effort may be altered dramatically as addi- tional field and laboratory work results in the dis- covery of new fossil-bearing localities and analy- sis of new specimens, respectively. Publication of data on non-mammalian fossils will also add to the robustness of the findings presented here. ACKNOWLEDGEMENTS First, we extend our gratitude to the following avo- cational fossil collectors: Aura Baker† (and vol- unteers from the Myrtle Beach Fossil Club who helped her), Mace Brown (Director, Mace Brown Museum of Natural History, College of Charles- ton), Michael Bruggeman, Alan Devier, Doris Holt†, Derwin Hudson, Jeremy Jacobs, Daniel Lesesne, James Malcolm, Vance McCollum, John Metts†, Frank Morning, Ray Ogilvie, Bruce Orr, Bill Palmer†, Richard Patterson, Sammy Peek, Alan Schoemaker, Matthew Swilp, Jason Thomp- son, and Susan Wallace (extensive discussions with V. McCollum and M. Swilp were particularly helpful); plus Jason Osborne and Aaron Alford, founders and principals of Paleo Quest, a non- profit (503c) organization established to enhance paleontological collections, research, and educa- tion through “citizen science.” It is only through the enthusiastic and professional efforts of all the above, and their understanding of the importance of having fossils curated in public repositories, that scientific study of many of South Carolina’s fossil vertebrates has been possible. († = Deceased) Additional gratitude is extended to the fol- lowing colleagues who took time from their busy schedules to identify molluscs and/or microfossils from samples of stratigraphic units for which we sought refined age determinations, and for help- ful discussions on SC stratigraphy: Lyle Camp- bell (University of South Carolina, Spartanburg), 196 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) David Campbell (University of North Carolina, Chapel Hill), Matthew Campbell (Erskine Col- lege, Due West, SC), William Doar (SC Depart- ment of Natural Resources), W. B. Harris (Univer- sity of North Carolina, Wilmington), Tom Cronin (USGS), and especially Jean Self-Trail and Lucy Edwards (USGS) who we bombarded with sedi- ment samples for which we needed microfossil analyses. Matthew Campbell is also thanked for allowing us to include his unpublished lists of mol- luscs from the Ashley and Chandler Bridge forma- tions that have resulted from his extensive studies of these units. Special gratitude is extended to Matthew Gibson for assisting LBA and AES with study of specimens under his care at The Charleston Museum, and to Richard Hulbert for similar assis- tance during visits by LBA and JLK to the Florida Museum of Natural History. Jonathan Bloch, Paul Morse, and David Steadman were also integral to our studies of specimens at the latter institu- tion. We also thank David Bohaska, Fred Grady, Robert Purdy, and Steve Gotte at the Smithsonian Institution, Washington, DC, and Stephanie Pierce and Jessica Cundiff at the Museum of Compara- tive Zoology, Harvard. Gary Morgan at the New Mexico Museum of Natural History and Science is thanked for helpful discussions on fossils and Neo- gene sites of Florida and also for help with identi- fication of some specimens. Matthew Mihlbachler provided helpful discussions on North American brontotheres and Jonathan Geisler was always generous in sharing his knowledge of SC’s fossil cetacean record. Charles Ciampaglio (Wright State University, Celina, Ohio) introduced LBA to the Clapp Creek locality in Kingstree. Adam Rountrey (University of Michigan Museum of Paleontology) provided photographs of the genoholotype of Pro- neofiber guildayi, as did Ted Daeschler (ANSP) for Phoca modesta and P. debilis, and Stephen God- frey (Calvert Marine Museum) for Agorophius pygmaeus. Finally, Nicholas Emord, Joshua Nevin, and Joshua Oglesby of Johnson and Johnson’s 3-D Printing Laboratory at UNF are thanked for the time they took from their busy schedules to CT scan some of the specimens. Michael Wood- burne is thanked for reviewing an early version of the manuscript, as are formal reviewers Jonathan Geisler, Richard Hulbert, and Gary Morgan, whose attention to details plus comments and suggestions resulted in a much improved final product. Funding for this project was provided in part by a grant to LBA from the Charleston Scientific and Cultural Education Fund. Finally, LBA would like to extend a personal debt of gratitude to his friend of over 55 years, Howard G. Costa. It was their childhood fos- sil forays around the Charleston area and the deep interest in natural history they shared during their formative years that in large part galvanized the first author’s early decision to become a scientist, with particular interests in paleontology. Both of us were in very large part inspired by our early asso- ciation with Albert E. Sanders. LITERATURE CITED Abbott, W. H., and G. W. Andrews. 1979. Middle Miocene marine diatoms from the Hawthorn Formation within Ridgeland Trough, South Carolina and Georgia. Micropaleontology 25:225–271. Agassiz, L. 1848. Dorudon serratus, and Sauroce- tus gibbesii. Proceedings of the Academy of Natural Sciences of Philadelphia 4:4–5. Ahearn, M. E. 1981. A revision of the North Amer- ican Hydrochoeridae. M. S. thesis, University of Florida, Gainesville, Florida. 99 p. Ahearn, M. E., and J. F. Lance. 1980. A new spe- cies of Neochoerus (Rodentia: Hydrochoeri- dae) from the Blancan (late Pliocene) of North America. Proceedings of the Biological Soci- ety of Washington 93:435–442. Akers, W. H. 1972. Planktonic foraminifera and biostratigraphy of some Neogene formations, northern Florida and the Atlantic Coastal Plain. Tulane Studies in Geology and Paleon- tology 9, 139 p. Albright, L. B., III. 1991. The vertebrate paleontol- ogy, taphonomy, and paleoecology of a new early Miocene site in the Fleming Formation near Toledo Bend Dam, Newton County, Tex- as, and its importance to Gulf Coastal Plain biostratigraphy. M. S. thesis. Louisiana State University, Baton Rouge. 319 p. Albright, L. B., III. 1996. A protocetid cetacean ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 197 from the Eocene of South Carolina. Journal of Paleontology 70:519–523. Albright, L. B., III. 1998. The Arikareean Land Mammal Age in Texas and Florida: southern extension of Great Plains faunas and Gulf Coastal Plain endemism. Pp. 160–183 in D. O. Terry, H. E. LaGarry, and R. M. Hunt, eds. Depositional Environments, Lithostratigra- phy, and Biostratigraphy of the White River and Arikaree Groups (Late Eocene to Early Miocene, North America). Geological Society of America Special Paper 325. Albright, L. B., III. 1999a. Biostratigraphy and ver- tebrate paleontology of the San Timoteo Bad- lands, southern California. University of Cali- fornia Publications in the Geological Sciences 144:1–139. Albright, L. B., III. 1999b. Ungulates of the To- ledo Bend Local Fauna (late Arikareean, early Miocene), Texas Coastal Plain. Bulletin of the Florida Museum of Natural History 42:1–80. Albright, L. B., III, A. E. Sanders, and J. H. Geisler. 2018. An unexpectedly derived odontocete from the Ashley Formation (up- per Rupelian) of South Carolina, U.S.A. Journal of Vertebrate Paleontology, DOI: 1080/02724634.2018.1482555. Albright, L. B., III, J. E. Starnes, and G. E. Phillips. 2016. The Jones Branch Local Fauna: an early Arikareean mammalian assemblage from the late Oligocene marginal marine basal Cata- houla Clay, Wayne County, Mississippi. Geo- logical Society of America Program, South- Central Section Meeting 21–22 March 2016. Volume 48, no.1, Abstract 11–4. Albright, L. B., III, M. O. Woodburne, T. J. Fremd, C. C. Swisher III, B. J. MacFadden, and G. R. Scott. 2008. Revised chronostratigraphy and biostratigraphy of the John Day Forma- tion (Turtle Cove and Kimberly members), Oregon, with implications for updated cali- bration of the Arikareean North American Land Mammal Age. The Journal of Geology 116:211–237. Allen, G. M. 1926. Fossil mammals from South Carolina. Bulletin of the Museum of Compar- ative Zoology 67:445–467. Anderson, A. J. 1961. Gliederung und paläogeog- raphische Entwicklung der Chattischen Stufe (Oberoligozän) im Nordseebecken. Meyniana 10:118–146. Anderson, A. J., W. Hinsch, E. Martini, C. Müller, and S. Ritzkowski. 1971. Chattian. Giornale di Geologia 37:69–79. Anthonissen, E. D. 2008. Late Pliocene and Pleis- tocene biostratigraphy of the Nordic Atlantic region. Newsletters on Stratigraphy 43:33–48. Appleby, C., W. B. Harris, G. Dietl, P. H. Kelly, K. A. Badyrka, and E. A. Graybill. 2010. North and South Carolina Plio-Pleistocene strontium isotope geochronology. Geological Society of America, Abstracts with Programs 42:137. Archibald, J. D., W. A. Clemens, P. D. Gingerich, D. W. Krause, E. H. Lindsay, and K. D. Rose. 1987. First North American land mammal ages of the Cenozoic Era. Pp. 24–76 in M. O. Woodburne, ed. Cenozoic Mammals of North America. University of California Press, Berkeley. Babiarz, J. P., H. T. Wheeler, J. L. Knight, and L. D. Martin. 2018. Smilodon from South Caro- lina: implications for the taxonomy of the ge- nus. Pp. 76–107 in L. Werdelin, H. G. McDon- ald, and C. A. Shaw, eds. Smilodon, the Iconic Sabertooth. Johns Hopkins University Press, Baltimore. Badyrka, K. A., P. H. Kelley, W. B. Harris, G. P. Dietl, and C. C. Visaggi. 2010. Paleontolog- ic and strontium isotopic age analysis of the Waccamaw Formation, Waccamaw River and Intracoastal Waterway, Horry County, South Carolina. Geological Society of America, Ab- stracts with Programs 42:187. Banks, R. S. 1977. Stratigraphy of the Eocene San- tee Limestone in three quarries of the Coastal Plain of South Carolina. South Carolina Geo- logical Survey Geological Notes 21:85–149. Barnes, L. G., and A. E. Sanders. 1996. Transition from archaeocetes to mysticetes: Late Oligo- cene toothed mysticetes from near Charleston, South Carolina. P. 24 in J. E. Repetski, ed. Ab- stracts of Papers, Sixth North American Pale- ontological Convention, Smithsonian Institu- tion, Washington D.C., June 9–12, 1996. The 198 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) Paleontological Society Special Publication No. 8. Barnosky, A. D. 1981. A skeleton of Mesoscalops (Mammalia: Insectivora) from the Miocene Deep River Formation, Montana, and a review of proscalopid moles: evolutionary, function- al, and stratigraphic relationships. Journal of Vertebrate Paleontology 1:285–339. Barnosky, A. D. 1982. A new species of Proscalops (Mammalia: Insectivora) from the Arikareean Deep River Formation, Meagher County, Mon- tana. Journal of Paleontology 56:1103–1111. Baskin, J. A. 1991. Early Pliocene horses from late Pleistocene fluvial deposits, Gulf Coastal Plain, south Texas. Journal of Paleontology 65:995–1006. Baum, G. R., J. S. Collins, R. M. Jones, B. R. Madlinger, and R. J. Powell. 1980. Correlation of the Eocene strata of the Carolinas. South Carolina Geology 24:19–27. Beard, K. C., and M. R. Dawson. 2001. Early Wasatchian mammals from the Gulf coastal plain of Mississippi: biostratigraphic and pale- obiogeographic implications. Pp. 75–94 in G. F. Gunnell, ed. Eocene Biodiversity: Unusual Occurrences and Rarely Sampled Habitats. Kluwer, New York. Beard, K. C., and M. R. Dawson. 2009. Early Wasatchian mammals of the Red Hot Lo- cal Fauna, uppermost Tuscahoma Formation, Lauderdale County, Mississippi. Annals of Carnegie Museum 78:193–243. Beaty, M. C., A. D. Cohen, and R. H. Willoughby. 2007. Paleoecology and depositional environ- ments of a middle Pleistocene (?) vertebrate fossil locality in Harleyville, SC. Geological Society of America Abstracts with Programs 39:96. Beatty, B. L., and J. Geisler. 2010. A stratigraphi- cally precise record of Protosiren (Protosire- nidae, Sirenia) from North America. Neues Jahrbuch für Geologie und Paleontologie. Ab- handlungen 258:185–194. Bell, C. J., E. L. Lundelius, Jr., A. D. Barnosky, R. W. Graham, E. H. Lindsay, De. R. Ruez, Jr., H. A. Semken, Jr., S. D. Webb, and R. J. Zakrze- wski. 2004. The Blancan, Irvingtonian, and Rancholabrean Mammal Ages. Pp. 232–314 in M. O. Woodburne, ed. Late Cretaceous and Cenozoic Mammals of North America, Bio- stratigraphy and Geochronology. Columbia University Press, New York. Bentley, C. C., J. L. Knight, and M. A. Knoll. 1994. The mammals of the Ardis Local Fauna (late Pleistocene), Harleyville, South Carolina. Brimleyana 21:1–35. Berggren, W. A. 1973. The Pliocene time scale: calibration of planktonic foraminifera and calcareous nannoplankton zones. Nature 243:391–397. Berggren, W. A., D. V. Kent, C. C. Swisher, III, and M-P. Aubry. 1995. A revised Cenozoic geochronology and chronostratigraphy. Pp. 129–212 in W. A. Berggren, D. V. Kent, and J. Hardenbol, eds. Geochronology, Time Scales and Global Stratigraphic Correlation. Society for Sedimentary Geology Special Publication No. 54. Bianucci, G. 1997. The Odontoceti (Mammalia Cetacea) from the Italian Pliocene. The Ziphi- idae. Palaeontographia Italica 84:163–192. Bishop, G. A., and B. T. Palmer. 2006. A new ge- nus and species of crab from the bryozoan bio- herms of the Eocene Santee Limestone, South Carolina, USA. Revista Mexicana de Ciencias Geologicas 23:334–337. Blackwelder, B. W. 1979. Stratigraphic revision of lower Pleistocene marine deposits of North and South Carolina. U.S. Geological Survey Bulletin 1482-A:52–61. Blackwelder, B. W. 1981. Late Cenozoic stages and molluscan zones of the United States, middle Atlantic Coastal Plain. Paleontological Society Memoir 12, 34 p. Blackwelder, B. W., and L. W. Ward. 1979. Strati- graphic revision of the Pliocene deposits of North and South Carolina. South Carolina Di- vision of Geology, Geologic Notes 23:33–49. Blow, W. C., and R. B. Manning. 1996. Prelimi- nary descriptions of 25 new decapod crusta- ceans from the middle Eocene of the Caroli- nas, U.S.A. Tulane Studies in Geology and Paleontology 29:1–26. Blow, W. H. 1969. Late middle Eocene to Recent ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 199 planktonic foraminiferal biostratigraphy. Pp. 199–422 in P. Brönnimann and H. H. Renz, eds. Proceedings of the 1st International Con- ference on Planktonic Microfossils, Geneva 1967, vol. 1. E. J. Brill, Leiden, Netherlands. Boessenecker, R. W., and J. H. Geisler. 2018. New records of the archaic dolphin Agorophius (Mammalia: Cetacea) from the upper Oligo- cene Chandler Bridge Formation of South Carolina, USA. PeerJ 6:e5290; DOI 10.7717/ peerj.5290. Boessenecker, R. W., E. Ahmed, and J. H. Geisler. 2017a. New records of the dolphin Albertoce- tus meffordorum (Odontoceti: Xenorophidae) from the lower Oligocene of South Carolina: encephalization, sensory anatomy, postcranial morphology, and ontogeny of early odontoce- tes. PloS ONE 12:e0186476; DOI 10.11371/ journal.pone.0186476. Boessenecker, R. W., D. Fraser, M. Churchill, and J. H. Geisler. 2017b. A toothless dwarf dolphin (Odontoceti: Xenorophidae) points to explo- sive feeding diversification of modern whales (Neoceti). Proceedings of the Royal Society B 284:20170531. Boessenecker, R. W., M. Churchill, E. Buchholt- zand, B. L. Beatty, and J. H. Geisler. 2018. New fossils from the Oligocene of South Carolina clarify the identity of “Squalodon” ti- edemani and the postcranial evolution of early odontocetes. Meeting Program and Abstracts, Society of Vertebrate Paleontology 78th An- nual Meeting, p. 93. Boessenecker, R. W., D. J. Ehret, D. J. Long, M. Churchill, E. Martin, and S. J. Boessenecker. 2019. The early Pliocene extinction of the me- ga-toothed shark Otodus megalodon: a view from the eastern North Pacific. PeerJ 7:e6088 DOI 10.7717/peerj.6088. Boessenecker, S. J., R. W. Boessenecker, and J. H. Geisler. 2018. Youngest record of the ex- tinct walrus Ontocetus emmonsi from the early Pleistocene of South Carolina and a review of North Atlantic walrus biochronology. Acta Pa- laeontologica Polonica 63:279–286. Brainard, R., P. Smith, and L. Drolet. 2009. Daniel Island surprise — sand lens lurking in Cooper marl, Charleston, SC. Pp. 1–18 in G. Almeris and B. Mariucci, eds. Rapid Excavation and Tunneling Conference Proceedings 2009. So- ciety for Mining, Metallurgy, and Exploration. Brunet, M. 1979. Les grands mammifères chefs de file de l’immigration Oligocene et le problème de la limite Éocène-Oligocène en Europe. Édi- tions de la Foundation Singer-Polignac, Paris, 281 p. Bybell, L. M. 1990. Calcareous nannoplankton from Pliocene and Pleistocene deposits in South Carolina. Pp. B1-B9 in Studies Related to the Charleston, SC, Earthquake of 1886 – Neogene and Quaternary Lithostratigraphy and Biostratigraphy. U.S. Geological Survey Professional Paper 1367–A. Bybell, L. M., K. J. Conlon, L. E. Edwards, N. O. Frederiksen, G. S. Gohn, and J. M. Self-Trail. 1998. Biostratigraphy and physical stratigra- phy of the USGS-Cannon Park core (CHN– 800), Charleston County, South Carolina. U.S. Geological Survey Open-file Report 98–246, 65 p. Campbell, D. C. 1995. New molluscan faunas from the Eocene of South Carolina. Tulane Studies in Geology and Paleontology 27:119–152. Campbell, D. C., and M. R. Campbell. 