Paludicola vol 8, issue 1 September 2010

PALUDICOLA SCIENTIFIC CONTRIBUTIONS of the ROCHESTER INSTITUTE OF VERTEBRATE PALEONTOLOGY

VOLUME 8 NUMBER 1

27 SEPTEMBER 2010 ISSN 1091-0263

PALUDICOLA SCIENTIFIC CONTRIBUTIONS OF THE ROCHESTER INSTITUTE OF VERTEBRATE PALEONTOLOGY Editors William W. Korth Rochester Institute of Vertebrate Paleontology 265 Carling Road Rochester, NY 14610 Phone (585) 482-0203 e-mail: [email protected] Judy A. Massare Department of Earth Sciences State University of New York, College at Brockport Brockport, NY 14420 Phone (585) 395-2419: Fax (585) 395-2416 e-mail: [email protected]

EDITORIAL POLICY Paludicola is published under the auspices of the Rochester Institute of Vertebrate Paleontology (RIVP). It is intended to serve as an “in-house” publication for students and professional paleontologists who are not affiliated with institutions that provide such a publication. All manuscripts considered for publication in Paludicola are expected to be original and have not been published elsewhere in any other form or are not being considered for publication anywhere else. This journal is intended as an outlet for papers on any aspect of Vertebrate Paleontology. Paludicola is published twice annually (Fall and Spring). All submitted manuscripts are subject to review by external qualified specialists and the editorial staff of Paludicola. The format of all submitted manuscripts should follow that of the Journal of Vertebrate Paleontology (printed in the first issue of each year). Questions about other editorial policies will be addressed by the editorial staff upon request. Please contact the editorial staff for prices of subscriptions and back issues (prices are subject to change). Back issues of volumes 1 – 6 can be purchased from PaleoPublications Inc. Prices and listings are available on-line at www.paleopubs.com.

Copyright  2010 by the Rochester Institute of Vertebrate Paleontology

Paludicola 8(1):1-7 September 2010 © by the Rochester Institute of Vertebrate Paleontology

FIRST REPORT OF FOSSIL AMPHIUMA (AMPHIBIA: CAUDATA: AMPHIUMIDAE) FROM SOUTH CAROLINA, USA JAMES L. KNIGHT 1 AND DAVID J. CICIMURRI 2 1

South Carolina State Museum, 301 Gervais Street, Columbia, South Carolina 29202 2 Campbell Geology Museum, Clemson University, Clemson, South Carolina, 29634

ABSTRACT Twenty-five fossil Amphiuma vertebrae have been collected from five sites in eastern South Carolina. These specimens constitute the first fossil records of this amphibian from the state and the northernmost record of the genus within the Atlantic Coastal Plain. Eleven specimens from one site in Dorchester County are assigned to the late Blancan North American Land Mammal Age (NALMA) and represent the oldest Pleistocene record (Gelasian Stage) of the genus. A vertebra from a second Dorchester County site was collected from strata thought to have been deposited during the middle Pleistocene (Ionian Stage), within the late Irvingtonian NALMA. Twelve specimens from two exposures of the Wando Formation represent middle Rancholabrean NALMA occurrences within the late Pleistocene (70,000 to 130,000 Ka; Tarantian Stage). A single specimen from Colleton County represents a new addition to the Edisto Beach faunal assemblage, which contains an assortment of fossil taxa ranging in age from late Miocene to late Pleistocene. All of the specimens reported herein are tentatively referred to Amphiuma means based on trunk vertebra morphology and Recent distributions of the three extant species.

fauna, the Ardis Local Fauna (LF), was collected at the Giant Cement quarry, just north of Harleyville, Dorchester County, from solution cavities in the upper Eocene Tupelo Bay Formation that were infilled with overlying fluvial sands and clays. The Ardis LF contains a wide variety of mammalian taxa (Bentley et al., 1994), turtles (Bentley and Knight, 1993, 1998), and birds (Chandler and Bentley, 2007). PlioPleistocene elasmobranch taxa are also receiving close analysis (i.e., Cicimurri and Knight 2009; Cicimurri and Knight unpublished data). Local faunas representing the late Blancan (lowermost Pleistocene), late Irvingtonian (middle Pleistocene), and Rancholabrean (middle to late Pleistocene) North American Land Mammal Ages (NALMA), in the collections of the South Carolina State Museum, are in various stages of study and will be reported on in the future. Although a considerable number of fossil collections have been made, a great deal of research remains to be conducted in order to claim even a partial understanding of Plio-Pleistocene ecologies and faunas in South Carolina. The purpose of this report is to discuss the significance of fossil amphibian remains, all referred to the genus Amphiuma, collected in South Carolina (Figure 1). These remains represent: 1) the oldest Pleistocene record of the Family Amphiumidae; 2) the northernmost Pleistocene record of Amphiuma from the

INTRODUCTION The southern portion of the Atlantic Coastal Plain, here defined as the region from the North Carolina-South Carolina state line south to the Georgia-Florida state line, contains a rich trove of Cenozoic vertebrate fossils, especially from the Pliocene and Pleistocene epochs. However, the fossils from this area have received comparatively little attention except from the hobby collecting community. Hulbert (2002) presented an overview of the vertebrate paleofaunas of Florida (listing over 1,100 taxa of fossil vertebrates) that provides some idea of the species that may also have inhabited Georgia and South Carolina. A review of the published records for South Carolina and Georgia revealed that only in the past three decades have concerted efforts been made to study the Plio-Pleistocene paleofaunas from this geographic region. Recent studies by Hulbert and Pratt (1998) and researchers from Georgia College and State University, Milledgeville, have begun to elucidate the structure of Pleistocene paleofaunas of Georgia (i.e., Mead et al., 2006; Parmely et al., 2007). In South Carolina the most popular hobby-collecting site is on Edisto Beach, and the Edisto Beach faunal assemblage (fa) was discussed by Roth and Laerm (1980), Holman (1995), and Sanders (2002). The best-known Pleistocene

1   

2 

PALUDICOLA, VOL. 8, NO. 1, 2010

  Atlantic Coastal Plain of the United States; 3) the first amphibian fossils reported from South Carolina; and 4) the first fossil amphibian records for both the Edisto Beach faunal assemblage and the Wando Formation. METHODS Amphiuma vertebrae have been collected from five sites in the South Carolina Coastal Plain (Fig. 1), and all of the specimens reported herein are housed in the collections of the South Carolina State Museum (SC) in Columbia. Vertebrae in the Walrus Ditch, Crowfield, and Rodent Ditch local faunas were obtained both by surface collecting of weather-exposed fossils and screen-washing in situ matrix (down to 0.25 mm mesh size). The specimen from the Camelot LF was recovered through screen-washing of matrix associated with a large sloth pelvis that was excavated at the site. Precise geographic information for these four sites is not listed here because they are currently on private property, but the information is on file at SC. The single specimen recovered from Edisto Beach had washed ashore and was found by a hobby collector while beachcombing. Fossil collecting is currently allowed in Edisto Beach State Park. SYSTEMATIC PALEONTOLOGY Order Caudata Family Amphiumidae Amphiuma sp. cf. A. means Garden, 1821 Figure 2 According to Gardner (2003), amphiumid trunk vertebrae are diagnostic and can be differentiated from those of other salamanders by a suite of features. Centra are amphicoelous with deeply concave cotyles (and we have observed that these are usually vertically oriented and oval; see Figure 2A, E, T) and the notochordal pit is retained (Figure 2E, F, J, K, O, P). Spinal foramina are absent. The neural crest is elongate (Figure 2H, M), moderately high, and often bears paired, posteriorly divergent neural spines (Figure 2B, Q). Transverse processes are weakly bicipitate on anterior-most trunk vertebrae but unicipitate on other trunk vertebrae. The ventral surface bears a longitudinal keel (Figure 2D, I, N, S) and a pair of well-developed anterior basapophyses (Fig. 2A, F, K, P). The trunk vertebrae and all but the posterior-most caudals of fossil and extant amphiumid salamanders bear a pair of dorsal postzygapophyseal crests adjacent to the neural crest (Figure 2C, G, H, L, M, Q, R), an apomorphy that is unique among salamanders (Gardner, 2003).

FIGURE 1. A, Geographic map of the eastern United States showing locations of some southeastern coastal states. B, Geographic map of South Carolina showing locations of the five fossil localities discussed in the text: 1 = Walrus Ditch site; 2 = Camelot site; 3 = Crowfield site; 4 = Rodent Ditch site; 5 = Edisto Beach site. Abbreviations: GA = Georgia; NC = North Carolina; SC = South Carolina. Base map in A adapted from Case (1994:text-fig. 1).