2003. Bi- otic patterns in Eocene-Oligocene mollusks of the Carolinas. Pp. 341–353 in D. R. Prothero, L. C. Ivany, and E. Nesbitt, eds. From Green- house to Icehouse: The Marine Eocene-Oligo- cene Transition. Columbia University Press, New York. Campbell, L. D. 1993. Pliocene molluscs from the Yorktown and Chowan River formations in Virginia. Virginia Division of Mineral Re- sources Publication 127:1–259. Campbell, M. R., and L. D. Campbell. 1995. Pre- liminary biostratigraphy and molluscan fauna of the Goose Creek Limestone of eastern South Carolina. Tulane Studies in Geology and Pale- ontology 27:53–100. Candy, I., D. C. Schreve, J. Sherriff, and G. J. Tye. 2014. Marine Isotope Stage 11: palaeocli- mates, palaeoenvironments and its role as an analogue for the current interglacial. Earth- Science Reviews 128:18–51. 200 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) Carranza-Castañeda, O., and W. E. Miller. 1988. Roedores caviomorfos de la Mesa Central de México, Blancano temprano (Plioceno tardío) de la Fauna Local Rancho Viejo, Estado de Guanajuato. Universidad Nacional Autónoma de México, Instituto de Geologia. Revista 7:182–199. Case, G. B. 1994. Fish remains from the Paleocene and early Eocene Bashi Formation of Merid- ian, Lauderdale County, Mississippi, Part II— Teleosteans. Palaeontographica Abteilung A 230:139–153. Case, G. B., T. D. Cook, and M. V. H. Wilson. 2011. A new genus and species of fossil mylioba- toid ray from the Fishburne Formation (lower Eocene/Ypresian) of Berkeley County, South Carolina. Historical Biology 23:139–144. Chandler, R. M., and J. L. Knight. 2009. Fossil birds of the Crowfield Local Fauna, late Pleis- tocene (Rancholabrean), South Carolina. Cur- rent Research in the Pleistocene 26:143–144. Churchill, M., J. H. Geisler, B. L. Beatty, and Anjali Goswami. 2018. Evolution of cranial telescop- ing in echolocating whales (Cetacea: Odonto- ceti). Evolution 75:1092-1108. Churchill, M., M. Martinez-Caceres, C. de Muizon, J. Mnieckowski, and J. H. Geisler. 2016. The origin of high-frequency hearing in whales. Current Biology 26:1–6. Cicimurri, D. J. 2010. On the dentition of Meridi- ania convexa Case (Myliobatoidea), an extinct early Eocene ray from the United States. Cai- nozoic Research 7:99–107. Cicimurri, D. J., and J. L. Knight. 2009a. New re- cord of an extinct fish, Fisherichthyes folmeri Weems (Osteichthyes), from the lower Eocene of Berkeley County, South Carolina, USA. Pa- leoBios 29:24–28. Cicimurri, D. J., and J. L. Knight. 2009b. Late Oligo- cene sharks and rays from the Chandler Bridge Formation, Dorchester County, South Carolina, USA. Acta Palaeontologica 54:627–647. Cicimurri, D. J., and J. L. Knight. 2009c. Two shark-bitten whale skeletons from coastal plain deposits of South Carolina. Southeastern Naturalist 8:71–82. Cicimurri, D. J., and J. L. Knight. 2019. Late Eo- cene (Priabonian) elasmobranchs from the Dry Branch Formation (Barnwell Group) of Aiken County, South Carolina, USA. Paleo- Bios 36:1-31. Cicimurri, D. J., J. L. Knight, and J. M. Self-Trail, and S. M. Ebersole. 2016. Late Paleocene glyptosaur (Reptilia: Anguidae) osteoderms from South Carolina, USA. Journal of Paleon- tology 90:147–153. Coccioni, R., A. Marsili, A. Montanari, A. Bel- lanca, R. Neri, D. M. Bice, H. Brinkhuis, N. Church, A. Macalady, A. McDaniel, A. Deino, F. Lirer, M. Sprovieri, P. Maiorano, S. Mone- chi, C. Nini, M. Nocchi, J. Pross, P. Rochette, L. Sagnotti, F. Tateo, Y. Touchard, S. Van Si- maeys, and G. L. Williams. 2008. Integrated stratigraphy of the Oligocene pelagic sequence in the Umbria-Marche basin (northeastern Apennines, Italy): a potential Global Strato- type Section and Point (GSSP) for the Rupe- lian/Chattian boundary. Geological Society of America Bulletin 120:487–511. Colquhoun, D. 1974. Cyclic surficial stratigraphic units of the middle and lower Coastal Plains, central South Carolina. Pp. 179–190 in R. Q. Oaks, Jr., and J. R. DuBar, eds. Post-Miocene Stratigraphy, Central and Southern Atlantic Coastal Plain. Utah State University Press, Logan. Colquhoun, D., A. Cohen, M. Katuna, J. Rine, M. Segall, and M. Waddell. 1993. Petrology, dep- ositional setting, stratigraphy, structure and age of the Upland Unit along the southeastern Atlantic coast regional cross-section (DNAG E-5 Corridor) and near Charleston, South Car- olina. Pp. 97–102 in V. A. Zullo, W. B. Har- ris, T. M. Scott, and R. W. Portell, eds. The Neogene of Florida and Adjacent Regions: Proceedings of the Third Bald Head Island Conference on Coastal Plain Geology. Florida Geological Survey Special Publication 37. Cooke, C. W. 1925. The coastal plain. Pp. 19–54 in L. LaForge, C. W. Cooke, A. Keith, and M. R. Campbell, eds. Physical Geography of Geor- gia. Georgia Geologic Survey Bulletin 42. Cooke, C. W. 1936. Geology of the Coastal Plain of South Carolina. U.S. Geological Survey ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 201 Bulletin 867:1–196. Cooke, C. W. 1943. Geology of the Coastal Plain of Georgia. U.S. Geological Survey Bulletin 941:1–121. Cooke, C. W. 1945. Geology of Florida. Florida Geological Survey Bulletin 29, 339 p. Cooke, C. W., and F. S. MacNeil. 1952. Tertiary stratigraphy of South Carolina. U.S. Geologi- cal Survey Professional Paper 243-B:19–29. Cope, E. D. 1867. An addition to the vertebrate fauna of the Miocene Period, with a synopsis of the extinct Cetacea of the United States. Proceedings of the Academy of Natural Sci- ences of Philadelphia 19:138–156. Cope, E. D. 1868. Second contribution to the his- tory of the vertebrate fauna of the Miocene Pe- riod of the United States. Proceedings of the Academy of Natural Sciences of Philadelphia 20:184–194. Cope, E. D. 1883. On a new extinct genus of Sire- nia from South Carolina. Proceedings of the Academy of Natural Sciences of Philadelphia 35:52–54. Cramer, B. S., J. R. Toggweiler, J. D. Wright, M. E. Katz, and K. G. Miller. 2009. Ocean over- turning since the Late Cretaceous: inferences from a new benthic foraminiferal isotope compilation. Paleoceanography 24:PA4216, doi:10.1029/2008PA001683. Cronin, T. M. 2009. Paleoclimates: Understanding Climate Change Past and Present. Columbia University Press, New York, 441 p. Cronin, T. M., P. R. Vogt, D. A. Willard, R. Tunell, J. Halka, M. Berke, and J. Pohlman. 2007. Rapid sea level rise and ice sheet response to 8200- year climate event. Geophysical Research Let- ters 34:L20603, doi:10.1029/2007GL031318. Cronin, T. M., L. M. Bybell, R. Z. Poore, B. W. Blackwelder, J. C. Liddicoat, and J. E. Hazel. 1984. Age and correlation of emerged Plio- cene and Pleistocene deposits, U.S. Atlantic Coastal Plain. Palaeogeography, Palaeoclima- tology, and Palaeoecology 47:21–51. Curry, D., C. G. Adams, M. C. Boulter, F. C. Dilley, F. E. Eames, B. M. Funnell, and M. K. Wells. 1978. A correlation of Tertiary rocks in the British Isles. Geological Society of London Special Report 12:1–72. Dall, W. H. 1890. Contributions to the Tertiary fauna of Florida, with especial reference to the Miocene Silex-beds of Tampa and the Plio- cene beds of the Caloosahatchie River, Part I: pulmonate, opisthobranchiate and orthodont gastropods. Transactions of the Wagner Free Institute of Science of Philadelphia 3:1–200. Dall, W. H. 1892. Contributions to the Tertiary fauna of Florida, with especial reference to the Miocene Silex-beds of Tampa and the Plio- cene beds of the Caloosahatchie River, Part II: streptodont and other gastropods, concluded. Transactions of the Wagner Free Institute of Science of Philadelphia 3:201–473. Dall, W. H. 1894. Notes on the Miocene and Plio- cene of Gay Head, Martha’s Vineyard, Mass., and on the “Land phosphate” of the Ashley River district, South Carolina. American Jour- nal of Science, 3rd series, vol. 48, no. 286, ar- ticle 42:296–301. Dall, W. H. 1896. On some new species of Scala. The Nautilus 9:111–112. Dall, W. H. 1903. Contributions to the Tertiary fau- na of Florida, with especial reference to the Si- lex beds of Tampa and the Pliocene beds of the Caloosahatchie River including in many cas- es a complete revision of the generic groups treated and their American Tertiary species, Part VI: concluding the work. Transactions of the Wagner Free Institute of Science of Phila- delphia 3:1219–1654. De Man, E., S. A. Simaeys, N. Vandenberghe, W. B. Harris, and J. M. Wampler. 2010. On the na- ture and stratigraphic position of the Rupelian and Chattian stratotypes in the southern North Sea basin. Episodes 33:3–14. Deméré, T. A. 1994. The Family Odobenidae: a phylogenetic analysis of fossil and living taxa. Pp. 99–123 in A. Berta and T. A. Deméré, eds. Contributions in Marine Mammal Paleontol- ogy Honoring Frank C. Whitmore, Jr. Pro- ceedings of the San Diego Society of Natural History 29. Deméré, T. A., A. Berta, and P. J. Adam. 2003. Pin- 202 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) nipedimorph evolutionary biogeography. Pp. 32–76 in L. J. Flynn, ed. Vertebrate Fossils and Their Context, Contributions in Honor of Richard H. Tedford. Bulletin of the American Museum of Natural History 279. De Muizon, C. 1988. Le polyphylétisme des Ac- rodelphidae, odontocètes longirostres du Miocène européen. Bulletin du Muséum Na- tional d’Histoire, Paris, séction C. Sciences de la Terre: Paléontologie, Géologie, et Minéralo- gie, série 4, 10(1):31–88. Deschamps, C. M., A. I. Olivares, E. C. Vieytes, and M. G. Vucetich. 2007. Ontogeny and diversity of the oldest capybaras (Rodentia: Hydrocho- eridae; late Miocene of Argentina). Journal of Vertebrate Paleontology 27:683–692. Deschamps, C. M., M. G. Vucetich, S. I. Montalvo, and M. A. Zárate. 2013. Capybaras (Rodentia, Hydrochoeridae, Hydrochoerinae) and their bearing in the calibration of the late Miocene- Pliocene sequences of South America. Journal of South American Earth Science 48:145–158. De Verteuil, L., and G. Norris. 1996. Miocene di- noflagellate stratigraphy and systematics of Maryland and Virginia. Micropaleontology 42, Supplement, 172 p. Doar, W. R., III, and C. G. St. C. Kendall. 2014. An analysis and comparison of observed Pleis- tocene South Carolina (USA) shoreline eleva- tions with predicted elevations derived from marine oxygen isotope stages. Quaternary Re- search 82:164–174. Dockery, D. T., III, and P. Lozouet. 2003. Mol- luscan faunas across the Eocene/Oligocene boundary in the North American Gulf Coastal Plain, with comparisons to those in the Eocene and Oligocene of France. Pp. 303–340 in D. R. Prothero, L. C. Ivany, and E. A. Nesbitt, eds. From Greenhouse to Icehouse: The Marine Eocene-Oligocene Transition. Columbia Uni- versity Press, New York. Domning, D. P. 1989a. Fossil sirenians from the Suwannee River, Florida and Georgia. Pp. 54– 60 in G. S. Morgan, ed. Miocene Paleontol- ogy and Stratigraphy of the Suwannee River Basin of North Florida and South Georgia. Southeastern Geological Society Guidebook 30. Tallahassee, Florida. Domning, D. P. 1989b. Fossil sirenia of the western Atlantic and Caribbean region. II. Dioplothe- rium manigaulti Cope, 1883. Journal of Verte- brate Paleontology 9:415–428. Domning, D. P. 1997. Fossil sirenia of the western Atlantic and Caribbean region.VI. Crenatosi- ren olseni (Reinhart, 1976). Journal of Verte- brate Paleontology 17:397–412. Domning, D. P., and B. L. Beatty. 2019. Fossil Si- renia of the West Atlantic and Caribbean re- gion. XII..Stegosiren macei, gen. et sp. nov. Journal of Vertebrate Paleontology DOI: 10.1080/02724634.2019.1650369 Domning, D. P., G. S. Morgan, and C. E. Ray. 1982. North American Eocene sea cows (Mammalia: Sirenia). Smithsonian Contributions to Paleo- biology 52:1–69. Dooley, A. C., Jr. 2003. A review of the eastern North American Squalodontidae (Mammalia: Cetacea). Jeffersonia 11:1–26. Dooley, A. C., Jr. 2005. A new species of Squa- lodon (Mammalia, Cetacea) from the middle Miocene of Virginia. Virginia Museum of Nat- ural History Memoirs 8:1–43. Downing, K. F., and R. S. White. 1995. The cingu- lates (Xenarthra) of the Leisey Shell Pit Local Fauna (Irvingtonian), Hillsborough County, Florida. Pp. 375–396 in R. C. Hulbert, G. S. Morgan, and S. D. Webb, eds. Paleontology and Geology of the Leisey Shell Pits, Early Pleistocene of Florida, Bulletin of the Florida Museum of Natural History 37, Part II. Dowsett, H. J. 2007. The PRISM paleoclimate reconstruction and Pliocene sea-surface tem- perature. Pp. 459–480 in M. Williams, A. M. Haywood, J. Gregory, and D. N. Schmidt, eds. Deep-time Perspectives on Climate Change: Marrying the Signal from computer Models and Biological Proxies. Micropaleontological Society Special Publication, Geological Soci- ety of London, UK. Dowsett, H. J., and T. M. Cronin. 1990. High eu- static sea level during the middle Pliocene: evidence from the southeastern U.S. Atlantic Coastal Plain. Geology 18:435–438. Dowsett, H. J., M. A. Chandler, and M. M. Robin- ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 203 son. 2009. Surface temperatures of the Mid- Pliocene North Atlantic Ocean: implications for future climate. Philosophical Transactions of the Royal Society A 367:69–84. Dowsett, H. J., M. A. Chandler, T. M. Cronin, and G. S. Dwyer. 2005. Middle Pliocene sea sur- face temperature variability. Paleoceanogra- phy 20:PA2014. Dowsett, H., M. Robinson, A. Haywood, U. Sal- zmann, D. Hill, L. Sohl, M. Chandler, M. Williams, K. Foley, and D. Stoll. 2010. The PRISM3D paleoenvironmental reconstruc- tion. Stratigraphy 7:123–139. Dowsett, H. J., A. M. Haywood, P. J. Valdes, M. M. Robinson, D. J. Lunt, D. J. Hill, D. K. Stoll, M. Chandler, and K. M. Foley. 2011. Sea surface temperatures of the mid-Piacenzian Warm Pe- riod: a comparison of PRISM3 and HadCM3. Palaeogeography, Palaeoclimatology, and Pal- aeoecology 309:83–91. Dowsett, H. J., A. Dolan, D. Rowley, R. Mou- cha, A. M. Forte, J. X. Mitrovica, M. Pound, U. Salzmann, M. Robinson, M. Chandler, K. Foley, and A. Haywood. 2016. The PRISM4 (mid-Piacenzian) paleoenvironmental recon- struction. Climate of the Past 12:1519–1538, doi:10.5194/cp-12-1519-2016. DuBar, J. R. 1971. Neogene stratigraphy of the lower Coastal Plain of the Carolinas: Myrtle Beach, S.C., Atlantic Coastal Plain Geological Association Field Conference 12, 128 p. DuBar, J. R. 1987. Geology of the Dongola 7.5 minute quadrangle, Horry and Marion coun- ties, South Carolina. South Carolina Geology 31:1–15. DuBar, J. R., and J. R. Solliday. 1963. Stratigraphy of the Neogene deposits, lower Neuse Estuary, North Carolina. Southeastern Geology 4:213– 233. DuBar, J. R., H. S. Johnson, B. Thom, and W. O. Hatchell. 1974. Neogene stratigraphy and morphology, south flank of the Cape Fear Arch, North and South Carolina. Pp. 139–173 in R. Q. Oaks, Jr., and J. R. DuBar, eds. Post- Miocene Stratigraphy, Central and Southern Atlantic Coastal Plain. Utah State University Press, Logan. DuBar, J. R., S. S. DuBar, L. M. Ward, B. W. Black- welder, W. H. Abbot, and P. F. Huddlestun. 1980. Cenozoic biostratigraphy of the outer coastal plain. Pp. 179–236 in R. W. Frey, ed. Excursions in Southeastern Geology. Geologi- cal Society of America 1980 Annual Meeting Field Trip Guidebook 1. Dubrovo, I. A., and A. E. Sanders. 2000. A new species of Patriocetus from the late Eocene of Kazakhstan. Journal of Vertebrate Paleontol- ogy 20:577–590. Dwyer, G. S., and M. A. Chandler. 2009. Mid- Pliocene sea level and continental ice volume based on coupled benthic Mg/Ca palaeotem- peratures and oxygen isotopes. Philosophical Transactions of the Royal Society A 367:157– 168. Dybkjaer, K., and S. Piasecki. 2008. A new Neo- gene biostratigraphy from Denmark. Geologi- cal Survey of Denmark and Greenland Bulle- tin 15:29–32. Edgar, K. M., P. A. Wilson, P. F. Sexton, S. J. Gibbs, A. P. Roberts, and R. D. Norris. 2010. New biostratigraphic, magnetostratigraphic and isotopic insights into the Middle Eocene Cli- matic Optimum in low latitudes. Palaeogeog- raphy, Palaeoclimatology, and Palaeoecology 297:670–682. Edmund, A. G. 1987. Evolution of the genus Holmesina (Pampatheriidae: Mammalia). Tex- as Memorial Museum Pearce-Sellards Series 45:1–20. Edwards, L. E. 1998. Dinoflagellates from the Black Mingo Group (Paleocene) of South Car- olina. Pp. 28–48 in A. E. Sanders, ed. Paleobi- ology of the Williamsburg Formation (Black Mingo Group; Paleocene) of South Carolina, U.S.A. Transactions of the American Philo- sophical Society 88. Edwards, L. E., L. M. Bybell, G. S. Gohn, and N. O. Fredericksen. 1997. Paleontology and Physical Stratigraphy of the USGS–Pregnall No. 1 core (DOR-208), Dorchester County, South Carolina. USGS Open-file Report 97– 145, 35 p. Edwards, L. E., N. O. Fredericksen, L. M. Bybell, T. G. Gibson, G. S. Gohn, D. Bukry, J. M. 204 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) Self-Trail, and R. J. Litwin. 2001. Overview of the biostratigraphy and paleoecology of sediments from five cores from Screven and Burke Counties, Georgia. Pp. B1–B19 in L. E. Edwards, ed. Geology and Paleontology of Five Cores from Screven and Burke Counties, Eastern Georgia. U.S. Geological Survey Pro- fessional Paper 1603. Edwards, L. E., G. S. Gohn, L. M. Bybell, P. G. Chirico, R. A. Christopher, N. O. Frederick- sen, D. C. Powell, J. M. Self-Trail, and R. E. Weems. 2000. Supplement to the preliminary stratigraphic database for subsurface sedi- ments of Dorchester County, South Carolina. U.S. Geological Survey Open-File Report 00–049–B. Edwards, L. E., G. S. Gohn, J. M. Self-Trail, D. C. Prowell, L. M. Bybell, L. P. Bardot, J. V. Firth, B. T. Huber, N. O. Fredericksen, and K. G. MacLeod. 1999. Physical stratigraphy, paleontology, and magnetostratigraphy of the USGS-Santee Coastal Reserve Core (CHN– 803), Charleston County, South Carolina. U.S. Geological Survey Open-File Report 99–0308, 66 p. Effinger, J. A. 1998. Entelodontidae. Pp. 375–380 in C. M. Janis, K. M. Scott, and L. L. Jacobs, eds. Evolution of Tertiary Mammals of North America, Volume 1: Terrestrial Carnivores, Ungulates, and Ungulatelike Mammals. Cam- bridge University Press, Cambridge, United Kingdom. Emslie, S. D. 1995. The fossil record of Arctodus pristinus (Ursidae: Tremarctinae) in Florida. Pp. 501–514 in R. C. Hulbert, G. S. Morgan, and S. D. Webb, eds. Paleontology and Geol- ogy of the Leisey Shell Pits, Early Pleistocene of Florida. Bulletin of the Florida Museum of Natural History 37, Part II. Erickson, B. R. 1998a. Crocodilians of the Black Mingo Group (Paleocene) of the South Caro- lina Coastal Plain. Pp. 196–214 in A. E. Sand- ers, ed. Paleobiology of the Williamsburg For- mation (Black Mingo Group; Paleocene) of South Carolina, U.S.A. Transactions of the American Philosophical Society 88. Erickson, B. R. 1998b. A palaeophid snake from the late Paleocene of South Carolina. Pp. 215–220 in A. E. Sanders, ed. Paleobiology of the Williamsburg Formation (Black Mingo Group; Paleocene) of South Carolina, U.S.A. Transactions of the American Philosophical Society 88. Ernissee, J. J., W. H. Abbott, and P. F. Huddles- tun. 1977. Microfossil correlation of the Coo- sawhatchie Clay (Hawthorn Formation, Mio- cene) of South Carolina and its equivalent in Georgia. Marine Micropaleontology 2:105– 119. Eshelman, R. E. 1975. Geology and paleontology of the early Pleistocene (late Blancan) White Rock Fauna from north-central Kansas. C. W. Hibbard Memorial vol. 4, Papers on Paleontol- ogy, Museum of Paleontology, University of Michigan 13:1–60. Eshelman, R. E., and F. C. Whitmore Jr. 2008. Early Pliocene (late Hemphillian) land mam- mals from the Lee Creek Mine, Aurora, North Carolina. Pp. 17–38 in C. E. Ray, D. J. Bo- haska, I. A. Koretsky, L. W. Ward, and L. G. Barnes, eds. Geology and Paleontology of the Lee Creek Mine, North Carolina IV. Virginia Museum of Natural History Special Publica- tion 14. Evans, H. F., T. Westerhold, H. Paulsen, and J. E. T. Channell. 2007. Astronomical ages for Mio- cene polarity chrons C4Ar-C5r (9.3–11.2 Ma), and for three excursion chrons within C5n.2n. Earth and Planetary Science Letters 256:455– 465. Fallaw, W. C., and V. Price. 1995. Stratigraphy of the Savannah River Site and vicinity. South- eastern Geology 35:21–58. Fields, S. E. 2010. The ground sloth Megalonyx (Xenarthra: Megalonychidae) from the Pleis- tocene (late Irvingtonian) Camelot Local Fau- na, Dorchester County, South Carolina. Trans- actions of the American Philosophical Society, New Series 100:1–76. Fields, S. E., H. G. McDonald, J. L. Knight, and A. E. Sanders. 2012. The Ground Sloths (Pilosa) of South Carolina. PalArch’s Journal of Verte- ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 205 brate Paleontology 9:1–19. Fierstine, H. L., and R. E. Weems. 2009. Paleontol- ogy of the Ashley and Chandler Bridge forma- tions of South Carolina, 4: analysis and new records of billfishes (Perciformes: Xiphioidei). Paleo Ichthyologica 11:43–88. Flynn, J. J., B. J. Kowallis, C. Nuñez, O. Carranza- Casteñada, W. E. Miller, C. C. Swisher III, and E. Lindsay. 2005. Geochronology of Hemp- hillian-Blancan aged strata, Guanajuato, Mex- ico, and implications for timing of the Great American Biotic Interchange. The Journal of Geology 113:287–307. Fordyce, R. E. 1981. Systematics of the odontocete whale Agorophius pygmaeus and the family Agorophiidae (Mammalia: Cetacea). Journal of Paleontology 55:1028–1045. Fordyce, R. E. 1994. Waipatia maerewhenua, new genus and new species (Waipatiidae, new fam- ily), an archaic late Oligocene dolphin (Ceta- cea: Odontoceti: Platanistoidea) from New Zealand; pp. 147–176 in A. Berta and T. A. Deméré (eds.), Contributions in Marine Mam- mal Paleontology Honoring Frank C. Whit- more, Jr. Proceedings of the San Diego Soci- ety of Natural History 29. Fordyce, R. E., and C. De Muizon. 2001. Evolu- tionary history of the cetaceans: a review. Pp. 169–233 in J.-M. Mazin and V. de Buffrenil, eds. Secondary Adaptations of Tetrapods to Life in Water. Verlag Dr. Friedrich Pfeil, Mu- nich. Forsten, A. 1975. The fossil horses of the Texas Gulf Coastal Plain: a revision. Pearce-Sellards Series 22:1–86. Franţescu, A. L., R. M. Feldman, and C. E. Sch- weitzer. 2010. A new genus and species of dromiid crab (Decapoda, Brachyura) from the middle Eocene of South Carolina. Pp. 255– 267 in C. H. J. M. Fransen, S. De Grave, and P. K. L. Ng, eds. Studies on Malacostraca: Lipke Bijdeley Holthuis Memorial Volume. Crusta- ceana Monographs 14, 754 p. Franz, R. 2014. The fossil record of North Ameri- can tortoises. Pp. 13–24 in D. C. Rostal, E. D. McCoy, and H. R. Mushinsky, eds. Biology and Conservation of North American Tortois- es. The Johns Hopkins University Press, Balti- more, Maryland. Frazier, M. K. 1981. A revision of the fossil Er- ethizontidae of North America. Bulletin of the Florida Museum of Natural History 27:1–76. Galazzo, F. B., E. Thomas, M. Pagani, C. Warren, V. Luciani, and L. Giusberti. 2014. The middle Eocene climatic optimum (MECO): a multi- proxy record of paleoceanographic changes in the southeast Atlantic (ODP site 1263, Walvis Ridge). Paleoceanography 29:1143–1161. Galusha, T., N. M. Johnson, E. H. Lindsay, N. D. Opdyke, and R. H. Tedford. 1984. Biostratig- raphy and magnetostratigraphy, late Pliocene rocks, 111 Ranch, Arizona. Geological Society of American Bulletin95:714–722. Geisler, J. H., and A. E. Sanders. 2003. Morpho- logical evidence for the phylogeny of Cetacea. Journal of Mammalian Evolution 10:23–129. Geisler, J. H., B. L. Beatty, and R. W. Boesseneck- er. 2018. New specimens of Coronodon haven- steini provide insights into the transition from raptorial to filter feeding in whales. Meeting Program and Abstracts, Society of Vertebrate Paleontology 78th Annual Meeting, p. 133. Geisler, J. H., M. W. Colbert, and J. L. Carew. 2014. A new fossil species supports an early origin for toothed whale echolocation. Nature 508:383–386. Geisler, J. H., A. E. Sanders, and Z.-X. Luo. 2005. A new protocetid whale (Cetacea: Archaeoce- ti) from the late middle Eocene of South Caro- lina. American Museum Novitates 3480:1–65. Geisler, J. H., R. W. Boessenecker, M. Brown, and B. L. Beatty. 2017. The origin of filter feeding in whales. Current Biology 27:1–7. Geisler, J. H., M. R. McGowen, G. Ynag, and J. Gatesy. 2011. A supermatrix analysis of ge- nomic, morphological, and paleontological data from crown Cetacea. BMC Evolutionary Biology 11:1–33. Gibbard, P. L., M. J. Head, M. J. C. Walker, and the Subcommission on Quaternary Stratigraphy. 2010. Formal ratification of the Quaternary System/Period and the Pleistocene Series/Ep- och with a base at 2.58 Ma. Journal of Quater- nary Science 25:96–102. 206 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) Gibbes, R. W. 1845. Description of the teeth of a new fossil animal found in the green-sand of South Carolina. Proceedings of the Academy of Natural Sciences of Philadelphia 2:254– 256. Gibbes, R. W. 1851. A memoir on Mosasaurus and the three allied new genera, Holcodos, Cono- saurus, and Amophorosteus. Smithsonian Con- tributions to Knowledge 2 (article 5):1–13. Gibson, M. L., J. Mnieckowski, and J. H. Geisler. 2019. Tupelocetus palmeri, a new species of protocetid whale (Mammalia, Cetacea) from the middle Eocene of South Carolina. Journal of Vertebrate Paleontology. doi:10.1080/0272 4634.2018.1555165. Gibson, T. G., and L. M. Bybell. 1991. Paleocene- Eocene boundary sedimentation in the Po- tomac River Valley, Virginia and Maryland. IGCP Project 308, Fieldtrip Guidebook, 24 p. Gidley, J. W. 1907. Revision of the Miocene and Pliocene Equidae of North America. Bulletin of the American Museum of Natural History 23:865–934. Gillette, D. D. 1976. A new species of small cat from the late Quaternary of southeastern Unit- ed States. Journal of Mammalogy 57:664–676. Gillette, D. D., and C. E. Ray. 1981. Glyptodonts of North America. Smithsonian Contributions to Paleobiology 40, 255 p. Gillette, D. D., Ó. Carranza-Castañeda, R. S. White, G. S. Morgan, L. C. Thrasher, R. McCord, and G. McCullough. 2016. Ontogeny and sexual dimorphism of Glyptotherium texanum (Xen- arthra, Cingulata) from the Pliocene and Pleis- tocene (Blancan and Irvingtonian NALMA) of Arizona, New Mexico, and Mexico. Journal of Mammalian Evolution 23:133–154. Gingerich, P. D., M. Arif, M. A. Bhatti, M. An- war, and W. J. Sanders. 1997. Basilosaurus drazindai and Basilosaurus hussaini, new Archaeoceti (Mammalia, Cetacea) from the middle Eocene Drazinda Formation, with a re- vised interpretation of ages of whale-bearing strata in the Kirthar Group of the Sulaiman Range, Punjab (Pakistan). Contributions from the Museum of Paleontology, University of Michigan 30:55–81. Godfrey, S. J., M. D. Uhen, J. E. Osborne, and L. E. Edwards. 2016. A new specimen of Ago- rophius pygmaeus (Agorophiidae, Odontoce- ti, Cetacea) from the early Oligocene Ashley Formation of South Carolina, USA. Journal of Paleontology 90:154–169. Gohn, G. S. 1988. Late Mesozoic and early Ce- nozoic geology of the Atlantic Coastal Plain: North Carolina to Florida. Pp. 107–130 in R. E. Sheridan and J. A. Grow, eds. The Atlantic Continental Margin, U.S. Geological Society of America Decade of North American Geol- ogy, vol. 1–2, Boulder, Colorado. Gohn, G. S., J. E. Hazel, L. M. Bybell, and L. E. Edwards. 1983. The Fishburne Formation (lower Eocene), a newly defined subsurface unit in the South Carolina Coastal Plain. Pp. C1–C16 in Contributions to Stratigraphy, U.S. Geological Society Bulletin 1537–C. Gradstein, F. M., J. G. Ogg, M. D. Schmitz, and G. M. Ogg, eds. 2012. The Geologic Time Scale 2012. Elsevier, Amsterdam, 1144 p. Graybill, E. A., W. B. Harris, P. Kelley, G. Dietl, and C. C. Visaggi. 2009. Age of the Duplin and Waccamaw Formations, Cape Fear River basin, North Carolina. Geological Society of America, Abstracts with Programs 41:45. Gunnell, G. F., T. M. Bown, J. H. Hutchison, and J. I. Bloch. 2008. Lipotyphla. Pp. 89–125 in C. M. Janis, G. F. Gunnell, and M. D. Uhen, eds. Evolution of Tertiary Mammals of North America Volume 2: Small Mammals, Xen- arthrans, and Marine Mammals. Cambridge University Press, New York. Harlan, R. 1834. Notice of fossil bones found in the Tertiary formation of the state of Louisi- ana. Transactions of the American Philosophi- cal Society Philadelphia 4:397–403. Harris, W. B, and P. D. Fullagar. 1987. Rb-Sr radio- metric ages of glauconite from Eocene carbon- ate, Berkeley County, South Carolina. South Carolina Geology 31:51–57. Harris, W. B, and P. D. Fullagar. 1991. Middle Eo- cene and late Oligocene isotopic dates of glau- conitic mica from the Santee River area, South ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 207 Carolina. Southeastern Geology 32:1–19. Harris, W. B, and V. A. Zullo. 1991. Eocene and Oligocene stratigraphy of the Outer Coastal Plain. Pp. 251–262 in J. W. Horton, Jr., and V. A. Zullo, eds. The Geology of the Carolinas: Carolina Geological Survey Fiftieth Anniver- sary Volume. The University of Tennessee Press, Knoxville. Harrison, J. A. 1978. Mammals of the Wolf Ranch Local Fauna, Pliocene of the San Pedro Val- ley, Arizona. Occasional Papers of the Muse- um of Natural History, University of Kansas 73:1–18. Hay, O. P. 1923. The Pleistocene of North America and its vertebrated animals from the states east of the Mississippi River and from the Cana- dian Provinces east of longitude 95°. Carnegie Institution of Washington, Washington, D.C., 499 p. Haywood, A. M., H. J. Dowsett, P. J. Valdes, D. J. Lunt, J. E. Francis, and B. W. Sellwood. 2009. Introduction. Pliocene climate, processes and problems. Philosophical Transactions of the Royal Society A 367:3–17. Hazel, J. E., and nine others. 1977. Biostratigraphy of the deep core (Clubhouse Crossroads Core hole 1) near Charleston, South Carolina. Pp. 71–96 in D. W. Rankin, ed. Studies Related to the Charleston, South Carolina Earthquake of 1886: a Preliminary Report. U.S. Geological Survey Professional Paper 1028–E. Heron, S. D., Jr., G. C. Robinson, and H. S. John- son. 1965. Clays and opal-bearing claystones of the South Carolina Coastal Plain. S.C. State Development Board, Division of Geology, Bulletin 31:1–66. Hibbard, C. W. 1959. Late Cenozoic microtine ro- dents from Wyoming and Idaho. Papers of the Michigan Academy of Sciences, Arts, and Let- ters 44:1–40. Hibbard, C. W., and W. W. Dalquest. 1973. Proneo- fiber, a new genus of vole (Cricetidae: Roden- tia) from the Pleistocene Seymour Formation of Texas, and its evolutionary significance. Journal of Quaternary Research 3:269–274. Hilgen, F. J., L. J. Lourens, and J. A. Van Dam. 2012. The Neogene Period. Pp. 923–978 in F. M. Gradstein, J. G. Ogg, M. D. Schmitz, and G. M. Ogg, eds. The Geologic Time Scale 2012. Elsevier, Amsterdam. Hirschfeld, S. E., and D. S. Webb. 1968. Plio-Pleis- tocene megalonychid sloths of North America. Bulletin of the Florida State Museum 12:213– 296. Hocking, D. P., F. G. Marx, E. M. G. Fitzgerald, and A. R. Evans. 2017. Ancient whales did not filter feed with their teeth. Biology Letters 13: 20170348, doi:10.1098/rsbl.2017.0384. Holmes, F. S. 1860. Post-Pleiocene [sic] Fossils of South Carolina. Russell and Jones, Charleston, South Carolina, 122 p. Huddlestun, P. F. 1988. A revision of the lithostrati- graphic units of the Coastal Plain of Georgia, the Miocene through the Holocene. Georgia Geologic Survey Bulletin 104, 162 p. Hulbert, R. C., Jr. 1990. The taxonomic status of Hipparion minus Sellards, 1916 (Mammalia, Equidae). Journal of Paleontology 64:855– 856. Hulbert, R. C., Jr. 1993. Late Miocene Nannip- pus (Mammalia: Perissodactyla) from Florida, with a description of the smallest hipparionine horse. Journal of Vertebrate Paleontology 13:350–366. Hulbert, R. C., Jr. 1995. The giant tapir, Tapirus haysii, from Leisey Shell Pit 1A and other Florida Irvingtonian localities. Pp. 515–551 in R. C. Hulbert, G. S. Morgan, and S. D. Webb, eds. Paleontology and Geology of the Leisey Shell Pits, Early Pleistocene of Florida. Bul- letin of the Florida Museum of Natural History 37, Part II. Hulbert, R. C., Jr. 1997. A new late Pliocene porcu- pine (Rodentia: Erethizontidae) from Florida. Journal of Vertebrate Paleontology 17:623– 626. Hulbert, R. C., Jr. 2001. The Fossil Vertebrates of Florida. University Press of Florida, Gaines- ville, 350 p. Hulbert, R. C., Jr. 2010. A new early Pleistocene ta- pir (Mammalia: Perissodactyla) from Florida, with a review of Blancan tapirs from the state. 208 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) Florida Museum of Natural History Bulletin 49:67-126. Hulbert, R. C., Jr., and G. S. Morgan. 1993. Quan- titative and qualitative evolution in the giant armadillo Holmesina (Edentata: Pampatheri- idae) in Florida. Pp. 134–177 in R. A. Martin and A. D. Barnosky, eds. Morphologic Change in Quaternary Mammals of North America. Cambridge University Press, Cambridge. Hulbert, R. C., Jr., and A. E. Pratt. 1998. New Pleistocene (Rancholabrean) vertebrate faunas from coastal Georgia. Journal of Vertebrate Paleontology 18:412–429. Hulbert, R. C., Jr., A. Kerner, and G. S. Morgan. 2014. Taxonomy of the Pleistocene giant bea- ver Castoroides (Rodentia: Castoridae) from the Southeastern United States. Florida Mu- seum of Natural History Bulletin 53:26–43. Hulbert, R. C., Jr., R. M. Petkewich, G. A. Bishop, D. Bukry, and D. P. Aleshire. 1998. A new mid- dle Eocene protocetid whale (Mammalia: Ce- tacea: Archaeoceti) and associated biota from Georgia. Journal of Paleontology 72:907–926. Hutchison, J. H. 1972. Review of the Insectivore from the early Miocene Sharps Formation of South Dakota. Los Angeles County Natural History Museum, Contributions in Science 235:1–16. Hutchison, J. H. 1976. The Talpidae (Insectivora, Mammalia): evolution, phylogeny, and classi- fication. Ph.D dissertation. University of Cali- fornia, Berkeley. 36 p. Hutchison, J. H., and R. E. Weems. 1998. Paleo- cene turtle remains from South Carolina. Pp. 165–195 in A. E. Sanders, ed. Paleobiology of the Williamsburg Formation (Black Mingo Group; Paleocene) of South Carolina, U.S.A. Transactions of the American Philosophical Society 88. Idris, F. M., and V. J. Henry. 