KNIGHT AND CICIMURRI—AMPHIUMA FROM SOUTH CAROLINA

3

  1. Walrus Ditch LF Material Examined—SC 89.245.51, one vertebra (Figure 2A-E), SC 89.245.50, bulk lot of ten vertebrae. Comments—The centrum length of the largest specimen in the sample measures 8.55 mm. It is likely that several individuals are represented in our sample because the vertebrae were found at different areas within the collection site. The Walrus Ditch LF was collected from the bank of a ditch flowing through a residential neighborhood adjacent to Summerville, Dorchester County. Remains of a considerable number of taxa have been recovered from the site, including terrestrial and marine mammals, birds, turtles, squamates, crocodilians, elasmobranchs, and teleost fish. Only the elasmobranchs have been studied in detail thus far (Cicimurri and Knight, unpublished data). The Amphiuma vertebrae were associated with mammalian taxa recognized by Morgan and Hulbert (1995) as late Blancan NALMA indicator taxa, including Nannippus peninsulatus, Holmesina floridana, and Canis lepophagus. The late Blancan NALMA falls within the Gelasian Stage, and until recently this stage was part of the Pliocene Epoch. The Pliocene-Pleistocene boundary was placed at 1.8 Ma (see Haq et al., 1977), but on 30 June, 2009, the International Union of Geological Sciences ratified a request from the International Commission on Stratigraphy to move the Pliocene-Pleistocene boundary from 1.8 Ma to 2.6 Ma (see web page at www.stratigraphy.org). The Gelasian Stage, an interval of time of roughly 800,000 years, is therefore now part of the Pleistocene Epoch. The occurrence of Amphiuma in the Walrus Ditch LF should therefore be considered as the oldest Pleistocene record of the genus (Holman, 2006; Parmley et al., 2007).

2. Camelot LF Material Examined- SC 2004.1.151, one vertebra (Figure 2F-J). Comments—The preservation of this specimen is nearly perfect, with only minor damage to the anterior part of the left prezygopophysis, the distal end of the left transverse process, and the right neural spine. The centrum measures 7.54 mm in length, and the width of the vertebra between the tips of the right and left transverse processes measures 12.41 mm. The Camelot local fauna was collected in an active limestone quarry (Giant Cement) located less than five km north of Harleyville, Dorchester County. The fossils are derived from fluvial sands and gravels of an undescribed formation that represent point bar, crevasse splay, and oxbow lakes that were part of a meandering river system. These fluvial deposits are

located under 3-4 m of unfossiliferous sediment and are exposed when overburden is removed in order to quarry the limestone. Based on the mammalian component, which is very similar to the Coleman IIA LF of Florida (Morgan and Hulbert, 1995), the Camelot LF is assignable to the late Irvingtonian NALMA and is 450,000 - 400,000 years old. This interval of time is within the Ionian Stage (middle Pleistocene), and the fossiliferous deposit is temporally equivalent to the Ladson Formation found near the South Carolina coast. 3,4. Crowfield and Rodent Ditch local faunas Material Examined—Crowfield local fauna: SC 2009.1.30, one vertebra (Fig. 2, K-O), SC 2009.1.31 bulk lot of ten vertebrae. Rodent Ditch local fauna: SC 99.43.1, one vertebra. Comments—The Crowfield and Rodent Ditch local faunas were derived from the Wando Formation. The largest specimen in the Crowfield sample measures 9.40 mm in centrum length, whereas the single specimen in the Rodent Ditch LF measures 4.65 mm in centrum length. The Crowfield material was collected from the Crowfield Lake construction site located between Summerville and Goose Creek, Berkeley County. The lake, developed as an ornamental feature for the Crowfield neighborhood, was dug in such a fashion that a gentle slope was formed that cut through the Wando Formation and exposed the significant disconformity with the underlying upper Oligocene (Chattian) Chandler Bridge Formation. The Rodent Ditch collection site is less than 20 km from the Crowfield site, and the Wando Formation is exposed in creek and ditch banks in a number of places in the immediate region. The Wando Formation consists of a series of sand, clay, and silt facies that are thought to be of fluvial to estuarine origin (Weems and Lemon, 1988). The formation has been dated to 130,000 to 70,000 Ka and is therefore predominantly within the Tarantian Stage (late Pleistocene). The mammalian component of the Crowfield and Rodent Ditch local faunas is assignable to the middle Rancholabrean NALMA. Fossil birds in the Crowfield LF include a grebe and at least four species of duck (Chandler and Knight,2009), and these provide additional evidence for the presence of an aquatic environment nearby. Our specimens represent the first record of a salamander from the Wando Formation, as well as the first published amphibian records in the Crowfield and Rodent Ditch local faunas (other amphibian taxa are known from these faunas and will be reported on at a later date).

4 

PALUDICOLA, VOL. 8, NO. 1, 2010

  5. Edisto Beach faunal assemblage Material Examined—SC 2005.57.1, one vertebra (Figure 2P-T) Comments—The specimen measures 8.05 mm in centrum length. The Edisto Beach faunal assemblage has long been considered to represent the Rancholabrean NALMA based on the mammals and turtles identified from the locality (Roth and Laerm, 1980; Holman, 1995; Sanders, 2002). However, we believe that there is some difficulty in assigning the various components of the assemblage to only a single NALMA. Edisto Beach specimens from other vertebrate groups, particularly the elasmobranchs and teleosts (currently under study), indicate a considerably broader temporal distribution spanning a time from some point in the Blancan NALMA (or even Hemphillian) well into or through the Rancholabrean (JLK and DJC unpublished data). Thus, the Edisto Beach faunal assemblage is heterochronic, and the Amphiuma vertebra could be as old as Pliocene or even late Miocene in age. This specimen represents the first fossil amphibian record for the assemblage. DISCUSSION The Amphiumidae are a small family of aquatic salamanders endemic to North America (Gardner, 2003). The fossil record of the group is primarily known from the Pleistocene to Recent of the southeastern United States (i.e., Meylan, 1995; Parmley et al., 2007), but there are records from the late Maastrichtian or early Paleocene (Lancian or Puercan NALMA) of Montana, the late Paleocene of Wyoming, and the middle Miocene of Texas (Salthe, 1973; Holman, 1977; Estes 1981; Gardner 2003). The identification of a partial skeleton from the lower Eocene (Wasatchian NALMA) of western Wyoming (Rieppel and Grande, 1998) as an amphiumid was refuted by Gardner (2003) and Holman (2006). In North America, five species have been assigned to the genus Amphiuma, including two fossil and three extant species. Of the fossil species, A. jepsoni was reported from the Polecat Bench Formation (late Paleocene, Tiffanian NALMA) of Park County, Wyoming (Estes, 1969). This taxon is known from incomplete skeletons and, contrary to the view of Rieppel and Grande (1998), Gardner (2003) considered A. jepsoni valid. Amphiuma antica was described by Holman (1977) from the Fleming Formation (middle Miocene, Barstovian NALMA) of Polk County, Texas. However, this taxon is based on a badly preserved trunk vertebra (Holman, 1977: fig. 2) and is considered a nomen dubium (Gardner, 2003; Parmley et al., 2007). A sixth Amphiuma species, A. nordica, was reported from the German Pleistocene, but the taxon is also a