1995. Shallow Ceno- zoic seismic stratigraphy and structure: South Carolina lower coastal plain and continental shelf. Geological Society of America Bulletin 107:762–778. Jackson, D. R. 1978. Evolution and fossil record of the Chicken Turtle Deirochelys, with a re- evaluation of the genus. Tulane Studies in Zo- ology and Botany 20:35–55. Janis, C. M., G. F. Gunnell, and M. D. Uhen. eds. 2008. Appendix 1: Unified locality listing. Pp. 694–736 in C. M. Janis, G. F. Gunnell, and M. D. Uhen, eds. Evolution of Tertiary Mammals of North America Volume 2: Small Mammals, Xenarthrans, and Marine Mammals. Cam- bridge University Press, New York. Johnson, C. W. 1931. New fossil species of the ge- nus Epitonium from South Carolina. The Nau- tilus 45:6–10. Jones, D. S., P. A. Mueller, D. A. Hodell, and L. A. Stanley. 1993. 87Sr/86Sr geochronology of Oli- gocene and Miocene marine strata in Florida. Pp. 15–26 in V. A. Zullo, W. B. Harris, T. M. Scott, and R. W. Portell, eds. The Neogene of Florida and Adjacent Regions, Proceedings, 3rd Bald Head Island Conference on Coastal Plains Geology. Florida Geological Survey Special Publication 37. Katuna, M. P., J. H. Geisler, and D. J. Colquhoun. 1997. Stratigraphic correlation of Oligocene marginal marine and fluvial deposits across the middle and lower coastal plain, South Car- olina. Sedimentary Geology 108:181–194. Kelley, P. H., H. A. Baker, J. R. Depriest, D. S. Friend, J. H. Gagnon, E. S. Gould, B. M. McK- one, R. L. Salmon, and E. L. Scher. 2010a. Ecological structure of a molluscan assem- blage from the upper Waccamaw Formation, Shallotte, North Carolina, and the Pleistocene Extinction. Geological Society of America, Abstracts with Programs 42:256. Kelley, P. H., E. E. Biesack, B. M. Fetter, W. M. Keeler, J. M. Kendall, T. A. Moroney, R. H. Morrow, D. J. Morse, and G. P. Dietl. 2010b. A molluscan assemblage from the Waccamaw Formation, northeastern South Carolina: di- versity and stratigraphic implications. Geo- logical Society of America, Abstracts with Programs 42:187. Kellogg, R. 1923. Description of an apparently new toothed cetacean from South Carolina. Smith- sonian Contributions to Knowledge 76:1–7. Kellogg, R. 1923. Description of two squalodonts ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 209 recently discovered in the Calvert Cliffs, Maryland, and notes on the shark-toothed ce- taceans. Proceedings of the United States Na- tional Museum 62, art. 16:1–69. Kellogg, R. 1936. A review of the Archaeoceti. Carnegie Institution of Washington Publica- tion 482:1–366. Kidane, T., Y.-I. Otofuji, F. H. Brown, K. Takemoto, and G. Eshete. 2007. Two normal paleomag- netic polarity intervals in the lower Matuyama Chron recorded in the Shungura Formation (Omo Valley, southwest Ethiopia). Earth and Planetary Science Letters 262:240–256. Kier, P. M. 1980. The echinoids of the middle Eo- cene Warley Hill Formation, Santee Lime- stone, and Castle Hayne Limestone of North and South Carolina. Smithsonian Contribu- tions to Paleobiology 39:1–102. Kite, L. E. 1982. Tertiary stratigraphy of the Oak- wood quadrangle, Aiken County, South Caro- lina. Pp. 65–78 in P. G. Nystrom, Jr. and R. H. Willoughby, eds. Geological investigations related to the stratigraphy in the kaolin mining district, Aiken County, South Carolina. Caro- lina Geological Society, Field Trip Guidebook 1982. Knight, J. L., and D. J. Cicimurri. 2010. First report of fossil Amphiuma (Amphibia: Caudata: Am- phiumidae) from South Carolina, USA. Palu- dicola 8:1–7. Kohn, M. J., M. P. McKay, and J. L. Knight. 2005. Dining in the Pleistocene – who’s on the menu? Geology 33:649–652. Kohno, N., and C. E. Ray. 2008. Pliocene walrus- es from the Yorktown Formation of Virginia and North Carolina, and a systematic revi- sion of the North Atlantic Pliocene walruses. Pp. 39–80 in C. E. Ray, D. J. Bohaska, I. A. Koretsky, L. W. Ward, and L. G. Barnes, eds. Geology and Paleontology of the Lee Creek Mine, North Carolina IV. Virginia Museum of Natural History Special Publication 14. Koretsky, I. A., and L. G. Barnes. 2008. Phocidae. Pp. 542–556 in C. M. Janis, G. F. Gunnell, and M. D. Uhen, eds. Evolution of Tertiary Mam- mals of North America Volume 2: Small Mam- mals, Xenarthrans, and Marine Mammals. Cambridge University Press, New York. Koretsky, I. A., and C. E. Ray. 2008. Phocidae of the Pliocene of Eastern USA. Pp. 81–140 in C. E. Ray, D. J. Bohaska, I. A. Koretsky, L. W. Ward, and L. G. Barnes, eds. Geology and Paleontology of the Lee Creek Mine, North Carolina IV. Virginia Museum of Natural His- tory Special Publication 14. Koretsky, I. A., and A. E. Sanders. 2002. Paleontol- ogy of the late Oligocene Ashley and Chan- dler Bridge formations of South Carolina, 1: Paleogene pinniped remains; the oldest known seal (Carnivora: Phocidae). Pp. 179–183 in R. J. Emry, ed. Cenozoic Mammals of Land and Sea: Tributes to the Career of Clayton E. Ray. Smithsonian Contributions to Paleobiology 93, Smithsonian Institution Press, Washing- ton, D.C. Kornicker, L. S. 1979. The question mark in tax- onomic literature. Journal of Paleontology 53:761. Krantz, D. E. 1991. A chronology of Pliocene sea- level fluctuations: the U.S. middle Atlantic Coastal Plain record. Quaternary Science Re- views 10:163–174. Ksepka, D. T. 2014. Flight performance of the larg- est volant bird. Proceedings of the National Academy of Science 111:10624–10629. Kurtén, B. 1966. Pleistocene bears of North Amer- ica. 1. Genus Tremarctos, spectacled bears. Acta Zoologica Fennica 115:1-120. Kurtén, B. 1984. Geographic differentiation in the Rancholabrean dire wolf (Canis dirus Leidy) in North America. Pp. 218–227 in H. H. Geno- ways and M. R. Dawson, eds. Contributions in Quaternary Vertebrate Paleontology: a Vol- ume in Memorial to John E. Guilday. Special Publication, Carnegie Museum 8. Kurtén, B., and E. Anderson. 1980. Pleistocene Mammals of North America. Columbia Uni- versity Press, New York, 442 p. Lambert, O. 2004. Systematic revision of the Mio- cene long-snouted dolphin Eurhinodelphis longirostris de Bus, 1872 (Cetacea, Odontoce- ti, Eurhinodelphinidae). Bulletin de L’Institut 210 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) Royal des Sciences Naturelles de Belgique. Sciences de la Terre 74:147–174. Laurito, C. A., A. L. Valerio, and E. A. Pérez Gam- boa. 2005. Los Xenarthras fósiles de la locali- dad de Buenos Aires de Palmares (Blancano tardío–Irvingtoniano temprano) provincia de Alajuela, Costa Rica. Revista Geológica de América Central 33:83–90. Lawrence, K. T., T. D. Herbert, C. M. Brown, M. E. Raymo, and A. M. Haywood. 2009. High- amplitude variations in North Atlantic seas surface temperature during the early Pliocene warm period. Paleoceanography 24:PA2218, doi:10.1029/2008PA001669. Laws, R. A. 1988. Upper Claibornian-Jacksoni- an calcareous nannofossils from the Savan- nah and Santee basins. Geological Society of America Abstracts with Programs 20:276. Leidy, J. 1853. Observations on remains of extinct Cetacea, from the Green Sand of New Jersey, and from South Carolina and Virginia. Pro- ceedings of the Academy of Natural Sciences of Philadelphia 6:377–378. Leidy, J. 1854. Remarks on several fossils indicat- ing new species of extinct Mammalia. Pro- ceedings of the Academy of Natural Sciences of Philadelphia 7:89–90. Leidy, J. 1856. Notice of remains of two species of seals. Proceedings of the Academy of Natural Sciences of Philadelphia 8(6):265. Leidy, J. 1859, 1860. Description of vertebrate fossils. Pp. 99-110 (1859) and pp. 111–122 (1860) in F. S. Holmes, ed. Post-Pleiocene [sic] Fossils of South Carolina. Russell and Jones, Charleston, South Carolina. Leidy, J. 1868a. Notice of some extinct cetaceans. Proceedings of the Academy of Natural Sci- ences of Philadelphia 20:196–197. Leidy, J. 1868b. Notice of some remains of extinct pachyderms. Proceedings of the Academy of Natural Sciences of Philadelphia 20:230–233. Leidy, J. 1869. On the extinct mammalian fauna of Dakota and Nebraska, including an account of some allied forms from other localities, to- gether with a synopsis of the mammalian re- mains of North America. Journal of the Acad- emy of Natural Sciences of Philadelphia, 2nd series, 7:1–472. Leidy, J. 1876a. Remarks on fossils from the Ashley Phosphate Beds. Proceedings of the Academy of Natural Sciences of Philadelphia 28:80–81. Leidy, J. 1876b. Remarks on fossils from the Ashley Phosphate Beds. Proceedings of the Academy of Natural Sciences of Philadelphia 28:86–87. Leidy, J. 1876c. Remarks on vertebrate fossils from the Phosphate Beds of South Carolina. Proceedings of the Academy of Natural Sci- ences of Philadelphia 28:114–115. Leidy, J. 1877. Description of vertebrate remains, chiefly from the Phosphate Beds of South Car- olina. Journal of the Academy of Natural Sci- ences of Philadelphia, 2nd series, 8:209–262. Leidy, J. 1890. Mastodon and capybara of South Carolina. Proceedings of the Academy of Nat- ural Sciences of Philadelphia 42:184–185. Liddicoat, J. C., and N. D. Opdyke. 1981. Mag- netostratigraphy of sediments in the Atlantic Coastal Plain and Pacific Coast of the United States as an aid for dating tectonic deforma- tion. U.S. Geological Survey Open File Report No. 81–232:25 p. Liddicoat, J. C., L. McCartan, R. E. Weems, and E. M. Lemon. 1981. Paleomagnetic investigation of Pliocene and Pleistocene sediments in the Charleston, South Carolina area. Geological Society of America Abstracts with Programs 13:12, 29. Lindsay, E. H., N. D. Opdyke, and N. M. John- son. 1984. Blancan-Hemphillian land mam- mal ages and late Cenozoic mammal dispersal events. Annual Review of Earth and Planetary Science 12:445–488. Lisiecki, L. E., and M. E. Raymo. 2005. A Plio- Pleistocene stack of 57 globally distributed benthic δ18O records. Paleoceanography 20:PA1003. Lister, A. M., and A. V. Sher. 2015. Evolution and dispersal of mammoths across the Northern Hemisphere. Science 350:805–809. Lofgren, D. L., J. A. Lillegraven, W. A. Clemens, P. D. Gingerich, and T. E. Williamson. 2004. Pa- leocene biochronology: the Puercan through ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 211 Clarkforkian Land Mammal Ages. Pp. 43–105 in M. O. Woodburne, ed. Late Cretaceous and Cenozoic Mammals of North America: Bio- stratigraphy and Geochronology. Columbia University Press, New York. Lourens, L., F. Hilgen, N. J. Shackleton, J. Laskar, and D. Wilson. 2004. The Neogene Period. Pp. 409–440 in F. M. Gradstein, J. G. Ogg, and A. G. Smith, eds. A Geologic Time Scale 2004. Cambridge University Press, Cambridge, UK. Lucas, S. G. 2010. Case 3515. Rhynchotherium Falconer, 1868 (Mammalia; Proboscidia): Pro- posed conservation of usage by designation of Rhynchotheirum falconeri Osborn, 1923 as the type species. Bulletin of Zoological Nomen- clature 67:158–162. Lucas, S. G., and G. S. Morgan. 2008. Taxonomy of Rhynchotherium (Mammalia, Proboscidea) from the Miocene-Pliocene of North America. Pp. 71–88 in S. G. Lucus, G. S. Morgan, J. A. Spielmann, and D. R. Prothero, eds. Neogene Mammals. New Mexico Museum of Natural History and Science Bulletin 44. Lucas, S. G., R. J. Emry, and S. E. Foss. 1998. Taxonomy and distribution of Daeodon, an Oligocene-Miocene entelodont (Mammalia: Artiodactyla) from North America. Proceed- ings of the Biological Society of Washington 111:425–435. Ludt, C. J., W. Schroeder, O. Rottmann, and R. Kuehn. 2004. Mitochondrial DNA phylogeog- raphy of red deer (Cervus elaphus). Molecular Phylogenetics and Evolution 31:1064–1083. Lundelius, E. L. 1960. Mylohyus nasutus, long- nosed peccary of the Texas Pleistocene. Bul- letin of the Texas Memorial Museum 1:1–40. Lyell, C. 1845. On the White Limestone of South Carolina and Georgia, and the Eocene strata of other parts of the U.S., with appendix, on the corals, by Mr. Lonsdale. Proceedings of the Geological Society of London 4 (1843– 1845):563–576. MacFadden, B. J. 1984. Systematics and phylog- eny of Hipparion, Neohipparion, Nannippus, and Cormohipparion (Mammalia, Equidae) from the Miocene and Pliocene of the New World. Bulletin of the American Museum of Natural History 179:1–196. MacFadden, B. J. 1998. Equidae. Pp. 537–559 in C. M. Janis, K. M. Scott, and L. L. Jacobs, eds. Evolution of Tertiary Mammals of North America, Volume 1: Terrestrial Carnivores, Ungulates, and Ungulatelike Mammals. Cam- bridge University Press, Cambridge, UK. MacFadden, B. J., and G. S. Morgan. 2003. New oreodont (Mammalia, Artiodactyla) from the late Oligocene (early Arikareean) of Florida. Bulletin of the American Museum of Natural History 279:368–396. MacFadden, B. J., and J. S. Waldrop. 1980. Nan- nippus phlegon (Mammalia, Equidae) from the Pliocene (Blancan) of Florida. Bulletin of the Florida State Museum, Biological Scienc- es 25:1–37. Malde, H. E. 1959. Geology of the Charleston phosphate area, South Carolina. U.S. Geologi- cal Survey Bulletin 1079, 105 p. Manning, E. M. 1997. An early Oligocene rhinoc- eros jaw from the marine Byram Formation of Mississippi. Mississippi Geology 18:13–31. Mansfield, W. C. 1935. New Miocene gastropods and scaphopods from Alaqua Creek Valley, Florida. Florida Geological Survey Bulletin 12:1–64. Martin, L. D. 1972. The microtine rodents of the Mullen assemblage from the Pleistocene of north central Nebraska. Bulletin of the Uni- versity of Nebraska State Museum 9:173-182. Martin, L. D., B. M. Gilbert, and D. B. Adams. 1977. A cheetah-like cat in the North Ameri- can Pleistocene. Science 195:981–982. Martin, R. A. 1969. Taxonomy of the giant Pleisto- cene beaver Castoroides from Florida. Journal of Paleontology 43:1033–1041. Martin, R. A. 1974. Fossil mammals from the Cole- man IIA fauna, Sumter County. Pp. 35–99 in S. D. Webb, ed. Pleistocene Mammals of Florida. University Press Florida, Gainesville. Martin, R. A. 1975. Giant Pleistocene beavers and the Waccasassa River, Levy County, Florida. Bulletin of the New Jersey Academy of Sci- ences 20:29–30. 212 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) Martin, R. A., and J. M. Sneed. 1989. Late Pleis- tocene records of caribou and elk from Geor- gia and Alabama. Georgia Journal of Science 47:117–122. Martin, R. A., F. Marcolini, and F. Grady. 2009. The early Pleistocene Hamilton Cave muskrats and a review of muskrat size change through the late Neogene. Paludicola 7:61–66. Martin, R. A., P. Peláez-Campomanes, J. G. Honey, D. L. Fox, R. J. Zakrzewski, L. B. Albright, E. H. Lindsay, N. D. Opdyke, and H. T. Good- win. 2008. Rodent community change at the Pliocene-Pleistocene transition in southwest- ern Kansas and identification of the Microtus immigration event on the Central Great Plains. Palaeogeography, Palaeoclimatology, and Pal- aeoecology 267:196–207. May, S. R., A. M. Sarna-Wojcicki, E. H. Lindsay, M. O. Woodburne, N. D. Opdyke, E. Wan, D. B. Wahl, and H. Olson. 2014. Geochronol- ogy of the Upper Alturas Formation, northern California: implications for the Hemphillian- Blancan North American Land Mammal Age boundary. Palaeontologia Electronica 17.3.43A:1–13. McCartan, L. 1990. Introduction. Pp. 1–5 in Studies Related to the Charleston, SC, Earthquake of 1886 – Neogene and Quaternary Lithostratig- raphy and Biostratigraphy. U.S. Geological Survey Professional Paper 1367–A. McCartan, L., E. M. Lemon, Jr., and R. E. Weems. 1984. Geologic map of the area between Charleston and Orangeburg, South Carolina. U.S. Geological Survey Miscellaneous Inves- tigations Series map I–1472, scale 1:250,000. McCartan, L., R. E. Weems, and E. M. Lemon, Jr. 1980. The Wando Formation (upper Pleisto- cene) in the Charleston, S.C., area. U.S. Geo- logical Survey Bulletin 1502 A:A110–A116. McCartan, L., R. E. Weems, and E. M. Lemon, Jr. 1990. Quaternary stratigraphy in the vicinity of Charleston, South Carolina, and its relation- ship to local seismicity and regional tectonism. Pp. A1–A39 in Studies Related to the Charles- ton, SC, Earthquake of 1886 – Neogene and Quaternary Lithostratigraphy and Biostratig- raphy. U.S. Geological Survey Professional Paper 1367–A. McCartan, L., J. P. Owens, B. W. Blackwelder, B. J. Szabo, D. F. Belknap, N. Kriausakul, R. M. Mitterer, and J. F. Wehmiller. 1982. Compari- son of amino acid racemezation geochronom- etry with lithostratigraphy, biostratigraphy, uranium-series coral dating, and magneto- stratigraphy in the Atlantic Coastal Plain of the Southeastern United States. Quaternary Research 18:337–359. McDonald, J. N., and C. E. Ray. 1993. Records of Musk Oxen from the Atlantic Coastal Plain of North America. The Mosasaur (Delaware Val- ley Paleontological Society) 5:1–18. McDonald, J. N., C. E. Ray, and F. Grady. 1996. Pleistocene caribou (Rangifer tarandus) in the eastern United States: new records and range extensions. Pp. 406–430 in K. M. Stewart and K. L. Seymour, eds. Palaeoecology and Pal- aeoenvironments of the Late Cenozoic Mam- mals: Tributes to the Career of C. S. (Rufus) Churcher. University of Toronto Press, Toron- to, Canada. McDonald, J. N., C. E. Ray, and M. W. Ruddell. 2000. New records and range extensions of Cervalces, Rangifer, and Bootherium in the Southeastern United States. Current Research in the Pleistocene 17:131–133. McLeod, S. A., and L. G. Barnes. 2008. A new ge- nus and species of Eocene protocetid archaeo- cete whale (Mammalia, Cetacea) from the Atlantic Coastal Plain. Pp. 73–98 in X. Wang and L. G. Barnes, eds. Geology and Vertebrate Paleontology of Western and Southern North America: Contributions in Honor of David P. Whistler. Natural History Museum of Los An- geles County Science Series 41. Melchior, R. C. 1998. Paleobotany of the Wil- liamsburg Formation (Paleocene) at the Santee Rediversion Site, Berkeley County, South Car- olina. Pp. 49–121 in A. E. Sanders, ed. Paleo- biology of the Williamsburg Formation (Black Mingo Group, Paleocene) of South Carolina, U.S.A. Transactions of the American Philo- sophical Society 88. ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 213 Mihlbachler, M. C. 2008. Species taxonomy, phy- logeny, and biogeography of the Brontoth- eriidae (Mammalia: Perissodactyla). Bulletin of the American Museum of Natural History 311:1–475. Miller, G. S. 1923.The telescoping of the cetacean skull. Smithsonian Miscellaneous Collections 76:1–71. Miller, K. G., J. D. Wright, and R. G. Fairbanks. 1991. Unlocking the icehouse: Oligocene- Miocene oxygen isotopes, eustasy and mar- gin erosion. Journal of Geophysical Research 96:6829–6848. Miller, K. G., M. A. Kominz, J. V. Browning, J. D. Wright, G. S. Mountain, M. E. Katz, P. J. Sugarman, B. S. Cramer, N. Christie-Blick, and S. F. Pekar. 2005. The Phanerozoic record of global sea-level change. Science 310:1293– 1298. Mones, A. 1984. Estudios sobre la Familia Hydro- choeridae, XIV. Revisión systemática (Mam- malia: Rodentia). Senckenbergiana Biologica 65:1–17. Mones, A., and J. Ojasti. 1986. Hydrochoerus hy- drochaeris. Mammal Species 264:1–7. Morgan, G. S. 1989. Miocene vertebrate fau- nas from the Suwannee River Basin of north Florida and south Georgia. Pp. 26–53 in G. S. Morgan, ed. Miocene Paleontology and Stra- tigraphy of the Suwannee River Basin of north Florida and south Georgia, Southeastern Geo- logical Society Guidebook 30, Tallahassee. Morgan, G. S. 1993. Mammalian biochronology and marine-nonmarine correlations in the Neogene of Florida. Florida Geological Sur- vey Special Publication 37:55–66. Morgan, G. S. 1994. Miocene and Pliocene marine mammals from the Bone Valley Formation of central Florida. Pp. 239–268 in A. Berta and T. A. Deméré, eds. Contributions in Marine Mammal Paleontology Honoring Frank C. Whitmore, Jr. Proceedings of the San Diego Society of Natural History 29. Morgan, G. S. 2002. Late Rancholabrean mammals from southernmost Florida, and the Neotropi- cal influence in Florida Pleistocene faunas. Pp. 15–38 in R. J. Emry, ed. Cenozoic Mam- mals of Land and Sea: Tributes to the Career of Clayton E. Ray. Smithsonian Contributions to Paleobiology 93. Morgan, G. S. 2005. The Great American Biotic Interchange in Florida. Pp. 271–311 in R. C. Hulbert, G. S. Morgan, and J. A. Baskin, eds. Cenozoic Vertebrates of the Americas: Papers to Honor S. David Webb, Bulletin of the Flor- ida Museum of Natural History 45. Morgan, G. S. 2008. Vertebrate fauna and geo- chronology of the Great American Biotic In- terchange in North America. Pp. 93–140 in S. G. Lucas, G. S. Morgan, J. A. Spielmann, and D. R. Prothero, eds. Neogene Mammals. New Mexico Museum of Natural History and Sci- ence Bulletin 44. Morgan, G. S., and S. D. Emslie. 2010. Tropical and western influences in vertebrate faunas from the Pliocene and Pleistocene of Florida. Quaternary International 217:143–158. Morgan, G. S., and A. H. Harris. 2015. Pliocene and Pleistocene vertebrates of New Mexico. Pp. 233–428 in S. G. Lucas and R. M. Sul- livan, eds., Fossil Vertebrates in New Mexico. New Mexico Museum of Natural History and Science Bulletin 68. Morgan, G. S., and R. C. Hulbert, Jr. 1995. Over- view of the geology and vertebrate biochro- nology of the Leisey Shell Pit Local Fauna, Hillsborough County, Florida. Pp. 1–92 in R. C. Hulbert, G. S. Morgan, and S. D. Webb, eds. Paleontology and Geology of the Leisey Shell Pits, Early Pleistocene of Florida. Bul- letin of the Florida Museum of Natural History 37, Part I. Morgan, G. S., and J. A. White. 1995. Small mam- mals (Insectivora, Lagomorpha, and Rodentia) from the early Pleistocene (Irvingtonian) Lei- sey Shell Pit Local Fauna, Hillsborough Coun- ty, Florida. Pp. 397–461 in R. C. Hulbert, G. S. Morgan, and S. D. Webb, eds. Paleontology and Geology of the Leisey Shell Pits, Early Pleistocene of Florida. Bulletin of the Florida Museum of Natural History 37, Part II. Morgan, G. S., and R. S. White, Jr. 2005. Miocene 214 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) and Pliocene vertebrates from Arizona. Pp. 115–136 in A. B. Heckert and S. G. Lucas, eds. Vertebrate Paleontology in Arizona, New Mexico Museum of Natural History and Sci- ence Bulletin 29. Morgan, G. S., B. J. MacFadden, and M. Martínez. 2016. Quaternary gomphotheres (Mammalia: Proboscidea: Gomphotheriidae) from the con- tinental shelf, Pearl Islands, Panama. Quater- nary International 392:335-348. Mudelsee, M., and M. E. Raymo. 2005. Slow dy- namics of the Northern Hemisphere glaciation. Paleoceanography 20:PA4022 (1–14). Murphey, P. C., T. S. Kelly, K. R. Chamberlain, K. Tsukui, and W. C. Clyde. 2018. Mammals from the earliest Uintan (middle Eocene) Tur- tle Bluff Member, Bridger Formation, south- western Wyoming, USA, Part 3: Marsupialia and a reevaluation of the Bridgerian-Uintan North American Land Mammal Age. Palae- ontologia Electronica, article 21.2.25A, doi. org/10.26879/804. Myrick, A. C. 1979. Variation, taphonomy, and adaptation of the Rhabdosteidae (= Eurhino- delphidae) (Odontoceti, Mammalia) from the Calvert Formation of Maryland and Virginia. Ph.D. dissertation. University of California, Los Angeles. 347 p. Naish, T. R., and 32 others. 2001. Orbitally induced oscillations in the East Antarctic ice sheet at the Oligocene/Miocene boundary. Nature 413:719–723. Nowak, R. M. 1979. North American Quaternary Canis. Monograph of the Museum of Natural History, University of Kansas 6:1–154. Nowak, R. M. 2002. The original status of wolves in eastern North America. Southeastern Natu- ralist 1:95–130. Nystrom, P. G., Jr., R. H. Willoughby, and L. K. Price. 1991. Cretaceous and Tertiary stratigra- phy of the Upper Coastal Plain, South Caro- lina. Pp. 221–240 in J. W. Horton, Jr., and V. A. Zullo, eds. The Geology of the Carolinas: Carolina Geological Survey Fiftieth Anniver- sary Volume. The University of Tennessee Press, Knoxville. Olsen, S. 1957. A new beak-jawed mastodont from Florida. Journal of the Palaeontological Soci- ety of India 2:131–135. Osborn, H. F. 1918. Equidae of the Oligocene, Miocene, and Pliocene of North America, iconographic type revision. Memoirs of the American Museum of Natural History, New Series 2:1–217. Packard, A. S., Jr. 1871. Geology of the phosphate beds of South Carolina. Bulletin of the Essex Institute 3:55–58. (also in American Natural- ist, 1872, volume 6:55–58). Pälike, H., R. D. Norris, J. O. Herrie, P. A. Wilson, H. K. Coxall, C. H. Lear, N. J. Shackleton, A. K. Tripati, and B. S. Wade. 2006. The heart- beat of the Oligocene climate system. Science 314:1894–1898. Palmer, A. A. 1988. Cenozoic radiolarians from Ocean Drilling Program Leg 101, Bahamas (Site 627 and 628) and surrounding region. Pp. 105–116 in J. A. Austin and W. Schlager, eds. Proceedings of the Ocean Drilling Program, Scientific Results 101. Texas A&M Univer- sity, College Station. Parmalee, P. W., and R. W. Graham. 2002. Addi- tional records of the giant beaver, Castoroides, from the mid-South: Alabama, Tennessee, and South Carolina. Pp. 65–72 in R. J. Emry, ed. Cenozoic Mammals of Land and Sea: Tributes to the Career of Clayton E. Ray. Smithsonian Contributions to Paleobiology 93. Paula Couto, C. de. 1954. Megaterios intertropicals de Pleistoceno. Anais da Academia Brasileria de Cieñcias 26:447–463. Pérez, M. E., M. C. Vallejo-Pareja, J. D. Carrillo, and C. Jaramillo. 2017. A new Pliocene capy- bara (Rodentia, Caviidae) from northern South America (Guajira, Columbia), and its implica- tions for the Great American Biotic Interchange. Journal of Mammalian Evolution 24:111–125. Petiver, J. 1705. An account of animals and shells sent from South Carolina to Mr. James Petiver, F.R.S. Philosophical Transactions of the Royal Society 24(299):1952–1960. Petuch, E. J. 2003. Cenozoic Seas: The View from Eastern North America. CRC Press, Boca Ra- ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 215 ton, Florida, 308 p. Petuch, E. J., and C. E. Roberts. 2007. The Geolo- gy of the Everglades and Adjacent Areas. CRC Press, Boca Raton, Florida, 212 p. Pillans, B., and P. Gibbard. 2012. The Quaternary Period. Pp. 979–1010 in F. M. Gradstein, J. G. Ogg, M. D. Schmitz, and G. M. Ogg, eds. The Geologic Time Scale 2012. Elsevier, Amster- dam. Polly, P. D. 1997. Ancestry and species definition in paleontology: a stratocladistic analysis of Paleocene-Eocene Viverravidae (Mamma- lia, Carnivora) from Wyoming. Contributions from the Museum of Paleontology, The Uni- versity of Michigan 30:1–53. Popenoe, P., V. J. Henry, and F. M. Idris. 1987. Gulf trough – the Atlantic connection. Geol- ogy 15:327–332. Pooser, W. K. 1965. Biostratigraphy of Cenozoic Ostracoda from South Carolina. The Univer- sity of Kansas Paleontological Contributions, Arthropoda, 8:1–80. Portell, R. W., G. H. Means, and R. C. Hulbert, Jr. 2012. SMR Aggregates, Inc., Sarasota, Flori- da. Southeastern Geological Society Field Trip Guidebook 56. Tallahassee, 24 p. Post, K., O. Lambert, and G. Bianucci. 2008. First record of Tusciziphius crispus (Cetacea, Ziphi- idae) from the Neogene of the US east coast. Deinsea 12:1-10. Powell, R. J. 1984. Lithostratigraphy, deposition- al environment, and sequence framework of the middle Eocene Santee Limestone, South Carolina Coastal Plain. Southeastern Geology 25:79–100. Powell, R. J., and G. R. Baum. 1982. Eocene bio- stratigraphy of South Carolina and its rela- tionship to Gulf Coast zonations and global changes in coastal onlap. Geological Society of America Bulletin 93:1099–1108. Pratt, A. E., and R. M. Petkewich. 1989. Fossil vertebrates from the Marks Head Formation (Lower Miocene) of southeastern Georgia. Journal of Vertebrate Paleontology 9 (supple- ment to No. 3):35A. Prothero, D. R. 1998. Rhinocerotidae. Pp. 595–605 in C. M. Janis, K. M. Scott, and L. L. Jacobs, eds. Evolution of Tertiary Mammals in North America, Volume 1:Terrestrial Carnivores, Ungulates, and Ungulatelike Mammals. Cam- bridge University Press, Cambridge. Prothero, D. R. 2005. The Evolution of North American Rhinoceroses. Cambridge Univer- sity Press. Cambridge, 218 p. Prothero, D. R. 2009. The early evolution of the North American peccaries (Artiodactyla: Ta- yassuidae). Pp. 509–542 in L. B. Albright III, ed. Papers on Geology, Vertebrate Paleontol- ogy, and Biostratigraphy in Honor of Michael O. Woodburne. Museum of Northern Arizona Bulletin 65, Flagstaff. Prothero, D. R., and R. J. Emry. 2004. The Chadro- nian, Orellan, and Whitneyan North American Land Mammal Ages. Pp. 156–168 in M. O. Woodburne, ed. Late Cretaceous and Cenozo- ic Mammals of North America, Biostratigra- phy and Geochronology. Columbia University Press, New York. Prothero, D. R., and E. M. Manning. 1987. Mio- cene rhinoceroses from the Texas gulf coastal plain. Journal of Paleontology 61:388–423. Puri, H. S., and R. O. Vernon. 1964. Summary of the geology of Florida and a guidebook to the classic exposures: Florida Geological Survey Special Publication 5, 312 p. Putney, T. R., M. P. Katuna, and M. S. Harris. 2004. Subsurface stratigraphy and geomorphology of the Grand Strand, Georgetown and Horry counties, South Carolina. Southeastern Geol- ogy 42:217–236. Rabeder, G., and F. Steininger. 1975. Die direkten biostratigraphishen korrelationsmoglichkeiten von Saugierterfaunen aus dem Oligo-Miozän der Centralen Paratethys. Proceedings of the Sixth Congress on Mediterranean Neogene Stratigraphy, Bratislava, Yugoslavia 1:177– 183. Raffi, I., J. Backman, E. Fornaciari, H. Pälike, D. Rio, L. Lourens, and F. Hilgen. 2006. A review of calcareous nannofossil astrobiochronology encompassing the past 25 million years. Qua- ternary Science Reviews 25:3113–3137. 216 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) Ramírez-Cruz, G. A., and M. Montellano-Balleste- ros. 2014. Two new glyptodont records (Mam- malia: Cingulata) from the late Pleistocene of Tamaulipas and Tlaxcala, Mexico: implica- tions for the taxonomy of the genus Glypto- therium. The Southwestern Naturalist, 59(4): 522–530. Ray, C. E. 1965. A glyptodont from South Caro- lina. Charleston Museum Leaflet 27:1–12. Ray, C. E. 1967. Pleistocene mammals from Ladds, Bartow County, Georgia. Bulletin of the Geor- gia Academy of Science 25:120–150. Ray, C. E. 1976a. Geography of phocid evolution. Systematic Zoology 25:391–406. Ray, C. E. 1976b. Phoca wymani and other Tertiary seals (Mammalia: Phocidae) described from the Eastern Coastal Plain of North America. Smith- sonian Contributions to Paleobiology 28:1–36. Ray, C. E. 1983. Editor. Geology and Paleontol- ogy of the Lee Creek Mine, North Carolina, I. Smithsonian Contributions to Paleobiology 53, 529 p. Ray, C. E. 1987. Editor. Geology and Paleontol- ogy of the Lee Creek Mine, North Carolina, II. Smithsonian Contributions to Paleobiology 61, 283 p. Ray, C. E. 2001. Editor. Geology and Paleontol- ogy of the Lee Creek Mine, North Carolina, III. Smithsonian Contributions to Paleobiol- ogy 90, 365 p. Ray, C. E. and A. E. Sanders. 1984. Pleistocene ta- pirs in the eastern United States. Pp. 283–315 in H. H. Genoways and M. R. Dawson, eds. Contributions in Quaternary Vertebrate Pale- ontology: a Volume in Memorial to John E. Guilday. Carnegie Museum of Natural History Special Publication 8, Pittsburgh. Ray, C. E., A. Whetmore, and D. H. Dunkle. 1968. Fossil vertebrates from the marine Pleistocene of southeastern Virginia. Smithsonian Miscel- laneous Collections 153:1–25. Ray, C. E., D. J. Bohaska, I. A. Koretsky, L. W. Ward, and L. G. Barnes, eds. 2008. Geol- ogy and Paleontology of the Lee Creek Mine, North Carolina, IV. Virginia Museum of Natu- ral History Special Publication 14, 527 p. Raymo, M. E., and J. X. Mitrovica. 2012. Collapse of polar ice sheets during the stage 11 intergla- cial. Nature 483:453–456. Repenning, C. A. 1992. Allophaiomys and the age of the Olyor Suite, Krestovka sections, Yakutia. U.S. Geological Survey Bulletin 2037:1–98. Repenning, C. A., and F. Grady. 1988. The mi- crotine rodents of the Cheetah Room Fauna, Hamilton Cave, West Virginia, and the sponta- neous origin of Synaptomys. U.S. Geological Survey Bulletin 1853:1–28. Repenning, C. A., T. R. Weasma, and G. R. Scott. 1995. The early Pleistocene (latest Blancan– earliest Irvingtonian) Froman Ferry Fauna and history of the Glenns Ferry Formation, south- western Idaho. U.S. Geological Survey Bulle- tin 2105:1-86. Robertson, J. S. 1976. Late Pleistocene mammals from Haile XV A, Alachua County, Florida. Bulletin of the Florida State Museum, Biologi- cal Sciences 20(3):111–186. Robinson, M. M., H. J. Dowsett, G. S. Dwyer, and K. T. Lawrence. 2008. Reevaluation of mid-Pliocene North Atlantic sea surface tem- peratures. Paleoceanography 23:PA3213, doi:10.1029/2008PA001608. Rose, K. D. 1999. Part 7. Fossil mammals from the early Eocene Fisher/Sullivan site. Pp. 133–138 in R. E. Weems and G. J. Grimsley, eds. Early Eocene Vertebrates and Plants from the Fisher/ Sullivan Site (Nanjemoy Formation) Stafford County, Virginia. Virginia Division of Mineral Resources Publication 152, Charlottesville. Rose, K. D. 2000. Land mammals from the late Paleocene Aquia Formation: the first early Cenozoic mammals from Maryland. Proceed- ings of the Biological Society of Washington 113:855–863. Rose, K. D. 2010. New marsupial from the early Eocene of Virginia. Journal of Paleontology 84:561–565. Rossbach, T. J., and J. G. Carter. 1991. Milluscan biostratigraphy of the Lower River Bend For- mation at the Martin Marietta quarry, New Bern, North Carolina. Journal of Paleontology 65:80–118. ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 217 Roth, J. A., and J. Laerm. 1980. A late Pleistocene vertebrate assemblage from Edisto Island, South Carolina. Brimleyana 3:1–29. Rothausen, K. 1968. Die systematische Stellung der europäischen Squalodontidae (Odonto- ceti, Mammalia). Paläontologische Zeitschrift 42:83–104. Rothausen, K. 1971. Cetotheriopsis tobieni n. sp., der erste paläogene Bartenwal (Cetotheriidae, Mysticeti, Mammalia) nördlich des Tethysrau- mes. Abhandlung des Hessische Landesampt- esn für Bodenforschung 60:131–148. Rovere, A., P. J. Hearty, J. Austermann, J. X. Mi- trovica, J. Gale, R. Moucha, A. M. Forte, and M. E. Raymo. 