nomen dubium because the holotype specimen is thought to belong to a teleost fish (Estes, 1981; Gardner, 2003). Genetic analyses by Bonett et al. (2009) indicate that the three extant species of Amphiuma diverged from a common ancestor four to ten million years ago, and A. means diverged from A. pholeter one to four million years ago. It seems reasonable to assume, therefore, that fossil Amphiuma remains occurring in Pleistocene through late Miocene strata can be assigned to one of the three extant species. We tentatively identify the South Carolina Amphiuma fossils as A. means based largely on the Recent distribution of the three extant species and because our material closely compares to a trunk vertebra of A. means illustrated by Gardner (2003, fig. 3F-H). Additionally, the large vertebrae in our sample are comparable in size and morphology to those of several Recent specimens of A. means we examined in the collection of the Georgia College and State University. Modern Amphiuma salamanders occur along the Gulf Coast and southeastern Atlantic Coast of the United States (Conant and Collins, 1998; Petranka, 1998; Bonett et al., 2009). These salamanders have an elongate and superficially eel-like body, and a single gill slit but no external gills (Salthe, 1973; Duellman and Trueb, 1986; Conant and Collins, 1998). The onetoed Amphiuma, A. pholeter, is a small species (up to 33 cm in length) that occurs in disjunct populations from the Mobile Bay region of Alabama eastward through the central western coast of Florida (Petranka, 1998; Conant and Collins, 1998; Bonett et al., 2009). The three-toed Amphiuma, A. tridactylum, is a large species (up to 106 cm in length) that ranges from the lower Mississippi River valley of southeastern Missouri southwest to Houston Bay, and southeast to western Mississippi (Petranka, 1998; Conant and Collins, 1998; Bonett et al., 2009). The two-toed Amphiuma, A. means, is the largest of the three extant species (up to 116.2 cm in length) and occurs in the Gulf Coastal Plain from eastern Louisiana into Florida, and into the Atlantic Coastal Plain from Florida to southeastern Virginia (Conant and Collins, 1998; Bonett et al., 2009). The presence of Amphiuma in a paleofauna can serve as a paleoecological indicator. This large salamander is paedomorphic, obligate aquatic, gillbreathing, and occupies a wide variety of aquatic habitats, including bottomland swamps, bayous, cypress swamps, drainage ditches in suburban and agricultural areas, permanent ponds and lakes, isolated ephemeral wetlands, wet prairies and marshes, small streams, and individuals often inhabit crayfish burrows (Lannoo, 2005). Amphiuma only very rarely leaves the water, but individuals may move overland during and following heavy rains (Lannoo, 2005). Gibbons

KNIGHT AND CICIMURRI—AMPHIUMA FROM SOUTH CAROLINA

5

 

FIGURE 2. Trunk vertebrae referred to Amphiuma sp. cf. A. means. Numbers in parentheses occurring after site names refer to locations on Figure 1B. A-E, SC 89.245.51, Walrus Ditch site (1). F-J, SC 2004.1.151, Camelot site (2). K-O, SC 2009.1.30, Crowfield site (3). P-T, SC 2005.57.1, Edisto Beach site (5). Anterior views in A, F, K, and P; dorsal views in B, G, L, and Q; right lateral views in C, H, and M, left lateral view in R (flipped horizontally); ventral views in D, I, N, and S (flipped horizontally); posterior views in E, J, O, and T. Anterior is at right in dorsal, lateral, and ventral views. Scale bars = 5 mm.

__________________________________________________________________________________________ and Semlitsch (1991), working in the Savannah River Site (SRS) in south-central South Carolina, collected specimens in pitfall traps that were located away from aquatic situations. Snodgrass et al. (1999), also working at the SRS, found that the occurrence of A. means in depression wetlands decreased as the distance from the nearest intermittent habitat increased. Petranka (1998) observed A. tridactylum as much as 12 m away from water. The habitat preferences of extant

Amphiuma strongly suggest that the discovery of Amphiuma in a paleofauna will serve as a good indicator of standing or slowly flowing bodies of water in a coastal or more inland setting. The above hypothesis is supported by the Amphiuma fossils within the Walrus Ditch, Crowfield, and Rodent Ditch local faunas. These local faunas contain a mixture of marine organisms (elasmobranchs, cetaceans, and sirenians) and terrestrial taxa like

6 

PALUDICOLA, VOL. 8, NO. 1, 2010

  artiodactyls, perissodactyls, canids, felids, and xenarthrans. These suggest very close proximity to a marine shoreline, where the remains of marine and terrestrial animals were mixed together. The compositions of the turtle communities within each LF, coupled with other associated fossils, further indicate the existence of standing or slowly moving bodies of fresh water in some portion of the habitat. CONCLUSIONS Twenty-five fossil Amphiuma vertebrae tentatively identified as A. means were collected from coastal plain strata of South Carolina. Isolated Amphiuma vertebrae collected from Pleistocene and Holocene sites elsewhere have been referred to A. means or simply Amphiuma sp. (see summary in Gardner, 2003). Parmley et al. (2007) stated that specific identifications based on vertebral morphology could not be clearly supported because of similarities between vertebrae of the three extant species. Previously published reports of Pleistocene Amphiuma sp. or A. means are from Florida (Brattstrom, 1953; Weigel, 1962; Hirschfield, 1969; Holman, 1995; Meylan, 1995), Georgia (Hulbert and Pratt, 1998; Parmley et al., 2007), and Texas (Holman, 1965; Slaughter and McClure, 1965). If the Texas material is correctly identified as A. means (Holman, 1965), then the western range of the species has been reduced by several hundred kilometers. The South Carolina fossils represent the northernmost Pleistocene record of Amphiuma in the Atlantic Coastal Plain, and with the placement of the Plio-Pleistocene boundary shifted to 2.6 Ma (previously 1.8 Ma), the specimens from the Walrus Ditch LF are the oldest Quaternary record of the genus (early Pleistocene; Gelasian Stage; late Blancan NALMA). ACKNOWLEDGEMENTS We thank D. Parmley and J. Gardner for providing critical reviews of the original manuscript. Parmley was also very helpful in discussions on the identification of Amphiuma vertebrae and arranged for access to Recent comparative skeletons. R. Doyle donated the Edisto Beach specimen. V. McCollum, J. Jacobs, F. Grady, D. Bohaska, R. Purdy, G. KcKee, M. Swilp, and L. Eberle are thanked for their field assistance and/or donation of material from the Crowfield and Walrus Ditch sites. R. Baulcomb sorted concentrates from the Camelot site. Thanks are also extended to our wives, Karin and Christian, for putting up with husbands who collect fossils.

LITERATURE CITED Bentley, C.C., and J.L. Knight. 1993. The oldest spotted turtle: Clemmys guttata (Testudines: Emydidae), from the late Pleistocene (Rancholabrean) Ardis local fauna, Dorchester County, South Carolina. South Carolina Geology 36:59-63. Bentley, C.C., and J.L. Knight. 1998. Turtles (Reptilia: Testudines) of the Ardis local fauna, late Pliestocene (Rancholabrean) of South Carolina. Brimleyana 25:3-33. 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. Bonett, R.M., P.T. Chippindale, P.E. Moler, R.W. Van Devender, and D.B. Wake. 2009. Evolution of gigantism in amphiumid salamanders. PLoS ONE 4(5):1-10. www.plosone.org. Brattstrom, B.H. 1953. Records of Pleistocene reptiles and amphibians from Florida. Quarterly Journal of the Florida Academy of Sciences 16:243-248. Case, G.R. 1994. Fossil fish remains from the late Paleocene Tuscahoma and early Eocene Bashi formations of Meridian, Lauderdale County, Mississippi. Palaeontographica Abteilung A 230:97-138. Chandler , R., and C. Bentley. 2007. Fossil Birds of the Ardis Local Fauna, Late Pleistocene, South Carolina. Current Research in the Pleistocene 24:8-9. Chandler, R., and J.L. Knight.2009. Fossil Birds of the Crowfield Local Fauna, late Pleistocene (Rancholabrean), South Carolina. Current Research in the Pleistocene 26:143-144. Cicimurri, D.J., and J.L. Knight. 2009. Two sharkbitten whale skeletons from Coastal Plain deposits of South Carolina. Southeastern Naturalist 8:71-82. Conant, R., and J.T. Collins. 1998. A Field Guide to Reptiles and Amphibians: Eastern and Central North America. 3rd edition (expanded), Houghton-Mifflin Company, Boston, Massachusetts. 634 pp. Duellman, W.E., and L. Trueb. 1986. Biology of Amphibians. McGraw Hill, New York. 630 pp. Estes, R. 1969. The fossil record of amphiumid salamanders. Breviora 322:1-11. Estes, R. 1981. Gymnophiora, Caudata. Encyclopedia of Paleoherpetology, Part 2. Gustav Fischer Verlag, Stuttgart, Germany. p. 1-115. (P. Wellnhofer ed.) Gardner, J.D. 2003. The fossil salamander Proamphiuma cretacea Estes (Caudata;