2015. Mid-Pliocene shorelines of the US Atlantic Coastal Plain – an improved elevation database with comparison to Earth model predictions. Earth Science Reviews 145:117–131. Rovere, A., M. E. Raymo, J. X. Mitrovica, P. J. Hearty, M. J. O’Leary, and J. D. Inglis. 2014. The Mid-Pliocene sea-level conundrum: gla- cial isostasy, eustasy and dynamic topography. Earth and Planetary Science Letters 387:27– 33. Rowley, D. B., A. M. Forte, R. Moucha, J. X. Mi- trovica, N. A. Simmons, and S. P. Grand. 2013. Dynamic topography change of the eastern United States since 3 million years ago. Sci- ence 340:1560–1563. Ruddy, D. 2013. Recording the beginnings of South Carolina river diving. Quarterly Report- er, South Carolina Institute of Archaeology and Anthropology 4(1):6–7. Ruffin, E. 1843. Report of the commencement and progress of the Agricultural Survey of South Carolina for 1843. Columbia, S.C., p. 120. Sanders, A. E. 1974. A paleontological survey of the Cooper Marl and Santee Limestone near Harleyville, South Carolina – preliminary re- port. South Carolina Division of Geology, Geologic Notes 18:4–12. Sanders, A. E. 1980. Excavation of Oligocene ma- rine fossil beds near Charleston, South Caro- lina. National Geographic Society Research Reports 12:601–621. Sanders, A. E. 1996. The systematic position of the primitive odontocete Xenorophus sloanii (Mammalia, Cetacea) and two new taxa from the late Oligocene of South Carolina, U.S.A. Sixth North American Paleontological Con- vention Abstracts of Papers, The Paleontologi- cal Society Special Publication No. 8:338. Sanders, A. E., ed. 1998a. Paleobiology of the Wil- liamsburg Formation (Black Mingo Group; Paleocene) of South Carolina, U.S.A. Trans- actions of the American Philosophical Society 88, 276 p. Sanders, A. E. 1998b. Paleobiology of the Wil- liamsburg Formation (Black Mingo Group; Paleocene) of South Carolina: Summary and Conclusions. Pp. 255–268 in A. E. Sanders, ed. Paleobiology of the Williamsburg Forma- tion (Black Mingo Group; Paleocene) of South Carolina, U.S.A. Transactions of the American Philosophical Society 88. Sanders, A. E. 1998c. Introduction. Pp. 1–8 in A. E. Sanders, ed. Paleobiology of the Williamsburg Formation (Black Mingo Group; Paleocene) of South Carolina, U.S.A. Transactions of the American Philosophical Society 88. Sanders, A. E. 2002. Additions to the Pleistocene Mammal Faunas of South Carolina, North Carolina, and Georgia. Transactions of the American Philosophical Society 92:1–152. Sanders, A. E., and W. D. Anderson, Jr. 1999. Nat- ural History Investigations in South Carolina from Colonial Times to the Present. University of South Carolina Press, Columbia, 333 p. Sanders, A. E., and L. G. Barnes. 2002a. Paleontol- ogy of the late Oligocene Ashley and Chan- dler Bridge Formations of South Carolina, 2: Micromysticetus rothauseni, a primitive ceto- theriid mysticete (Mammalia: Cetacea). Pp. 271–294 in R. J. Emry, ed. Cenozoic Mam- mals of Land and Sea: Tributes to the Career of Clayton E. Ray. Smithsonian Contributions to Paleobiology 93, Smithsonian Institution Press, Washington, D.C. Sanders, A. E., and L. G. Barnes. 2002b. Paleontol- ogy of the late Oligocene Ashley and Chandler Bridge Formations of South Carolina, 3: Eo- 218 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) mysticetidae, a new family of primitive mys- ticetes (Mammalia: Cetacea). Pp. 313–356 in R. J. Emry, ed. Cenozoic Mammals of Land and Sea: Tributes to the Career of Clayton E. Ray. Smithsonian Contributions to Paleobiol- ogy 93, Smithsonian Institution Press, Wash- ington, D.C. Sanders, A. E., and J. H. Geisler. 2015. A new basal odontocete from the upper Rupelian of South Carolina, U.S.A., with contributions to the systematics of Xenorophus and Mirocetus (Mammalia, Cetacea). Journal of Vertebrate Paleontology 35:e890107 (22 p.). Sanders, A. E., and M. J. Katuna. 2000. Proposal for revision of the stratigraphic nomenclature of the Santee Limestone (Eocene) of South Carolina. Abstracts with Programs, Geologi- cal Society of America, 49th Annual Meeting, Southeastern Section, March 2000, 32(2):A– 71. Sanders, A. E., R. E. Weems, and L. B. Albright, III. 2009. Formalization of the middle Pleisto- cene “Ten Mile Hill Beds” in South Carolina with evidence for placement of the Irvingto- nian-Rancholabrean boundary. Pp. 363–370 in L. B. Albright, III, ed. Papers on Geology, Vertebrate Paleontology, and Biostratigraphy in Honor of Michael O. Woodburne. Museum of Northern Arizona Bulletin 65, Flagstaff. Sanders, A. E., R. E. Weems, and E. M. Lemon, Jr. 1982. Chandler Bridge Formation - a new Oli- gocene stratigraphic unit in the lower coastal plain of South Carolina. Contributions to Stra- tigraphy, U.S. Geological Survey Bulletin 1529-H:105–124. Santana, F. R. de, D. J. Cicimurri, and J. A. Barbosa. 2011. Mew material of Apocopodon sericeus Cope, 1886 (Myliobatiformes, Myliobatidae) from the Paraiba Basin (northeastern Brazil), and South Carolina (USA) with a reanalysis of the species. PalArch’s Journal of Vertebrate Paleontology 8:1–20. Schoch, R. M. 1985. Preliminary description of a new late Paleocene land-mammal fauna from South Carolina, USA. Postilla 196:1–13. Schoch, R. M. 1998. Late Paleocene land-mammals from the Williamsburg Formation (Black Min- go Group) of South Carolina. Pp 229–245 in A. E. Sanders, ed. Paleobiology of the Williams- burg Formation (Black Mingo Group; Paleo- cene) of South Carolina, U.S.A. Transactions of the American Philosophical Society 88. Schubert, B. W. 2008. Dental evolution in North American short-faced bears. Journal of Verte- brate Paleontology 28:138A. Schubert, B. W., R. C. Hulbert, Jr., B. J. MacFad- den, M. Searle, and S. Searle. 2010. Giant short-faced bears (Arctodus simus) in Pleisto- cene Florida USA, a substantial range exten- sion. Journal of Paleontology 84:79–87. Schwimmer, D. R., A. E. Sanders, B. R. Erickson, and R. E. Weems. 2015. A Late Cretaceous dinosaur and reptile assemblage from South Carolina, U.S.A. Transactions of the American Philosophical Society 105, 157 p. Scott, T. M., and W. D. Allmon, eds. 1992. Plio- Pleistocene Stratigraphy and Paleontology of Southern Florida. Florida Geological Survey Special Publication 36, 194 p. Secord, R., P. D. Gingerich, M. E. Smith, W. C. Clyde, P. Wilf, and B. S. Singer. 2006. Geo- chronology and mammalian biostratigraphy of middle and upper Paleocene continental strata, Bighorn Basin, Wyoming. American Journal of Science 306:211–245. Short, R. A., S. C. Wallace, and L. G. Emmert. 2019. A new species of Teleoceras (Mamma- lia, Rhinocerotidae) from the late Hemphillian of Tennessee. Florida Museum of Natural His- tory Bulletin 56:183–260. Shotwell, J. A. 1970. Pliocene mammals of south- east Oregon and adjacent Idaho. Bulletin of the Museum of Natural History, University of Oregon 17:1–103. Simpson, G. G. 1929. Pleistocene mammal fauna of the Seminole Field, Pinellas County, Flori- da. Bulletin of the American Museum of Natu- ral History 56:561–599. Simpson, G. G. 1942. The beginnings of vertebrate paleontology in North America. Proceed- ings of the American Philosophical Society 86:130–188. ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 219 Sloan, E. 1908. Catalogue of the mineral localities of South Carolina. South Carolina Geological Survey Bulletin 2, 505 p. Snyder, S. W., L. L. Mauger, and D. Ames. 2001. Benthic foraminifera and paleoecology of the Pliocene Yorktown and Chowan River forma- tions, Lee Creek Mine, North Carolina, USA. Journal of Foraminiferal Research 31:244– 274. Sosdian, S., and Y. Rosenthal. 2009. Deep-sea tem- perature and ice volume changes across the Pliocene-Pleistocene climate transitions. Sci- ence 325:306–310. Spearing, K. D. 2006. Testing the validity of the fossil felid species Felis amnicola. Geological Society of America Abstracts with Programs 38:63. Stirton, R. A. 1940. Phylogeny of North American Equidae. University of California Publica- tions, Bulletin of the Department of Geologi- cal Sciences 25:165–198. Stock, C. 1950. Bears from the Pleistocene cave of San Josecito, Nuevo Leon, Mexico. Journal of the Washington Academy of Science 40:317– 321. Storer, J. E. 2003. The eastern Beringian vole Mi- crotus deceitensis (Rodentia, Muridae, Arvi- colinae) in late Pliocene and early Pleistocene faunas of Alaska and Yukon. Quaternary Re- search 60:84–93. Sussman, D. R., F. W. Croxen III, H. G. McDon- ald, and C. A. Shaw. 2016. Fossil porcupine (Mammalia, Rodentia, Erethizontidae) from the El Golfo de Santa Clara, Sonora, Mexico, with a review of the taxonomy of the North American erethizontids. Natural History Mu- seum of Los Angelese County Contributions to Science 524:1-29. Szabo, B. J. 1985. Uranium-series dating of fossil corals from marine sediments of southeastern United States Atlantic Coastal Plain. Bulletin of the Geological Society of America 96:398– 406. Tedford, R. H. 1970. Principles and practices of mammalian geochronology in North America. Proceedings of the North American Paleonto- logical Convention Pt. F:666–703. Tedford, R. H. 1981. Mammalian biochronology of the late Cenozoic basins of New Mexico. Geo- logical Society of America Bulletin 92:1008– 1022. Tedford, R. H., and M. E. Hunter. 1984. Miocene marine-nonmarine correlations, Atlantic and Gulf Coastal Plains, North America. Palaeo- geography, Palaeoclimatology, Palaeoecology 47:129–151. Tedford, R. H., X. Wang, and B. E. Taylor. 2009. Phylogenetic systematics of the North Ameri- can fossil Caninae (Carnivora: Canidae). Bul- letin of the American Museum of Natural His- tory 325:1–218. Thewissen, J. G. M. 1990. Evolution of Paleocene and Eocene Phenacodontidae (Mammalia, Condylarthra). University of Michigan Papers on Paleontology 29:1–107. True, F. W. 1907. Remarks on the type of the fos- sil cetacean Agorophius pygmaeus (Mül- ler). Smithsonian Miscellaneous Collections 1694:1–8. Tseng, Z. J., and J. H. Geisler. 2016. The first fossil record of borophagine dogs (Mammalia, Car- nivora) from South Carolina, U.S.A. Journal of Vertebrate Paleontology 36: doi:10.1080/02 724634.2015.1062022. Tsukui, K., and W. C. Clyde. 2012. Fine-tuning the calibration of the early to middle Eocene geo- magnetic polarity time scale: paleomagnetism of radioisotopically dated tuffs from Laramide foreland basins. Geological Society of Ameri- ca Bulletin 124:870–885. Tuomey, M. 1847. Notice of the discovery of a cranium of the zeuglodon. Proceedings of the Academy of Natural Sciences of Philadelphia 3:151–153. Tuomey, M. 1848. Report on the Geology of South Carolina. Columbia: A. S. Johnson. 1–293. Tuomey, M., and F. S. Holmes. 1857. Pleiocene [sic] Fossils of South Carolina. Russell and Jones, Charleston, South Carolina, 152 p. Uhen, M. D. 1999. New species of protocetid ar- chaeocete whale, Eocetus wardii (Mammalia: Cetacea) from the middle Eocene of North 220 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) Carolina. Journal of Paleontology 73:512–528. Uhen, M. D. 2008. Archaeoceti. Pp. 557–565 in C. M. Janis, G. F. Gunnell, and M. D. Uhen, eds. Evolution of Tertiary Mammals of North America Volume 2: Small Mammals, Xen- arthrans, and Marine Mammals. Cambridge University Press, New York. Uhen, M. D. 2013. A review of North American Basilosauridae. Pp. 1–45 in J. Ebersole and T. Ikejiri, eds. Contributions to Alabama Paleon- tology, Alabama Museum of Natural History Bulletin 31. The University of Alabama, Tus- caloosa. Uhen, M. D., and P. D. Gingerich. 2001. New ge- nus of Dorudontine archaeocete (Cetacea) from the middle-to-late Eocene of South Caro- lina. Marine Mammal Science 17:1–34. Van Beneden, P. –J. 1877. Description des Osse- ments Fossiles des Environs d’Anvers, pre- mière partie: Pinnipèdes ou Amphithériens. Musée Royal d’Histoire Naturelle de Belgique Annales I, 88 p. Vandenberghe, N., F. J. Hilgen, and R. P. Speijer. 2012. The Paleogene Period. Pp. 855–922 in F. M. Gradstein, J. G. Ogg, M. D. Schmitz, and G. M. Ogg, eds. The Geologic Time Scale 2012. Elsevier, Amsterdam. Van Nieuwenhuise, D. S., and D. J. Colquhoun. 1982. The Paleocene-Eocene Black Mingo Group of the east central coastal plain of South Carolina. South Carolina Geology 26:47–67. Vanuxem, L. 1826. Report on a mineralogical and geological examination of the state of South Carolina. Pp. 25–30 in R. Mills, ed. Statis- tics of South Carolina, Including a View of its Natural, Civil, and Military History, General and Particular. Hurlbut and Lloyd, Charleston, South Carolina. Van Valkenburgh, B., F. Grady, and B. Kurtén. 1990. The Plio-Pleistocene cheetah-like cat Miraci- nonyx inexpectatus of North America. Journal of Vertebrate Paleontology 10:434–454. Vélez-Juarbe, J. 2014. Ghosts of seagrasses past: using sirenians as a proxy for historical dis- tribution of seagrasses. Palaeogeography, Pal- aeoclimatology, Palaeoecology 400:41–49. Vélez-Juarbe, J., and D. P. Domning. 2014a. Fossil sirenia of the west Atlantic and Caribbean Re- gion: X. Priscosiren atlantica, gen. et sp. nov. Journal of Vertebrate Paleontology 34:951– 964. Vélez-Juarbe, J., and D. P. Domning. 2014b. Fos- sil sirenia of the west Atlantic and Caribbean Region: IX. Metaxytherium albifontanum, sp. nov. Journal of Vertebrate Paleontology 34:444–464. Vélez-Juarbe, J., and D. P. Domning. 2015. Fos- sil sirenia of the west Atlantic and Carib- bean Region: XI. Callistosiren boriquen- sis, gen. et sp. nov. Journal of Verte- brate Paleontology 35:1, e885034, DOI: 10.1080/02724634.2014.885034. Vélez-Juarbe, J., D. P. Domning, and N. D. Py- enson. 2012. Iterative evolution of sympat- ric seacow (Dugongidae, Sirenia) assem- blages during the past ~26 million years. PLos ONE 7(2):e31294. doi:10.1371/journal. pone.0031294. Voorhies, M. R. 1974. The Pliocene horse Nannip- pus minor in Georgia: geologic implications. Tulane Studies in Geology and Paleontology 11:109–113. Voorhies, M. R. 1990. Vertebrate Paleontology of the Proposed Norden Reservoir Area, Brown, Cherry, and Keya Paha Counties, Nebraska. Technical Report 82-09, Division of Archeo- logical Research, University of Nebraska, Lin- coln. V+138 p. + A1–593. Vucetich, M. G., Deschamps, C. M., and M. E. Pérez. 2015. The first capybaras (Rodentia, Caviidae, Hydrochoerinae) involved in the Great American Biotic Interchange. Ameghin- iana 52:324–333. Wade, B. S., and H. Pälike. 2004. Oligocene cli- mate dynamics. Paleoceanography 19 PA4019. doi:10.1029/2004PA001042. Wade, B. S., P. N. Pearson, W. A. Berggren, and H. Pälike. 2011. Review and revision of Cenozoic planktonic foraminiferal biostratigraphy and calibration to the geomagnetic polarity and as- tronomical time scale. Earth-Science Reviews 104:111–142. ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 221 Wang, X., R. H. Tedford, and B. E. Taylor. 1999. Phylogenetic systematics of the Borophaginae (Carnivora: Canidae). Bulletin of the Ameri- can Museum of Natural History 243:1–391. Ward, L. W. 2008. Synthesis of paleontological and stratigraphic investigations at the Lee Creek Mine, Aurora, N.C. (1958-2007). Pp. 325–432 in C. E. Ray, D. J. Bohaska, I. A. Koretsky, L. W. Ward, and L. G. Barnes, eds. Geology and Paleontology of the Lee Creek Mine, North Carolina IV. Virginia Museum of Natural His- tory Special Publication 14. Ward, L. W., and G. W. Andrews. 2008. Stratigra- phy of the Calvert, Choptank, and St. Marys formations (Miocene) in the Chesapeake Bay area, Maryland and Virginia. Virginia Muse- um of Natural History Publications, Memoir 9, 170 p. Ward, L. W., and B. W. Blackwelder. 1980. Strati- graphic revision of Miocene and lower Plio- cene beds of the Chesapeake Group, middle Atlantic Coastal Plain. Contributions to Stra- tigraphy, Geological Survey Bulletin 1482–D. Ward, L. W., and B. W. Blackwelder. 1987. Late Pliocene and early Pleistocene Mollusca from the James City and Chowan River formations at the Lee Creek Mine. Pp. 113–283 in C. E. Ray, ed. Geology and Paleontology of the Lee Creek Mine, North Carolina, II. Smithsonian Contributions to Paleobiology 61. Ward, L. W., and P. F. Huddlestun. 1988. Age and stratigraphic correlation of the Raysor For- mation, late Pliocene, South Carolina. Tulane Studies in Geology and Paleontology 21:59– 75. Ward, L. W., R. H. Bailey, and J. G. Carter. 1991. Pliocene and early Pleistocene stratigraphy, depositional history, and molluscan paleobio- geography of the Coastal Plain. Pp. 274–289 in J. W. Horton and V. A. Zullo, eds. The Geol- ogy of the Carolinas. Carolina Geological So- ciety 50th Anniversary Volume, University of Tennessee Press, Knoxville. Ward, L. W., B. W. Blackwelder, G. S. Gohn, and R. Z. Poore. 1979. Stratigraphic revision of Eocene, Oligocene, and lower Miocene forma- tions of South Carolina. South Carolina Divi- sion of Geology, Geologic Notes 23:2–33. Warren, L. 1998. Joseph Leidy: The Last Man Who Knew Everything. Yale University Press, New Haven, Connecticut, 320 p. Webb, S. D., R. C. Hulbert Jr., G. S. Morgan, and H. F. Evans. 