KNIGHT AND CICIMURRI—AMPHIUMA FROM SOUTH CAROLINA

7

  Amphiumidae) and relationships within the Amphiumidae. Journal of Vertebrate Paleontology 23:769-782. Gibbons, J.W., and R.D. Semlitsch. 1991. Guide to the Reptiles and Amphibians of the Savanna River Site. University of Georgia Press, Athens. 144 pp. Haq, B.U., W.A. Berggren, and J.A. van Couvering. 1977. Corrected age of the Pliocene/Pleistocene boundary. Nature 269:483-488. Hirschfield, S.E. 1969. Vertebrate fauna of Nichol’s Hammock, a natural trap. Quarterly Journal of the Florida Academy of Sciences 31:177-189. Holman, J.A. 1965. A small Pleistocene herpetofauna from Houston, Texas. Texas Journal of Science 27:418-423. Holman, J.A. 1977. Amphibians and reptiles from the Gulf Coast Miocene of Texas. Herpetologica 33:391-403. Holman, J.A. 1995. Pleistocene Amphibians and Reptiles in North America. Oxford University Press, New York. 243 pp. Holman, J.A. 2006. Fossil Salamanders of North America. Indiana University Press, Bloomington. 256 pp. Hulbert, R., III. 2002. The Fossil Vertebrates of Florida. University Press of Florida, Gainesville. 368 pp. Hulbert, R., and A.E. Pratt. 1998. New Pleistocene (Rancholabrean) vertebrate faunas from coastal Georgia. Journal of Vertebrate Paleontology 18:412-429. International Commission on Stratigraphy. 2009. www.stratigraphy.org. Lannoo, M. 2005. Amphibian Declines – The Conservation Status of United States Species. University of California Press, Berkeley. 1115pp. Mead, A.J., R.A. Bahn, R.M. Chandler, and D. Parmley. 2006. Preliminary comments on the Pleistocene vertebrate fauna from Clark Quarry, Brunswick, Georgia. Current Research in the Pleistocene 23:174-176. Meylan, P.A. 1995. Pleistocene amphibians and reptiles from the Leisey Shell Pit, Hillsborough

County, Florida. Bulletin of the Florida Museum of Natural History 37(pt. I, 9):273-297. Morgan, G.S., and R.C. Hulbert, III. 1995. Overview of the Geology and vertebrate biochronology of the Leisey Shell Pit Local Fauna, Hillsborough Caounty, Florida. Bulletin of the Florida Museum of natural History 37(pt. I, 9):1-92. Parmley, D., J. Clark, and A.J. Mead. 2007. Amphiuma (Caudata: Amphiumidae) from the Pleistocene Clark Quarry Local Fauna of coastal Georgia. Georgia Journal of Science 65:76-81. Petranka, J.W. 1998. Salamanders of the United States and Canada. Smithsonian Institution Press, Washington, D.C. 587 pp. Rieppel, O., and L. Grande. 1998. A well-preserved amphiumid (Lissamphibia: Caudata) from the Eocene Green River Formation of Wyoming. Journal of Vertebrate Paleontology 18:700-708. Roth, J.A., and J.A. Laerm. 1980. A late Pleistocene vertebrate assemblage from Edisto Island, South Carolina. Brimleyana 3:1-29. Sanders, A.E. 2002. Additions to the Pleistocene mammal faunas of South Carolina, North Carolina, and Georgia. Transactions of the American Philosophical Society, 92, 152 pp. Salthe, S.N. 1973. Amphiumidae, Amphiuma. Catalogue of American Amphibians and Reptiles 147:1-4. Slaughter, B.H., and W.L. McClure. 1965. The Sims Bayou local fauna: Pleistocene of Houston. Texas Journal of Science 17:404-417. Snodgrass, J. W., J. W. Ackerman, A. L. Bryan, Jr., and J. Burger. 1999. Influence of hydroperiod, isolation, and heterospecifics on the distribution of aquatic salamanders (Siren and Amphiuma) among depressional wetlands. Copeia 1999:107–113. Weems, R.E., and Lemon, E.M., Jr. 1988. Geologic map of the Ladson Quadrangle, Berkeley, Charleston, and Dorchester counties, South Carolina. U.S. Geological Survey, Geologic Quadrangle Map GQ-1630, 1:24,000 scale. Weigel, R.D. 1962. Fossil vertebrates of Vero, Florida. Florida Geological Survey Special Publication 10:1-59.

Paludicola 8(1):8-13 September 2010 © by the Rochester Institute of Vertebrate Paleontology

MAMMALS FROM THE BLUE ASH LOCAL FAUNA (LATE OLIGOCENE), SOUTH DAKOTA. RODENTIA PART 6: FAMILY CASTORIDAE AND ADDITIONAL EOMYIDAE WITH A SUMMARY OF THE COMPLETE RODENT FAUNA William W. Korth Rochester Institute of Vertebrate Paleontology, 265 Carling Road, Rochester, New York 14610

ABSTRACT A single beaver tooth (Castoridae) and several isolated teeth referable to the Eomyidae are described from the Blue Ash anthill fauna of South Dakota. The castorid tooth is referred to Agnotocastor sp. Four isolated upper molars are referred to a new species Orelladjidaumo amplus. It is distinguished from the only other known and type species of this genus O. xylodes (Orellan) by its much larger size and longer mesolophs. The remainder of the eomyid teeth are referred to Paradjidaumo sp. This species, reported previously from this fauna, may represent a new species distinguished by the crown-height of the cheek teeth, shich is greater than ina any other Paradjidaumo population. However, the record is too poor to establish this difference with confidence. A total of 42 species of rodents have been recognized from the Blue Ash anthill fauna representing 10 different families. The rodent taxonomic composition suggests that it is more likely Whitneyan than Arikareean in age.

INTRODUCTION

SYSTEMATIC PALEONTOLOGY

Preliminary faunal lists of the Blue Ash local fauna of South Dakota (Martin, 1974; Simpson, 1985) recognized from 19 or 22 species of fossil rodents (also see Korth, 2007a:table 1). A large collection of predominantly isolated teeth was collected from anthills in the Blue Ash horizon in the 1970s and constituted more than 1000 specimens (in the permanent collections of the Carnegie Museum of Natural History [CM]. The vast majority of these specimens were of rodents. Recently, in a series of papers, 40 species of rodents have been described from the anthill fauna (Korth, 2007a, b, 2008a, 2009a, 2009b, 2010). The final sample of isolated molars of rodents from this fauna is described below and includes two additional species of rodents not previously recognized, bringing the total of rodent species to 42, nearly twice the number identified in the preliminary studies. The provincial age of the fauna has been in question, with disagreement over whether it represents late Whitneyan or early Arikareean North American Land Mammal Age (NALMA). Now, the complete rodent fauna can be compared to others and the age determined with greater likelihood. Dental terminology for castorids follows Stirton (1935) and for eomyids follows Wood and Wilson (1936). Upper teeth are designated by capital letters, lower teeth are designated by lower-case letters (e.g., M1 and m1).

Family Castoridae Agnotocastor Stirton, 1935 Agnotocastor sp. indet. (Figure 1) Referred Specimen—CM 84637, right M3. Measurements—L = 2.68 m; W = 2.60 mm. Description—The single M3 is only slightly longer than wide, mesodont in crown-height, and rooted. It is squared on the buccal side, and tapers lingually along the posterior wall. The occlusal surface is complex. There is a distinct mesoflexus that extends more than half the width of the tooth and it curves posteriorly at its lingual end. The hypoflexus extends from the posterolingual corner of the tooth obliquely across the tooth to just lingual to its center. There are three small, complex fossettes on the surface of the tooth: one anterior to the mesoflexus and two posterior to it. A narrow flexus originates near the center of the anterior border of the tooth and extends posterolingually, isolating the protocone but not fusing with the hypoflexus. Discussion—Species belonging to the Palaeocastorinae, Agnotocastorinae, and Anchitheriomyinae have all been reported from the Whitneyan and Arikareean (Stirton, 1935; Martin, 1987; Korth, 1998, 2001a, 2001b). The Blue Ash M3, CM 84637, most closely approaches the morphology of Agnotocastor in its lower crown-height and complexity 8

KORTH—ADDITIONAL RODENTS FROM SOUTH DAKOTA

of the occlusal pattern. A. praeteraedens has been reported from the Whitneyan (Stirton, 1935; Korth, 2001a), but the Blue Ash specimen is slightly smaller. The Blue Ash M3 is also smaller than those referred to the Orellan A. coloradensis and A. readingi (Korth, 1996, 2001a), and it is also markedly smaller and lower-crowned than the Whitneyan anchitheriomyine, Oligotheriomys (Korth, 1998). _________________________________________

FIGURE 1. Agnotocastor sp. from the Blue Ash fauna. CM 84637, right M3. A, occlusal view. B, lingual view. Bar scale = 1 mm. _______________________________________________________

Family Eomyidae Orelladjidaumo Korth, 1989 Orelladjidaumo amplus n. sp. (Figure 2A, B; Table 1) Type Specimen—CM 84658, left M1 or M2. Referred Specimens—CM 84662, CM 84659, and CM 84660, isolated M1 or M2. Diagnosis—Upper molars larger than in type species, O. xylodes; mesolophs long on upper molars, extending to buccal margin of the tooth (short in O. xylodes).

9

Etymology—Latin, amplus, large. Description—The upper molars referred here are larger than any other eomyid recognized from the Blue Ash fauna. They are nearly as long as wide and squared in occlusal outline (Table 1). The crowns are mesodont as in Paradjidaumo, but are higher crowned lingually than buccally (unilateral hypsodonty). CM 84662 is the least worn specimen; the lingual height of the tooth is 1.6 mm and the buccal height is only 0.8 mm, half the lingual height. On the more worn specimens, the lingual and buccal heights are not as disparate, but the lingual measurement is always greater. The anterior cingulum is short, extending from the center of the anterior margin of the tooth to well short of the buccal border. The buccal cusps (paracone, metacone) are anteroposteriorly compressed and the lingual cusps (protocone and hypocone) are obliquely compressed (anterolingual-posterobuccal). The protoloph and metaloph run directly lingual from the paracone and metacone, respectively, joining their lingual cusps at their anteriobuccal corners. The mesoloph is long, extending to the buccal margin of the tooth and parallel to the major lophs. The posterior cingulum extends buccally from the hypocone, ending along the posterior margin of the metacone. Discussion—These specimens are referred to Orelladjidaumo based on the diagnosis of the genus (Korth, 1989): upper molars mesodont and unilaterally hypsodont, and nearly as long as wide. They are distinguished from the type species from the Orellan of Nebraska, O. xylodes, by their 25-30% larger size and longer mesolophs. The crown-height and occlusal morphology of O. amplus is very similar to that described for species of Paradjidaumo (Wood, 1937; Black, 1965; Setoguchi, 1978; Korth, 1980) but differs from those in having upper molars nearly as long as wide and being more hypsodont lingually than buccally. In species of Paradjidaumo, the M1 and M2 have anteroposterior lengths that range from 78 – 87% of the transverse width of the tooth. In both species of Orelladjidaumo, the lengths of the upper molars are 97-99% of their width. Although it is unusual that only upper molars are preserved of O. amplus, there are no eomyid lower teeth from the Blue Ash fauna of the size expected for this species. Korth (1994) listed “Metadjidaumo” cedrus (Korth, 1981) from the Orellan of Colorado as questionably referable to Orelladjidaumo. However, he later suggested that “M.” cedrus was more likely referable to the Arikareean genus Neoadjidaumo (Korth, 2008b). Therefore, the only two definite species of Orelladjidaumo are O. xylodes (type) and O. amplus.

10

PALUDICOLA, VOL. 8, NO. 1, 2010

FIGURE 2. Orelladjidaumo amplus and Paradjidaumo sp. from the Blue Ash fauna. A, B, O. amplus, CM 84658, left M1 or M2 (holotype). A, occlual view. B, posterior view (occlusal surface toward top of page). C-E. Paradjidaumo sp. C, D, CM 84663, right m1 or m2. C, occlusal view. D. lingual view. E, CM 84666, right m3 (occlusal view). Bar scale = 1 mm. __________________________________________________________________________________________________________________

Paradjidaumo Burke, 1934 Paradjidaumo sp. (Figure 2C-E; Table 1, 2) Additional Referred Specimens—CM 84663, 84664 - m1 or m2; CM 84667 – m3; CM 84665 - M1 or M2; and CM 84666, M3. (Also see Korth, 2007:33.) Discussion—The specimens referred to Paradjidaumo sp. do not differ in morphology or size from those previously identified from the Blue Ash fauna (Korth, 2007a); nor do they differ from the Orellan P. trilophus (Korth, 1980) except in crownheight of the lower molars. Dividing the height of the enamel on the lingual side of the tooth measured at the central valley by the maximum width of the tooth provides an index for crown-height. This method has

been used for heteromyid dentitions by Korth (1997) and Lindsay and Reynolds (2008). The lower molars from the Blue Ash fauna have an index that ranges from 0.37-0.43 (mean = 0.40; Table 2). The average crown-height for P. trilophus and the supposedly higher-crowned P. hypsodus is 0.35-0.36. However, some specimens referred to the latter two species range above the average height of the Blue Ash specimens. Despite overlap, the average height of specimens from the Blue Ash fauna is greaer than the average of other samples of Paradjidaumo. A larger sample from the Blue Ash fauna is necessary before it can be established with certainty that the molars are higher-crowned, thus representing a new species.

KORTH—ADDITIONAL RODENTS FROM SOUTH DAKOTA

11

TABLE 1. Measurements of Orelladjidaumo amplus and Paradjidaumo sp. from the Blue Ash fauna. Abbreviations: L, anteroposterior length; W, transverse width; ht, lingual crown-height. Since the first and second molars cannot be distinguished, measurements listed as M1 and m1 should be read as M1 or M2 and m1 or m2. Measurements in mm. M1L

M1W

Paradjidaumo sp. 84665 76294 76294

1.35 1.2 1.24

1.47 1.33 1.40

mean

1.26

1.40

84667 84663 84664

M3L

M3W

1.12

1.47

mean

m1L

m1W

ht

ht/W

1.50 1.45

1.40 1.50

0.6 0.55

0.43 0.37

1.48

1.45

0.58

0.4

84666 Orelladjidaumo amplus 84662 84658 84659 84660

1.55 1.60 1.65 1.68

m3L

m3W

1.30

1.40

1.62 1.68 1.70 1.73

mean 1.62 1.68 _______________________________________________________________________________________________ CONCLUSIONS With the addition of Orelladjidaumo amplus and Agnotocastor sp., the total number of rodent species recognized from the Blue Ash anthill fauna is 42 (Table 3). As noted previously, the fauna contains elements characteristic of Orellan, Whitneyan, and Arikareean North American Land Mammal Ages. However, with the full rodent fauna to consider, a more reliable age determination can be made. Of the 42 rodent species, nearly half of them (19) are unique to this fauna. Four species are otherwise exclusive to the Whitneyan (Cedromus wilsoni, Campestrallomys siouxensis, Kirkomys parvus, Eumys brachyodus); two are limited to the late Orellan and Whitneyan (Prosciurus magnus, Scottimus longiquus); and two are previously known from the Whitneyan and Arikareean (Scottimus kellamorum, Leptodontomys douglassi). Two additional species are only known from the Orellan (Protosciurus mengi, Eumys parvidens). Both of these are listed as “cf.” because they are slightly larger than the Orellan species.

TABLE 2. Crown-height index (ht/W) for species of Paradjidaumo. Abbreviation: N, number of specimens measured; MT, Montana; NE, Nebraska; WY, Wyoming. All other abbreviations as in Table 1. m1 or m2 ht/W

mean

range

N

P. trilophus (Chadronian, MT)

0.36

0.20-0.48

26

P. trilophus (Orellan, NE)

0.35

0.24-0.40

19

P. “hypsodus” (late Orellan, WY)

0.35

0.27-0.49

16

P. validus (Orellan, NE)

0.38

Blue Ash Paradijdaumo

0.40

0.37-0.43

2

1

____________________________________________ Seven species from the Blue Ash fauna are elsewhere restricted to the Arikareean, although three of these are listed as ”cf.” and are smaller or morphologically more primitive than the Arikareean species (Florentiamys sp., cf. F. kingi, Paciculus sp., cf. P. nebraskensis, Leidymys sp., cf. L. blacki). Similarly, nine species from Blue Ash that represent otherwise Arikareean genera are smaller or morphologically more primitive than the known Arikareean species (Niglarodon brachyodon, Allomys sp., Ansomys cyanotephrus, Miospermophilus sp.,