2008. Terrestrial mammals of the Palmetto Fauna (early Pliocene, latest Hemp- hillian) from the central Florida phosphate district. Pp. 293–312 in X. Wang and L. G. Barnes, eds. Geology and Vertebrate Paleon- tology of Western and Southern North Ameri- ca, Contributions in Honor of David P. Whis- tler. Natural History Museum of Los Angeles County, Science Series 41. Weems, R. E. 1999. Actinopterygian fish remains from the Fisher/Sullivan Site. Pp. 53–99 in R. E. Weems and G. J. Grimsley, eds. Early Eo- cene vertebrates and plants from the Fisher/ Sullivan Site (Nanjemoy Formation), Stafford County, Virginia. Virginia Division of Mineral Resources Publication 152. Weems, R. E., and K. M. Brown. 2017. More-com- plete remains of Procolpochelys charlestonen- sis (Oligocene, South Carolina), an occurrence of Euclastes (upper Eocene, South Carolina), and their bearing on Cenozoic pancheloniid turtle distribution and phylogeny. Journal of Paleontology 91:1228–1243. Weems, R. E., and L. M. Bybell. 1998. Geology of the Black Mingo Group (Paleocene) in the Kingstree and St. Stephen areas of South Car- olina. Pp. 9–27 in A. E. Sanders, ed. Paleobi- ology of the Williamsburg Formation (Black Mingo Group; Paleocene) of South Carolina, U.S.A. Transactions of the American Philo- sophical Society 88. Weems, R. E., and L. E. Edwards. 2001. Geology of Oligocene, Miocene, and younger deposits in the coastal area of Georgia. Georgia Geo- logic Survey, Department of Natural Resourc- es, Bulletin 131, 124 p. Weems, R. E., and L. E. Edwards. 2007a. Post- middle Miocene origin of modern landforms in the eastern Piedmont of Virginia. Stratigra- phy 4:35–48. 222 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) Weems, R. E., and L. E. Edwards. 2007b. The age and provenance of “Eschrichtius” cephalus Cope (Mammalia: Cetacea). Journal of Verte- brate Paleontology 27:752–756. Weems, R. E., and G. J. Grimsley, eds. 1999. Early Eocene Vertebrates and Plants from the Fisher/ Sullivan Site (Nanjemoy Formation) Stafford County, Virginia. Virginia Division of Mineral Resources Publication 152, Charlottesville, 159 p. Weems, R. E., and W. B. Harris. 2008. Major change in depositional style and paleoclimate in the southeastern United States at or very near the Oligocene-Miocene boundary. Inter- national Geological Congress, Abstracts, v. 33, Abstract 1323940. Weems, R. E., and J. L. Knight. 2013. A new species of Bairdemys (Pelomedusoides: Podocnemidi- dae) from the Oligocene (early Chattian) Chan- dler Bridge Formation of South Carolina, USA, and its paleobiogeographical implications for the genus. Pp. 289–303 in D. B. Brinkman, P. A. Holroyd, and J. D. Garner, eds. Morphology and Evolution of Turtles, Vertebrate Paleobiol- ogy and Paleoanthropology Series. Springer, Dordrecht. Weems, R. E., and E. M. Lemon, Jr. 1984a. Geo- logic map of the Mount Holly Quadrangle, Berkeley and Charleston Counties, South Car- olina. U.S. Geological Survey Geologic Quad- rangle map GQ–1579. Weems, R. E., and E. M. Lemon, Jr. 1984b. Geolog- ic map of the Stallsville Quadrangle, Dorches- ter and Charleston Counties, South Carolina. U.S. Geological Survey Geologic Quadrangle Map GQ–1581. Weems, R. E., and E. M. Lemon, Jr. 1988. Geolog- ic map of the Ladson Quadrangle, Berkeley, Charleston, and Dorchester counties, South Carolina. U.S. Geological Survey Map GQ– 1630. Weems, R. E., and E. M. Lemon, Jr. 1989. The Geology of the Bethera, Cordesville, Huger, and Kittredge quadrangles, Berkeley County, South Carolina. U.S. Geological Survey Mis- cellaneous Investigations Series Map I–1854. Weems, R. E., and E. M. Lemon, Jr. 1993. Geology of the Cainhoy, Charleston, Fort Moultrie, and North Charleston Quadrangles, Charleston and Berkeley Counties, South Carolina. U.S. Geological Survey Miscellaneous Investiga- tions Series Map I–1935. Weems, R. E., and E. M. Lemon, Jr. 1996. Geology of Clubhouse Crossroads and Osborn quad- rangles, Charleston and Dorchester Counties, South Carolina. U.S. Geological Survey Mis- cellaneous Investigation Map I–2491 (with text), scale 1:24,000. Weems, R. E., and W. C. Lewis. 2002. Structural and tectonic setting of the Charleston, South Carolina, region: evidence from the Tertiary stratigraphic record. Geological Society of America Bulletin 114:24–42. Weems, R. E., and A. E. Sanders. 1986. The Chan- dler Bridge Formation (Upper Oligocene) of the Charleston, South Carolina, region. Geo- logical Society of America, Centennial Field Guide, Southeastern Section:323–326. Weems, R. E., and A. E. Sanders. 2014. Oligocene pancheloniid sea turtles from the vicinity of Charleston, South Carolina, U.S.A. Journal of Vertebrate Paleontology 34:80–99. Weems, R. E., E. M. Lemon, Jr., and M. S. Nelson. 1997. Geology of the Pringletown, Ridgeville, Summerville, and Summerville Northwest 7.5-minute quadrangles, Berkeley, Charleston, and Dorchester Counties, South Carolina. U.S. Geological Survey Miscellaneous Investiga- tions map 2502, scale 1:24,000, text. Weems, R. E., W. B. Harris, A. E. Sanders, and L. E. Edwards. 2006. Correlation of Oligocene sea level cycles between western Europe and the southeastern United States. Pp. 205–206 in G. Camoin, A. Droxler, C. Fulthorpe, and K. Miller, eds. Sea Level Changes: Records, Pro- cesses, and Modeling: SEALAIX’06 (25-29 September, 2006, Presqué de Giens, France), Abstracts. Weems, R. E., E. M. Lemon, Jr., L. McCartan, L. M. Bybell, and A. E. Sanders. 1982. Recogni- tion and formalization of the Pliocene “Goose Creek Phase” in the Charleston, South Caro- ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 223 lina, area. Contributions to Stratigraphy, U.S. Geological Survey Bulletin 1529-H:137–148. Weems, R. E., E. M. Lemon, Jr., M. S. Nelson, G. S. Gohn, and B. B. Houser. 1987. Detailed sections from auger holes and outcrops in the Pringletown, Ridgeville, Summerville North- west, and Summerville quadrangles, South Carolina. U.S. Geological Survey Open File Report 87-524:1–89. Weems, R. E., L. B. Albright III, L. M. Bybell, D. J. Cicimurri, L. E. Edwards, W. B. Harris, W. C. Lewis, J. E. Osborne, A. E. Sanders, and J. M. Self-Trail. 2016. Stratigraphic revision of the Cooper Group and the Chandler Bridge and Edisto formations in the Coastal Plain of South Carolina. South Carolina Geology 49:1–24. Weishampel, D. B., and L. Young. 1996. Dinosaurs of the East Coast. The Johns Hopkins Univer- sity Press, Baltimore, Maryland, 275 p. Werdelin, L. 1985. Small Pleistocene felines of North America. Journal of Vertebrate Paleon- tology 5:194–210. Westgate, J. W. 2001. Paleoecology and biostra- tigraphy of marginal marine Gulf Coast Eo- cene vertebrate localities. Pp. 263–296 in G. F. Gunnell, ed. Eocene Biodiversity: Unusual Occurrences and Rarely Sampled Habitats. Topics in Geobiology 18. Springer Science + Business Media, New York. White, J. A. 1968. A new porcupine from the mid- dle Pleistocene of the Anza Borrego-Desert of California with notes on mastication in Co- endu and Erethizon. Los Angeles County Mu- seum Contributions in Science 136:1–15. White, J. A. 1970. Late Cenozoic porcupines (Mammalia, Erethizontidae) of North Ameri- ca. American Museum Novitates 2421:1–15. White, J. A. 1991. A new Sylvilagus (Mammalia: Lagomorpha) from the Blancan (Pliocene) and Irvingtonian (Pleistocene) of Florida. Journal of Vertebrate Paleontology 11:243–246. White, R. S., and G. S. Morgan. 2005 Arizona Blancan vertebrate faunas in regional perspec- tive. Pp. 117–138 in R. D. McCord, ed. Ver- tebrate Paleontology of Arizona. Mesa South- west Museum Bulletin 11. Mesa, AZ. Whitmore, F. C., Jr., and J. A. Kaltenbach. 2008. Neogene cetaceans of the Lee Creek phos- phate mine, North Carolina. Virginia Museum of Natural History Special Paper 14:181–269. Whitmore, F. C., Jr., and A. E. Sanders. 1976. Re- view of the Oligocene Cetacea. Systematic Zoology 25:304–320. Williams, M., A. M. Haywood, E. M. Harper, A. L. A. Johnson, T. Knowles, M. J. Leng, D. J. Lunt, B. Okamura, P. D. Taylor, and J. Zalasie- wicz. 2009. Pliocene climate and seasonality in North Atlantic shelf seas. Philosophical Trans- actions of the Royal Society A 367:85–108. Wilson, R. W. 1935. A new species of porcupine from the later Cenozoic of Idaho. Journal of Mammalogy 16:220–222. Winkler, A. J., and W. Gose. 2003. Mammalian fauna and paleomagnetics of the middle Ir- vingtonian (early Pleistocene) Fyllan Cave and Kitchen Door localities, Travis County, Texas. Pp 215–261 in B. W. Schubert, J. I. Mead, and R. W. Graham, eds. Ice Age Cave Faunas of North America. Indiana University Press, Bloomington and Indianapolis. Wood, H. E. II, W. Chaney, J. Clark, E. H. Colbert, G. L. Jepsen, J. B. Reeside Jr., and C. Stock. 1941. Nomenclature and correlation of the North American Tertiary. Bulletin of the Geo- logical Society of America 52:1–48. Woodburne, M. O., ed. 1987. Cenozoic Mammals of North America, Geochronology and Bio- stratigraphy. University of California Press, Berkeley, 336 p. Woodburne, M. O., ed. 2004. Late Cretaceous and Cenozoic Mammals of North America, Bio- stratigraphy and Geochronology. Columbia University Press, New York, 391 p. Woodburne, M. O. 2010. The Great American Bi- otic Interchange: dispersals, tectonics, climate, sea level, and holding pens. Journal of Mam- malian Evolution 17:245–264. Wright, D. B. 1995. Tayassuidae of the Irvingto- nian Leisey Shell Pit Local Fauna, Hillsbor- ough County, Florida. Pp. 603–620 in R. C. Hulbert, G. S. Morgan, and S. D. Webb, eds. 224 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) Paleontology and Geology of the Leisey Shell Pits, Early Pleistocene of Florida, Bulletin of the Florida Museum of Natural History 37, Part II. Wright, D. B. 1998. Tayassuidae. Pp. 389–401 in C. M. Janis, K. M. Scott, and L. L. Jacobs, eds. Evolution of Tertiary Mammals of North America, Volume 1: Terrestrial Carnivores, Ungulates, and Ungulatelike Mammals. Cam- bridge University Press, Cambridge. Zachos, J. C., B. P. Flower, and H. Paul. 1997. Orbitally paced climate oscillations across the Oligocene/Miocene boundary. Nature 388:567–570. Zachos, J., M. Pagini, L. Sloan, E. Thomas, and K. Billups. 2001. Trends, rhythms, and aberra- tions in global climate 65 Ma to Present. Sci- ence 292:686–692. Zalmout, I. S., and P. D. Gingerich. 2012. Late Eocene sea cows (Mammalia: Sirenia) from Wadi al Hitan in the western desert of Fayum, Egypt. University of Michigan Papers on Pale- ontology 37, 158 p. Zurita, A. E., D. D. Gillette, F. Cuadrelli, and A. A Carlini. 2018. A tale of two clades: compara- tive study of Glyptodon Owen and Glyptother- ium Osborn (Xenarthra, Cingulata, Glypto- dontidae). Geobios 51(3):247–258. ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 225 APPENDIX 1 All known non-marine mammalian taxa from South Carolina including age and unit from which the remains were recovered (or thought to have originated). *Denotes South Carolina as type locality; † Denotes specimens examined by LBA but retained in a private collection; Boldface denotes taxon previously unreported from South Carolina; “APBs”, “Ashley Phosphate Beds”; ARD, Ardis Local Fauna; “BPBs”, “Beaufort Phosphate Beds”; BMF, Black Mingo Fauna; CAM, Camelot Local Fauna; CB, Chandler Bridge Formation; CLP, Clapp Creek locality; COOS, Coosawhatchie Formation; CR, Cooper River; CRO, Crowfield Local Fauna; DUP, Duplin Formation; EB, Edisto Beach; EBEN, Ebenezer Formation; EF, Edisto Formation; GCL, Goose Creek Limestone; HAR, Harleyville Formation; IRV, Irvingtonian NALMA; JMS, “Jamestown beds”; LAD, Ladson Formation; MB, Myrtle Beach; PEN, Penholoway Formation; RAY, Raysor Formation; RLF, Ridgeville Local Fauna; RLB, Rancholabrean NALMA; TMH, Ten Mile Hill Formation; WAC, Waccamaw Formation; Wa0, Wasatchian 0 (earliest Wasatchian) NALMA; WAN, Wando Formation; WD, Walrus Ditch Local Fauna. Compiled from Leidy (1860, 1877); Hay (1923); Allen (1926); Ray (1965, 1967); Ray et al. (1968); Ray and Sanders (1984); Roth and Laerm (1980); Bentley et al. (1994); McDonald et al. (1996); Downing and White (1995); Schoch (1985, 1998); Sanders (2002); Kohn et al. (2005); Fields et al. (2012); and this paper. Taxon Fauna, Stratigraphic Unit, or Locality North American Land Mammal Age Marsupialia Didelphis virginiana ARD, CAM, CRO, EB Late Irvingtonian–Late Rancholabrean Eutheria incertae sedis *Mingotherium holtae BMF Late Tiffanian (Ti5) Xenarthra Megalonyx leptostomus CR, WD Early(?)–Late Blancan Megalonyx jeffersonii ARD, CAM, CRO, EB, LAD, WAN Late Irvingtonian–Late Rancholabrean Eremotherium laurillardi EB, LAD, PEN, TMH Irvingtonian-Late Rancholabrean Eremotherium eomigrans WD, RLF Late Blancan Paramylodon harlani CR, EB, ?WAN Late Irvingtonian(?)–Late Rancholabrean Glyptotherium texanum ? ?Middle Blancan Glyptotherium floridanum EB Late Rancholabrean Pachyarmatherium leiseyi WD Late Blancan Holmesina septentrionalis ARD, CAM, CRO, EB, TMH Late Irvingtonian–Late Rancholabrean Holmesina floridanus CLP, WD, RLF Late Blancan Dasypus bellus ARD, CAM, CRO, EB, TMH, WD Late Blancan–Late Rancholabrean Taeniodonta Ectoganus gliriformis lobdelli BMF Late Tiffanian (Ti5) Lagomorpha Sylvilagus floridanus ARD, CRO, EB Rancholabrean Sylvilagus palustris ARD, CAM, CRO, TMH Late Irvingtonian–Late Rancholabrean Sylvilagus webbi WD Late Blancan Lepus sp. CAM Late Irvingtonian Rodentia Sciurus carolinensis ARD, CAM, CRO Late Irvingtonian–Late Rancholabrean Sciurus sp. CRO Rancholabrean Ictidomys tridecemlineatus ARD, CRO Rancholabrean 226 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) APPENDIX 1 (Continued) Taxon Fauna, Stratigraphic Unit, or Locality North American Land Mammal Age Glaucomys volans ARD, CRO Rancholabrean Castoroides dilophidus CRO, EB, WAN Rancholabrean Castor canadensis ARD, EB, CRO, TMH, WAN Rancholabrean Thomomys sp. CAM Late Irvingtonian Oryzomys palustris ARD Rancholabrean Oryzomys sp. CRO Rancholabrean Peromyscus sp. ARD, CRO Rancholabrean Sigmodon hispidus CRO Rancholabrean Sigmodon bakeri CAM Late Irvingtonian Neotoma floridana ARD, CRO Rancholabrean Allophaiomys pliocaenicus WD Late Blancan Microtus pennsylvanicus ARD, CRO Rancholabrean Microtus pinetorum ARD, CRO Rancholabrean Microtus ochrogaster CRO Rancholabrean Synaptomys cooperi ARD, CRO Rancholabrean Synaptomys australis ARD, CRO Rancholabrean Neofiber alleni ARD, CRO, TMH, WAN Rancholabrean Neofiber ?alleni CAM Late Irvingtonian Ondatra idahoensis CLP Late Blancan Ondatra zibethicus ARD, CRO Rancholabrean Erethizon ?bathygnathum WD Rancholabrean Erethizon dorsatum CRO, EB Rancholabrean Phugatherium dichroplax DUP Middle Blancan *Neochoerus pinckneyi GCL, DUP, WAC, WD, RLF, CAM, TMH, WAN, ARD, EB Early Blancan–Rancholabrean Insectivora Sorex sp. cf. S. longirostris ARD, CRO Rancholabrean Sorex sp. cf. S. arcticus CRO Rancholabrean Sorex sp. cf. S. palustris CRO Rancholabrean Microsorex hoyi CRO Rancholabrean Blarina brevicauda ARD, CRO Rancholabrean Blarina sp. CRO Rancholabrean Cryptotis parva CRO Rancholabrean Scalopus aquaticus ARD, CRO Rancholabrean Scalopus sp. CAM Late Irvingtonian Condylura cristata ARD, CRO Rancholabrean Carnivora Didymictis proteus JMS Clarkforkian–earliest Wasatchian Leopardus amnicola WAN Rancholabrean ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 227 APPENDIX 1 (Continued) Taxon Fauna, Stratigraphic Unit, or Locality North American Land Mammal Age Miracinonyx ?trumani EB Late Rancholabrean Miracinonyx inexpectatus CAM, PEN, WAC Irvingtonian Smilodon fatalis ARD, CAM, EB Late Irvingtonian–Late Rancholabrean Smilodon gracilis WD Late Blancan Panthera atrox EB Late Rancholabrean Panthera onca augusta EB Late Rancholabrean Puma concolor EB Late Rancholabrean Lynx rufus ARD, CRO, EB Rancholabrean †Chasmaporthetes ossifragus ? Late Blancan–Early Irvingtonian Borophagus hilli RAY Early Blancan Canis lepophagus WD Late Blancan Canis armbrusteri CAM Late Rancholabrean Canis dirus ARD, EB, WAN Rancholabrean Canis latrans MB Late Irvingtonian Urocyon cinereoargenteus ARD, CAM, EB, CRO Late Irvingtonian–Late Rancholabrean Arctodus pristinus LAD, WD Late Blancan–Late Irvingtonian Tremarctos floridanus ARD, CRO, EB, MB Rancholabrean Ursus americanus CRO, WAN Rancholabrean Lontra canadensis ARD, CRO Rancholabrean Spilogale putorius ARD, CRO Rancholabrean Mephitis mephitis ARD, CRO Late Rancholabrean Conepatus robustus ARD Late Rancholabrean Neovison vison ARD, CRO Rancholabrean Procyon lotor ARD, CAM, EB, CRO Late Irvingtonian–Late Rancholabrean Procyon sp. WD Late Blancan Cetungulata *Phenacodus grangeri mccollumi BMF Late Tiffanian (Ti5) Perissodactyla †?Megacerops sp. HAR Late Chadronian Anchippus texanus EF? Late Arikareean Nannippus peninsulatus CLP Late Blancan Nannippus sp. WD Late Blancan Cormohipparion sp. CLP Late Blancan Equus “complicatus” ARD Late Rancholabrean Equus “fraternus” “APBs” Rancholabrean Equus “leidyi” “APBs” Rancholabrean Equus sp. CAM, CLP, CRO, EB, LAD, TMH, WAN, WD, RLF Blancan–Late Rancholabrean 228 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) APPENDIX 1 (Continued) Taxon