12

PALUDICOLA, VOL. 8, NO. 1, 2010

Zetamys sp., Downsimus sp., Paciculus dakotensis, Geringia copiosus, Hitonkala martintau). At the generic level, six have their last occurrence in the Whitneyan (Prosciurus, Oligospermophilus, Agnotocastor, Cedromus) or are limited to the Whitneyan (Disallomys, Kirkomys). There are only four species in the Blue Ash anthill fauna that are elsewhere restricted to the Arikareean (Nototamias sp., Protosciurus rachelae, Tylionomys voorhiesi, Campesterallomys dawsonae). Based on these occurrences, and stage of evolution of the species present, it appears most likely that the Blue Ash fauna is Whitneyan rather than Arikareean in age. Similar conclusions can be made when considering other mammalian groups such as marsupials, lagomorphs, and insectivorans (Korth, 2007b, 2009c). The only remaining part of the anthill fauna that is yet to be described is the Carnivora and a single artiodactyl tooth. ____________________________________________ TABLE 3. Rodent species recognized from the Blue Ash anthill fauna of South Dakota. (Species in bold type are unique to the fauna.) APLODONTIDAE Prosciurus clausulus P. magnus Campestrallomys siouxensis C. dawsonae Downsimus sp. Disallomys robustus D. intermedius Ansomys cyanotephrus Allomys sp. Niglarodon brachyodon SCIURIDAE Hesperopetes jamesi H. blacki Nototamias sp. Douglassciurus sapphirus Protosciurus sp. cf. P. mengi Protosciurus rachelae Miospermophilus sp Cedromus wilsoni Oligospermophilus emryi EUTYPOMYIDAE Eutypomys wilsoni CASTORIDAE Agnotocastor sp. EOMYIDAE Paradjidaumo sp. Orelladjidaumo amplus Leptodontomys sp. cf. L. douglassi Zophoapeomys indicus HELISCOMYIDAE Heliscomys medius Tylionomys voorhiesi (=Heliscomys sp.) FLORENTIAMYIDAE Florentiamys sp. cf. F. kingi Kirkomys parvus Hitonkala martintau GEOMYIDAE

Geomyid undetermined CRICETIDAE Eumys brachyodus E. sp., cf. E. parvidens Scottimus kellamorum S. sp., cf. S. longiquus Leidymys juxtaparvulus L. sp., cf. L. blacki Paciculus dakotensis P. sp., cf. P. nebraskensis Geringia copiosus Cricetid indeterminate FAMILY UNCERTAIN Zetamys sp. _______________________________________________________

ACKNOWLEDGEMENTS The specimens used in this study were loaned by Dr. M. R. Dawson of the Carnegie Museum of Natural History. Partial funding for this research was provided by private donations to the Rochester Institute of Vertebrate Paleontology. Photographic equipment was provided by Dr. W. Hallahan of the Biology Department, Nazareth College. Earlier versions of this paper were critically reviewed by Drs. L. B. Albright and L. Flynn. LITERATURE CITED Black, C. C. 1965. Fossil mammals from Montana. Pt. 2. Rodents from the early Oligocene Pipestone Springs local fauna. Annals of Carnegie Museum 38:1-48. Burke, J. J. 1934. New Duchesne River rodents and a preliminary survey of the Adjidaumidae. Annals of Carnegie Museum 23:391-398. Korth, W. W. 1980. Paradjidaumo (Eomyidae, Rodentia) from the Brule Formation, Nebraska. Journal of Paleontology, 54:943-951. Korth, W. W. 1981. Metadjidaumo (Eomyidae, Rodentia) from Colorado and Wyoming. Journal of Paleontology, 55:598-602. Korth, W. W. 1989. Geomyoid rodents (Mammalia) from the Orellan (Oligocene) of Nebraska. Natural History Museum of Los Angeles County Science Series, 33:31-46. Korth, W. W. 1994. The Tertiary Record of Rodents in North America. Plenum Press, 319 pp. Korth, W. W. 1996. Additional specimens of Agnotocastor readingi (Rodentia, Castoridae) from the Orellan (Oligocene) of Nebraska and a possible origin for the beavers. Paludicola, 1:1620. Korth, W. W. 1997. A new subfamily of primitive pocket mice (Heteromyidae, Rodentia) from the middle Tertiary. Paludicola, 1:33-66.

KORTH—ADDITIONAL RODENTS FROM SOUTH DAKOTA

Korth, W. W. 1998. A new beaver (Rodentia, Castoridae) from the Orellan (Oligocene) of North Dakota. Paludicola, 1:127-131. Korth, W. W. 2001a. Cranial morphology of some early beavers (Rodentia, Castoridae) from the Oligocene (Orellan and Whitneyan) of South Dakota. Paludicola, 3:40-50. Korth, W. W. 2001b. Comments on the systematics and classification of the beavers (Rodentia, Castoridae). Journal of Mammalian Evolution, 8:279-296. Korth, W. W. 2007a. Mammals from the Blue Ash local fauna (late Oligocene), South Dakota. Rodentia, Part 1: Families Eutypomyidae, Eomyidae, Heliscomyidae, and Zetamys. Paludicola, 6:31-40. Korth, W. W. 2007b. Mammals from the Blue Ash local fauna (late Oligocene), South Dakota, Marsupialia and Lagomorpha. Paludicola, 6:111-117. Korth, W. W. 2008a. Mammals from the Blue Ash local fauna (late Oligocene), South Dakota. Rodentia, Part 2: Families Florentiamyidae and Geomyidae. Paludicola, 7:14-25. Korth, W. W. 2008b. Eomyid rodents (Mammalia) from the early Arikareean (Oligocene) of western Nebraska. Paludicola, 6:144-154. Korth, W. W. 2009a. Mammals from the Blue Ash local fauna (late Oligocene), South Dakota. Rodentia, Part 3: Family Sciuridae. Paludicola, 7:47-60. Korth, W. W. 2009b. Mammals from the Blue Ash local fauna (late Oligocene), South Dakota. Rodentia, Part 4: Family Aplodontidae. Paludicola, 7:89-106. Korth, W. W. 2009c. Mammals from the Blue Ash local fauna (late Oligocene), South Dakota. Lypotyphla and additional Marsupialia. Paludicola, 7:78-88.

13

Korth, W. W. 2010. Mammals from the Blue Ash local fauna (late Oligocene), South Dakota. Rodentia, Part 5: Family Cricetidae. Paludicola, 7:117-136. Lindsay, E. H. and R. E. Reynolds. 2008. Heteromyid rodents from Miocene faunas of the Mojave Desert, southern California. Natural History Museum of Los Angeles County Science Series, 41: 213-235. Martin, L. D. 1974. New rodents from the Lower Miocene Gering Formation of western Nebraska. Occasional Papers of the Museum of Natural history, University of Kansas 32:1-12. Martin, L. D. 1987. Beavers from the Harrison Formation (Early Miocene) with a revision of Euhapsis, Dakoterra, 3:73-91. Setoguchi, T. 1978. Paleontology and geology of the Badwater Creek Area, central Wyoming. Part 16. The Cedar Ridge local fauna (late Oligocene). Bulletin of the Carnegie Museum of Natural History 9:1-61. Simpson, W. F. 1985. Geology and paleontology of the Oligocene Harris Ranch Badlands, southwestern South Dakota. Dakoterra 2:303333. Stirton, 1935. A review of the Tertiary beavers. University of California Publications in the Geological Sciences, 23:391–458. Wood, A. E. 1937. The mammalian fauna of the White River Oligocene. Part 2, Rodentia. Transactions of the American Philosophical Society 28:155-262. Wood, A. E. and R. W. Wilson. 1936. A suggested nomenclature for the cusps of the cheek teeth of rodents. Journal of Paleontology 10:388-391.

Paludicola 8(1):14-21 September 2010 © by the Rochester Institute of Vertebrate Paleontology

THE NORTH AMERICAN PROMIMOMYS IMMIGRATION EVENT

Robert A. Martin Department of Biological Sciences, Murray State University, Murray, Kentucky 42071

ABSTRACT Explosive radiation of the mammalian rodent family Arvicolidae (muskrats, voles and lemmings) at the end of the Miocene and throughout the remaining late Neogene provided an ideal group for biochronological correlations. Consecutive waves of arvicolid immigration into North America, plus some autochthonous cladogenesis, allows the development of a faunal chronology based almost exclusively on these animals. One thorny problem has been the relatively early appearance of Promimomys (= Prosomys) in North America, leaving the distinct possibility that arvicolids first evolved in the New World. A summary of biochronological and radioisotopic information shows that Promimomys dispersed to North America from Asia subsequent to about 5.5 Ma and gave rise to the Pliophenacomyinae between 5.5-5.0 Ma.

external age control (Repenning, 1987; Bell et al., 2004), and therefore establishing their approximate chronological position must be done with biostratigraphic information. In this essay I examine a series of North American mammalian assemblages ranging in age from about 7-4.9 Ma. The results support the revised immigration timing for Promimomys suggested by Repenning (1990) and Bell (2000); Proimimomys is absent from fossil assemblages older than about 5.5 Ma, and apparently gave rise to the North American endemic subfamily Pliophenacomyinae between about 5.5-5.0 Ma.