Fauna, Stratigraphic Unit, or Locality North American Land Mammal Age Equus “littoralis” “APBs” Rancholabrean Tapirus sp. RLF Late Blancan Tapirus haysii TMH, WAC, WD Late Blancan–Early Rancholabrean Tapirus veroensis ARD, CAM, CLP, CRO, EB, LAD, WAN Late Irvingtonian–Rancholabrean Subhyracodon mitis HAR Late Chadronian Aphelops ?malacorhinus ?COOS or ?EBEN ?Barstovian–Hemphillian Teleoceras ?guymonense ?COOS or ?EBEN ?Barstovian–Late Hemphillian Artiodactyla *Daeodon mento EF Late Arikareean Mylohyus fossilis ARD, MB, EB Late Rancholabrean Mylohyus sp. CAM, CRO, WD Late Irvingtonian–Rancholabrean Platygonus compressus CAM Late Irvingtonian Perchoerus sp. CB Late Arikareean Hemiauchenia macrocephala CAM, CRO, WD Late Blancan–Rancholabrean Palaeolama mirifica ARD, CAM, CRO, EB Late Irvingtonian–Late Rancholabrean Rangifer tarandus CRO, WAN Rancholabrean Cervalces scotti WAN Rancholabrean Cervus elaphus CLP, CRO, EB Rancholabrean Odocoileus virginianus ARD, CAM, CRO, EB, WD Late Blancan–Recent Bison antiquus ARD, EB Late Rancholabrean Bison sp. CRO Rancholabrean Bootherium bombifrons ARD, MB, WAN Rancholabrean Proboscidea Gomphothere “APBs”, “BPBs” Rancholabrean Rhynchotherium falconeri WD, RLF Late Blancan Cuvieronius hyodon TMH, WAN Rancholabrean Mammut americanum ARD, CRO, EB, WAN Rancholabrean Mammuthus columbi ARD, EB, WAN Rancholabrean ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 229 APPENDIX 2 Marine mammalian taxa from South Carolina, including age and unit from which the remains were recovered (or thought to have been recovered). “ ” denotes doubtful generic assignment; *Denotes South Carolina as type locality; GCL = Goose Creek Limestone; TBF = Tupelo Bay Formation; UOPD = Unnamed Offshore Pleistocene Deposits. NOTE: Not included in this list are dozens of yet-to-be- described cetacean taxa including new representatives of the Protocetidae, Basilosauridae, Xenorophidae, Agorophiidae, “Waipatiidae,” “Squalodontidae,” and Eomysticetidae, in addition to several new sirenian taxa of Protosirenidae and early members of the Dugongidae. Taxon Formation Stage/Age Cetacea Protocetidae Georgiacetus sp. cf. G. vogtlensis Santee Limestone Upper Lutetian/Lower Bartonian *Carolinacetus gingerichi Cross Mbr., TBF Upper Bartonian *Tupelocetus palmeri Cross Mbr., TBF Upper Bartonian Basilosauridae Basilosaurus cetoides Pregnall Mbr., TBF Lower Priabonian Basilosaurus sp. Pregnall Mbr., TBF Lower Priabonian Zygorhiza kochii Pregnall Mbr., TBF Lower Priabonian *Chrysocetus healyorum Pregnall Mbr., TBF Lower Priabonian *Dorudon serratus ?Pregnall Mbr., TBF Lower Priabonian Basilosauridae indet. Parkers Ferry Fm. Upper Priabonian Basilosaurus sp. ?Harleyville Fm. Upper Priabonian/Lower Rupelian Odontoceti Xenorophidae *Xenorophus sloanii Ashley Fm. Upper Rupelian Albertocetus meffordorum Ashley Fm. Upper Rupelian *Inermorostrum xenops Ashley Fm. Upper Rupelian *Cotylocara macei Chandler Bridge Fm. Upper Chattian *Echovenator sandersi Chandler Bridge Fm. Upper Chattian Ashleycetidae *Ashleycetus planicapitis Ashley Fm. Upper Rupelian Agorophiidae *Agorophius pygmaeus Ashley and Chandler Bridge Fms. Upper Rupelian/Upper Chattian Family indet. *Ediscetus osbornei Ashley Fm. Upper Rupelian “Squalodon” tiedemani (“Genus Y”?) Ashley and Chandler Bridge Fms. Upper Rupelian/Upper Chattian “Squalodon” atlanticus Chandler Bridge Fm. Upper Chattian Eurhinodelphinidae “Rhabdosteus latiradix” ?Marks Head Fm. Burdigalian 230 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) APPENDIX 2 (Continued) Taxon Formation Stage/Age Ziphiidae *Anoplanassa forcipata ?Marks Head Fm. Burdigalian *Dioplodon proprops ?Marks Head Fm. Burdigalian *Choneziphius chonops ?Marks Head Fm. Burdigalian *Choneziphius liops ?Marks Head Fm. Burdigalian *Choneziphius tracops ?Marks Head Fm. Burdigalian *Choneziphius macrops ?Marks Head Fm. Burdigalian *Eboroziphius coelops ?Marks Head Fm. Burdigalian *Ceterhinops longifrons ?Goose Creek Limestone Zanclean Tusciziphius crispus ?Goose Creek Limestone Zanclean Physeteridae *Dinoziphius carolinensis (= Physeter carolinensis) ?Marks Head Fm. Burdigalian Physeterula sp. ?Waccamaw Fm. Lower Pleistocene Scaldicetus sp. ?Waccamaw Fm. Lower Pleistocene Physeter macrocephalus UOPD (Edisto Beach) Upper Pleistocene–Holocene Delphinidae Tursiops truncatus UOPD (Edisto Beach) Upper Pleistocene–Holocene Stenella sp. UOPD (Edisto Beach) Upper Pleistocene–Holocene Pseudorca crassidens UOPD (Edisto Beach) Upper Pleistocene–Holocene Mysticeti *Coronodon havensteini Ashley and Chandler Bridge Fms. Upper Rupelian/Upper Chattian *Micromysticetus rothauseni Ashley Fm. Upper Rupelian *Eomysticetus whitmorei Chandler Bridge Fm. Upper Chattian *Eomysticetus carolinensis Chandler Bridge Fm. Upper Chattian Balaenopteridae Eschrichtius sp. ?Waccamaw Fm. Lower Pleistocene Gen. et sp. indeterminate Goose Creek Limestone Upper Zanclean Pinnipedia Odobenidae Ontocetus emmonsi GCL, Raysor, Waccamaw Fms. Upper Zanclean–Lower Pleistocene Odobenus rosmarus ?Wando Fm. Upper Pleistocene Odobenus sp. Ten Mile Hill Fm. Upper Middle Pleistocene Phocidae Monatherium sp. ?Ebenezer Fm. ?Tortonian Phocanella pumila ?Wabasso beds, ?GCL Zanclean Callophoca obscura ?Wabasso beds, ?GCL Zanclean Erignathus barbatus Socastee Fm. Upper Pleistocene ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 231 APPENDIX 2 (Continued) Taxon Formation Stage/Age Monachus tropicalis Socastee Fm., UOPD Upper Pleistocene Halichoerus grypus UOPD (Edisto Beach) Upper Pleistocene–Lower Holocene Sirenia Protosirenidae Protosiren sp. Cross Mbr., TBF Upper Bartonian Protosiren sp. Pregnall Mbr., TBF Lower Priabonian Dugongidae Eotheroides sp. Pregnall Mbr, TBF Lower Priabonian Priscosiren atlantica Ashley Fm. Upper Rupelian *Stegosiren macei Ashley and ?Chandler Bridge Fms. Upper Rupelian/?Upper Chattian Crenatosiren olseni Ashley and Chandler Bridge Fms. Upper Rupelian/Upper Chattian Metaxytherium albifontanum ?Ashley Fm., Chandler Bridge Fm. ?Upper Rupelian/Upper Chattian Metaxytherium sp. Ashley Fm. Upper Rupelian *Dioplotherium manigaulti Chandler Bridge Fm. Upper Chattian *“Halitherium” alleni Uncertain Uncertain Corystosiren sp. Uncertain Uncertain Trichechidae Trichechus manatus1 UOPD (Edisto Beach) Upper Pleistocene–Holocene 1Includes *Manatus antiquus Leidy, 1856 and *Manatus inornatus Leidy, 1873 (see Domning, 1989b). 232 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) APPENDIX 3 Recognized Faunas, Local Faunas, faunal assemblages, and/or stratigraphic units from which known non- marine mammalian taxa from SC originate, plus associated age and NALMA, with faunas from Florida mentioned throughout text included for temporal comparison. * denotes the primarily Rancholabrean aspect of the mammals from Edisto Beach, but see text for discussion regarding additional taxa of non- Rancholabrean age. South Carolina Florida Epoch NALMA Ashley River Phosphate Beds Mixed faunas See text Beaufort Phosphate Beds Mixed faunas See text Cooper River Mixed faunas See text Clapp Creek locality Mixed faunas See text Ardis Local Fauna Latest Pleistocene Late Rancholabrean Edisto Beach Fauna Mixed faunas Rancholabrean* Myrtle Beach Fauna Mixed faunas Rancholabrean* Crowfield Local Fauna Late Pleistocene Rancholabrean Wando Formation Late Pleistocene Rancholabrean Socastee Formation Late Pleistocene Rancholabrean Ten Mile Hill Formation Late Mid. Pleistocene Earliest Rancholabrean Coleman 2A LF Mid. Pleistocene Late Irvingtonian Camelot Local Fauna Mid. Pleistocene Late Irvingtonian Ladson Formation Mid. Pleistocene Late Irvingtonian Penholoway Formation Late Earl. Pleistocene Middle Irvingtonian Leisey Shell Pit LF Early Pleistocene Late Early Irvingtonian Haile 16A LF Early Pleistocene Earliest Irvingtonian Ridgeville Local Fauna Early Pleistocene Latest Blancan-E. Irv. Upper Waccamaw Formation Early Pleistocene Late Blancan Inglis 1A LF Early Pleistocene Latest Blancan De Soto Shell Pit Early Pleistocene Latest Blancan Haile 7C LF Early Pleistocene Middle late Blancan Withlacoochee River 1A Haile 15A LF Early Pleistocene Early Late Blancan Santa Fe River 1 LF Early Pleistocene Early Late Blancan Macasphalt Shell Pit LF Early Pleistocene Early Late Blancan Lower Waccamaw Formation Early Pleistocene Early Late Blancan Walrus Ditch Local Fauna Early Pleistocene Early Late Blancan Duplin Formation Late Pliocene Middle Blancan Raysor Formation Middle Pliocene Early Blancan Goose Creek Limestone Middle Pliocene Early Blancan Wabasso beds Palmetto Fauna Early Pliocene Latest Hemphillian Edisto Formation Latest Oligocene Late Arikareean (Ar3) ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 233 APPENDIX 3 (Continued) South Carolina Florida Epoch NALMA Chandler Bridge Formation White Springs LF Late Oligocene Late Arikareean (Ar3) Ashley Formation Early Oligocene Early Arikareean (Ar2) Harleyville Formation Latest Eocene Late Chadronian Parkers Ferry Formation Late Eocene Late Chadronian Tupelo Bay Formation Late Eocene Late Duchesnean-Early Chadronian Santee Limestone Middle Eocene Late Uintan “Jamestown beds” (= Chicora Mbr., Williamsburg Formation) Latest Paleocene-Earliest Eocene Clarkforkian-Earliest Wasatchian Black Mingo Fauna Late Paleocene Late Tiffanian (Ti5) 234 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) APPENDIX 4 Molluscan Fauna of the Ashley Formation The Ashley Formation contains abundant well-preserved valves of a small, thin-shelled, finely ribbed pectenid-like bivalve that often occurs in densely packed layers. Tuomey (1848) reported this lo- cally abundant mollusc as Pecten calvatus, an Eocene form, and one of three molluscan taxa upon which he based his referral of the Charleston marl beds to the Eocene. Dall (1894:301), however, noted that “The pecten referred to [by Tuomey] is not Pecten calvatus” and he expressed doubt that the sediments in which it occurred were of Eocene age, as indeed they are not. Dall (1894, 1896, 1903) considered the Ashely Formation to be Miocene in age, and therefore used Miocene names for the species. Dall (1894:300) re- ported “about twenty species” of molluscs from the Ashley Formation from an 1877 shipwreck in Maine that carried phosphate from “the east bank of the Ashely River about 10 miles from Charleston, SC.” That bivalve has recently been identified by M. Campbell as Eburneopecten subminutus (Aldrich, 1903). During his survey of the geology of South Carolina in the first decade of the 20th Century, State Ge- ologist Earle Sloan collected molluscs from the Ashley Formation at several localities near Charleston. On some occasions he was accompanied by T. W. Vaughn (USGS), who may well have provided many of the identifications of the specimens. In 1907 Sloan and Vaughn collected specimens from the Ashley Forma- tion in the Lambs and Ingleside marl pits and from an unspecified location near the Ingleside railroad sta- tion; but because of the nature of the publication, Sloan did not include detailed faunal lists in his Mineral Localities of South Carolina (Sloan, 1908), and for many years thereafter the molluscan fauna of this unit remained virtually unknown. Sloan’s extensive collection of geological and paleontological material from South Carolina, Virginia, and the Gulf Coast is housed in The Charleston Museum. C. W. Johnson (1931) described five new wentletraps from the vicinity of Charleston. The type specimens of four of them, Epitonium chamberlaini, E. charlestonensis, E. cooperensis, and E. subexpan- sum (now Sthenorytis subexpansum (Johnson, 1931)) were collected by E. Bernham Chamberlain of The Charleston Museum on a small island of dredgings of Ashley Formation sediments from the Cooper River opposite its west bank approximately 1.5 miles above the Etiwan Fertilizer works, north of Charleston (Johnson, 1931). The label with the holotype of the fifth taxon, E. raveneli Johnson, 1931 (ChM PI26233 [Ravenel No. 11 30]), discovered by Johnson in the Edmund Ravenel collection, states that the specimen is from the “Eocene” of “St. Johns Parish,” Berkeley County, indicating that it was collected from the Ashley Formation, then thought to be of Eocene age. The holotype of Sthenorhytis mazyckii (Dall, 1896), originally described by Dall as Scala (Sthenorhytis) mazyckii and considered Miocene age, was collected by W. G. Mazyck at Cainhoy, Berkeley County; but a specimen of the same taxon collected from the Ash- ley Formation at The Charleston Museum’s Chandler Bridge excavation site (Sanders, 1980; see below) demonstrates this taxon to be a late Rupelian (early Oligocene) form. In concluding his report of these taxa, Johnson (1931:10) noted that 35 species of foraminifera from matrix removed from the specimens were identified as Eocene forms by J. A. Cushman, “thus definitely determining the age of this formation.” But as detailed above, overwhelming evidence now proves the Ashley Formation to be of Oligocene age. More recently, extensive studies of unpublished collections of specimens from the Ashley For- mation at The Charleston Museum and the SCSM have been conducted by Dr. Matthew Campbell. In his assessment he noted that calcitic taxa were preserved as the original shell, but aragonitic taxa are preserved as internal or external molds resulting in identification only to the genus level. He also noted that earlier literature that published notifications of molluscs from the Ashley Formation often assigned Eocene through Miocene ages to this unit. In turn, these papers used species names that are now known to be restricted to different ages, including the Cretaceous to Eocene through Pliocene. The species with affinities listed below are taken from the published literature. ALBRIGHT ET AL.: Cenozoic vertebrate biostratigraphy of South Carolina 235 APPENDIX 4 (Continued) As Campbell further noted (pers. com. to LBA, August 2019), “examination of the original speci- mens is needed to determine whether or not sufficient morphological details are preserved to assign more appropriate names …” The following table provides his updated assessment of the molluscs from this unit. This represents the second extensive list of upper Rupelian molluscs from the Atlantic coast, subse- quent to the fauna from the Lower River Bend Formation of North Carolina (Rossbach and Carter, 1991). Bivalves Yoldia sp. Anomia sp. aff. A. simplex Astarte sp. aff. A. thomasii Nuculana sp. aff. N. flexuosa Anomia jugosa Lucina sp. Nuculana sp. Anomia sp. Cardiidae sp. Brachidontes mississippiensis Pododesmus sp. aff. P. philippi Cardium sp. Crenella sp. Ostrea sp. aff. O. carolinensis Dinocardium sp. aff. D. taphrium Modiolus sp. Ostrea sp. aff. O. podagrina Nemocardium? sp. Glycymeris sp. Ostrea sp. aff. O. queteleti Solen sp. Pteria sp. cf. P. argentea Ostrea sp. aff. O. thomasii Tellina sp. Aequipecten cocoana Ostrea sp. Antigona (Artena) undulata Amusium sp. “Ostrea” n. sp. Antigona (A.) n. sp. aff. A. lamellacea Chlamys sp. Pycnodonte paroxis Antigona (A.) sp. Eburneopecten subminutus Gryphaeostrea sp. aff. G. plicatella Callista sp. Aequipecten sp. cf. A. acanikos Phacoides? sp. Veneridae sp. Pecten sp. aff. P. humphreysii Phacoides sp. aff. P. contractus Anomalocardia? sp. Pecten sp. Cyclocardia sp. aff. C. granulata Corbula sp. Pseudamussium sp. Cyclocardia sp. aff. C. castrana Gastrochaena sp. Pseudamussium sp. aff. P. claibornense Cardita sp. Panopea sp. aff. P. elongata Plicatula? sp. Astarte sp. Panopea sp. Spondylus sp. Astarte sp. aff. A. vacina Thracia? sp. Gastropods Modulus? sp. Malea sp. aff. M. camura Ecphora sp. Turitella sp. cf. T. tampae Ficus sp. Levifusus? sp. Turitella sp. aff. T. bowenae Cirsotrema sp. Lyria sp. Calyptraea (Trochita) sp. cf. C. (T.) aperta Sthenorytis mazyckii Athleta sp. aff. A. petrosa Xenophora sp. cf. X. conchyliophora Sthenorytis subexpansum Athelta sp. Xenophora sp. Epitonium chamberlaini Olivella sp. Apiocypraea sp. Epitonium charlestonensis Prunum sp. Naticid sp. Epitonium raveneli Mitra sp. Galeodea petersoni Epitonium cooperensis Pyropsis sp. Galeodea sp. Chicoreus sp. aff. C. mississippiensis Conus sp. 236 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 57(2) APPENDIX 4 (Continued) Scaphopods Dentalium sp. Dentalium sp. aff. D. attenuatum Cephalopods Aturia alabamensis ABSTRACT INTRODUCTION METHODS AND MATERIALS A NOTE ON “HOBBY COLLECTING” VERTEBRATE FOSSILS IN SOUTH CAROLINA HISTORICAL BACKGROUND CENOZOIC VERTEBRATE FOSSIL-BEARING BEDS OF THE SOUTH CAROLINA COASTAL PLAIN Paleocene Series Eocene Series Note on the stratigraphic position of the archaeocete Dorudon serratus Oligocene Series Note on biostratigraphic correlations of Paleogene cetaceans from South Carolina Miocene Series Pliocene Series Pleistocene Series ADDITIONS TO THE CENOZOIC MAM- MALIAN FAUNA OF SOUTH CAROLINA MARSUPIALIA XENARTHRA LAGOMORPHA RODENTIA EULIPOTYPHLA CARNIVORA PERISSODACTYLA ARTIODACTYLA SIRENIA PROBOSCIDEA SUMMARY AND CONCLUSIONS ACKNOWLEDGEMENTS LITERATURE CITED APPENDIX 1 APPENDIX 2 APPENDIX 3 APPENDIX 4