INTRODUCTION Our understanding of the origin of the cosmopolitan family Arvicolidae has been complicated by the occasional assumption that Promimomys mimus (= Prosomys mimus Shotwell, 1956) appeared earlier in North America than it did elsewhere, despite the fact that a host of arvicolid-like cricetids (e.g., Baranarviomys, Microtodon [= Baranomys, Bjornkurtenia], Celadensia) are found in Europe and Asia but not in North America. Indeed, Repenning’s (1987) biochronology of arvicolid rodents began with the immigration of Promimomys to the United States at 6.7 + 0.5 Ma, his “Event 1” (Repenning, 1987, p. 239), leaving open the possibility that Eurasian Promimomys is the result of an as yet undiscovered early radiation of North American proto-arvicolids. In subsequent treatments, Repenning et al. (1990) and Bell (2000) placed the immigration of Promimomys (and Dispersal Event 1) between 5.5-4.8 Ma, but provided no substantiating evidence. In possible support of the older dispersal scenario, Zakrzewski and Harington (2001) reported a mandible of cf Baranomys from a locality on Ellesmere Island, high in the Canadian Arctic. The associated mammalian fauna pointed to an “early Pliocene” age, too late for ancestry to Promimomys, but suggestive nonetheless of a previously unknown high arctic dispersal and possible radiation of arvicolid-like taxa. Compounding the problem is the fact that the late Hemphillian localities with Promimomys, Christmas Valley and McKay Reservoir, Oregon, and Mailbox, Nebraska, lack

METHODS Table 1 was constructed from a variety of late Hemphillian and early Blancan local faunas. The information was taken from published sources (Shotwell, 1956, 1970; Jacobs, 1977; Baskin, 1979; May, 1981; Dalquest, 1983; Tedford et al., 2004; White, 1987, 1991; Voorhies, 1990), on-line museum collection databases (Pinole assemblage; Museum of Paleontology, University of California, Berkeley) and John Alroy’s on-line compilation of mammalian records known as The Paleobiology Database (http//paleodb.org). Radiosotopic calibration points were as follows: A fission-track date of 6.8 + 0.2 and an Ar/Ar date of 6.8+/- 0.03 Ma were reported from an ash just above the Coffee Ranch l.f. of Texas (Tedford et al., 2004). A range of K/Ar dates from the Quiburis Formation of 5.21-6.25 Ma within which the Redington l.f. is found was given by Jacobs (1977). A fission14

PALUDICOAL, VOL. 8, NO. 1, 2010

track glass date for an ash above the Santee l.f. of Nebraska with Protopliophenacomys parkeri was reported by Boellstorf (1976) as 5.0 + 0.2 Ma. A fossiliferous Blancan sequence from the Panaca Formation of Nevada was reported by Lindsay et al. (2002) in their redefinition of the Hemphillian-Blancan boundary. An Ar/Ar date of 4.69 Ma was derived from pumice tuff materials overlying sediments from which Ophiomys panacaensis was collected. The lowest record of O. panacaensis is from a level in the sequence just above a pumice with two Ar/Ar dates at 4.96 + 0.01 and 4.96 + 0.02 Ma. A tuff overlying the Pinole assemblage from the San Francisco Bay area of California is dated at 5.5 +0.2 Ma (Tedford et al., 2004).

RESULTS AND DISCUSSION The database in Table 1 is separated into those mammalian assemblages with and without arvicolid rodents. None of the assemblages >5.5 Ma contains arvicolids (Goniodontomys and Paramicrotoscoptes are not now considered to be arvicolids, based on characters of the mandible outlined by Repenning [1968]), and this difference conveniently basically correlates in time with the Miocene/Pliocene boundary (about 5.3 Ma). Nevertheless, the relatively dense mammalian record for the late Miocene does not demonstrate an obvious schism that could separate a more ancient “fauna” from one with arvicolids. Rather, each assemblage in Table 1 includes a few unique species that likely represent a combination of immigration and autochthonous cladogenetic events. Some small mammals, such as Lemoynea, Aneuroneomys, the eomyid rodent Kansasimys, and Paramicrotoscoptes and Goniodontomys are restricted to older local faunas, but many genera have species that range from the Hemphillian through the Blancan (e.g., Hypolagus, Nekrolagus, Spermophilus, Bensonomys, Perognathus, Onychomys, Repomys, Bensonomys, etc.). I am hesitant to use the large mammals for detailed biochronological purposes because their fossilization probability, and therefore their potential representation in fossil assemblages, is so much lower than with small mammals. Still, prearvicolid assemblages include the large mammals Aphelops, Dinohippus, Pliohippus, Astrohippus, Alforjas, Pediomeryx, Texoceras, Nimravides, and Barbourofelis. The rhinoceros Teleoceras, once considered to be a Hemphillian indicator, extends to the early Blancan, as it has been recovered with Ogmodontomys from the Pipe Creek Sinkhole of Indiana (Martin et al., 2002). Mylagaulid rodents remain as good Hemphillian indicators, as they are last seen in the Santee l. f. (Voorhies, 1990).

15

This information suggests that the North American Promimomys (= Prosomys) immigration event occurred between 5.5-5.0 Ma, considerably later than the original proposal of 6.67 Ma. Although more in line with European records of Promimomys, dispersal of this genus to the New World still seems early compared to the earliest records of Promimomys in Europe, which had been set at about 4.9 Ma, at the base of MN (European Mammal Neogene zone) 14 (Fejfar et al., 1998). However, Hordijk and de Bruijn (2009) recently established the occurrence of Promimomys cor in the Ptolemais lacustrine-lignitic sequence of Greece at a level in excess of 5.0 million years. As Promimomys cor appears to be somewhat morphologically advanced over P. insuliferus (Fejfar et al., 1998), we can look to an earlier date in Eurasia for the origination of Promimomys from its cricetid ancestor. Changzhu and Yingqi (2005) described a species of Promimomys, P. asiaticus, from a cave deposit in Dajushan Hill, Anhui Province, China, which they considered to exhibit the most primitive dental features of the genus. They concluded, based on the associated fauna, that the Dajushan Hill assemblage was deposited in early MN 14 which, with the new Ptolemais Promimomys records, should likely be pushed back from its current position at about 4.9 Ma (Fejfar et al., 1998) to at least 5.3 Ma. Despite the independent age assessment, these considerations collectively support the proposition of Repenning et al. (1990) and Bell (2000) that Promimomys originated in Asia subsequent to 5.5 Ma and dispersed soon after to both Europe and North America. Chaline et al. (1999) suggested that Ogmodontomys evolved from North American Promimomys mimus, but intermediates are not known from the North American fossil record, whereas they are between North American Promimomys and the Pliophenacomyinae (Martin et al., 2002; Martin, 2007). Ophiomys and Ogmodontomys are closely related and diverged rapidly after they appeared in North America. The first appearance datum of an Ophiomys-like m1 dental morphology (e.g., O. panacaensis; Mou, 1997) is used to define the Hemphillian/Blancan boundary (Lindsay et al., 2002) at approximately 4.98 Ma. A potential ancestor, Promimomys antiquus, has been described from Siberia (Zazhigin, 1980) and, at least for now, the Blancan presence of Ophiomys seems likely to represent a second dispersal event a few hundred thousand years after Promimomys entered the continent. ACKNOWLEDGMENTS This essay was prompted by discussions with the late C. Repenning and, more recently, with my French

16

MARTIN—PROMIMOMYS IMMIGRATION

colleague L. Viriot. I thank A. Barnosky, C. Bell, J. Honey, K. Hordijk, and H. de Bruijn for information and leads to various databases and D. Prothero, K. Hordijk, R. Zakrzewski, W. Korth and an anonymous reviewer for critiquing the manuscript. LITERATURE CITED Baskin, J. A. 1979. Small mammals of the Hemphillian age White Cone local fauna, northeastern Arizona. Journal of Paleontology 53:695-708. Bell, C. J. 2000. Biochronology of North American microtine rodents. Pp. 379-406 in J. S. Noller, J. M. Sowers and W. R. Lettis (eds.), Quaternary Geology Methods and Applications. American Geophyscial Union, Washington, D. C. Bell, C. J., Lundelius Jr., E. L., Barnosky, A. D., Graham, R. W., Lindsay, E. H., Ruez Jr., D. R., Semken Jr., H. A., Webb, S. D., Zakrzewski, R. J. 2004. The Blancan, Irvingtonian, and Rancholabrean Mammal Ages. Pp.232-314 in M. O. Woodburne (ed.), Late Cretaceous and Cenozoic Mammals of North America: Biostratigraphy and Geochronology. Columbia University Press, New York. Pp. 232-314. Boellstorf, J. 1976. The succession of late Cenozoic volcanic ashes in the Great Plains: a progress report. Pp. 37-71 in C. K. Bayne (ed.), Guidebook 24th Annual Meeting Midwestern Friends of the Pleistocene. Kansas Geological Survey, University of Kansas. Guidebook Series 1. Chaline, J., P. Brunet-Lecomte, S. Montuire, L. Viriot and F. Courant. 1999. Anatomy of the arvicoline radiation (Rodentia): palaeogeographical, palaeoecological history and evolutionary data. Annals Zoologica Fennici 36:239-267. Changzhu, J. and Y. Zhang. 2005. First discovery of Promimomys (Arvicolidae) in east Asia. Chinese Science Bulletin 50:327-332. Dalquest, W. W. 1983. Mammals of the Coffee Ranch local fauna Hemphillian of Texas. The PearceSellards Series. No. 38:1-41. Fejfar, O., W.-D. Heinrich, and E. H. Lindsay. 1998. Updating the Neogene rodent biochronology in Europe. Pp. 533-553 in T. Van Kolfschoten and P. Gibbard (eds.), The Dawn of the Quaternary. Mededelingen Netherlands Institute voor Toogepaste. Geowetenschappen TNO, 60. Hordijk, K. and H. de Bruijn. 2009. The succession of rodent faunas from the Mio/Pliocene lacustrine deposits of the Florina-Ptolemais-Servia Basin (Greece). Hellenic Journal of Geosciences

44:21-103. Jacobs, L. L. 1977. Rodents of the Hemphillian age Redington local fauna, San Pedro Valley, Arizona. Journal of Paleontology 51:505-519. Lindsay, E. H. , Mou, Y. and Downs, W. 2002. Recognition of the Hemphillian/Blancan boundary in Nevada. Journal of Vertebrate Paleontology 22:429-442. Martin, R. A. 2007. Arvicolidae. Pp. 480-497 in C. M. Janis, G. F. Gunnell and M. D. Uhen (eds.), Evolution of Tertiary Mammals of North America, Vol. 2. Cambridge University Press, New York. Martin, R. A., H. T. Goodwin, and J. O. Farlow. 2002. Late Neogene (late Hemphillian) rodents from the Pipe Creek Sinkhole, Grant County, Indiana. Journal of Vertebrate Paleontology 22:137-151. May, S. R. 1981. Repomys (Mammalia: Rodentia gen. nov.) from the late Neogene of California and Nevada. Journal of Vertebrate Paleontology 1:219-230. Mou, Y. 1998. Schmelzmuster of Mimomys panacaensis. Pp. 79–90 in Y. Tomida, L. J. Flynn and L. L. Jacobs (eds.), Advances in Vertebrate Paleontology and Geochronology. National Science Museum Monograph 14, Tokyo. Repenning, C. A. 1968. Mandibular musculature and the origin of the subfamily Arvicolinae (Rodentia). Acta Zoologica Cracoviensia 13:29-72. Repenning, C. A. 1987. Biochronology of the microtine rodents of the United States. Pp. 236-268 in M. O.Woodburne (ed.) Cenozoic Mammals of North America. University of California Press, Berkeley. Repenning, C. A., Fejfar, O. and Heinrich, W.-D.1990. Arvicolid rodent biochronology of the northern hemisphere. Pp. 385-418 in O. Fejfar and W.D. Heinrich (eds.); International Symposium Evolution, Phylogeny and Biostratigraphy of Arvicolids (Rodentia, Mammalia). Geological Survey, Prague. Shotwell, J. A. 1956. Hemphillian mammalian assemblage from northeastern Oregon. Bulletin of the Geological Society of America 67:717738. Shotwell, J. A. 1970. Pliocene mammals of southeast Oregon and adjacent Idaho. Bulletin Museum of Natural History, University of Oregon. 17:1103. Tedford, R. H., et al. 2004. Mammalian biochronology of the Arikareean through Hemphillian interval (late Oligocene through early Pliocene epochs). Pp. 169-231 in M. O. Woodburne (ed.), Late

PALUDICOAL, VOL. 8, NO. 1, 2010

Cretaceous and Cenozoic Mammals of North America: Biostratigraphy and Geochronology. Columbia University Press, New York. Voorhies, M. R. 1990. Vertebrate biostratigraphy of the Ogallala Group in Nebraska. Pp. 115-151 in T. C. Gustavson (ed.), Geological Framework and Regional Hydrology: Upper Cenozoic Blackwater Draw and Ogallala Formations, Great Plains. Bureau of Economic Geology, University of Texas, Austin. White, J. A. 1987. The Archaeolaginae (Mammalia, Lagomorpha) of North America, excluding Archaeolagus and Panolax. Journal of

17

Vertebrate Paleontology 7:425-450. White, J. A. 1991. North American Leporinae (Mammalia: Lagomorpha) from late Miocene (Clarendonian) to latest Pliocene (Blancan). Journal of Vertebrate Paleontology 11:67-89. Zakrzewski, R. J. and C. R. Harington, 2001. Unusual Pliocene rodent from the Canadian arctic islands. Journal of Vertebrate Paleontology, 21 (suppl. To no. 3):116A-117A. Zazhigin, V. 1980. Late Pliocene and Anthropogene rodents of the south of western Siberia. Academy of Sciences USSR, Geological Institute, Moscow, Transaction 339:1-156.

18

MARTIN—PROMIMOMYS IMMIGRATION

TABLE 1. Mammal species lists for select assemblages between about 7-4.6 million years ago. NALMA = North American Land Mammal Age, Ma = millions of years ago, sp. = referable to the genus, species unknown, cf = compares favorably with; roughly equivalent to “aff,” = has affinities with. Radioisotope dates are from the literature (see text). Epoch

Miocene

Pliocene

NALMA Arvicolids Age (Ma) Local faunas

absent

Hemphillian

Blancan

present

?>6.0

6.69

6.25-5.21

6.5-6.8

5.5

?

?

5.0

4.964.69

Cambridge

White Cone

Redington

Coffee Ranch

Pinole

McKay Reservoir

Mailbox

Santee

Panaca

Mammal taxa Domninoides

?

Notiosorex

sp.

Meterix

x

Aneuroneomys magnus

x

sp.

Paracryptotis Scalopus (Hesperoscalops Scalopus (Hesperoscalops) reficervus

x

x

x

x

x

x

cf

Eptesicus hemphillensis

x

Eptesicus

x

Myotis Pliometanastes

x

x

Scapanus Lemoynea biradicularis

x

x x

Megalonyx curvidens

cf

Megalonyx leptostomus Thinobadistes

x x

Ochotona spanglei Hypolagus ringoldensis

x x

cf

x

Hypolagus tedfordi

x x

Hypolagus vetus

x

Hypolagus edensis

x

cf

x

x x

Hypolagus oregonensis

x x

Hypolagus regalis Hypolagus

x x

Alilepus vagus

x

Lepoides lepoides

x

Pewelagus dawsoni

x x

Nekrolagus progressus Mylagaulus monodon

x

x sp.

cf

cf

sp.

Ceratogaulus hatcheri

x

Paenemarmota cf nevadensis

x

Paenemarmota sawrockensis

cf x

Marmota vetus

x

Marmota oregonensis

x

Spermophilus wilsoni

x

Spermophilus mckayensis

x

PALUDICOAL, VOL. 8, NO. 1, 2010

Local faunas

Cambridge

White Cone

Redington

Coffee Ranch

Spermophilus

x

x

x

x

Tamias

Pinole

19

McKay Reservoir

Santee

x

cf

x

Perognathus sargenti

x

Perognathus henryfieldi

x

Perognathus mclaughlini

x

Perognathus

x

Prodipodomys kansasensis

x x

Pliosaccomys dubius

x

x

x

x x

x

Parapliosaccomys oregonensis

x

Parapliosaccomys

x

Pliogeomys sulcatus

x

Pliogeomys

x

Thomomys

sp. x

Dipoides Dipoides smithi

x

x

x

x

x

x

x x

Dipoides williamsi

x

Castor cf californicus

sp.

Kansasimys rogersi

x

Kansasimys wilsoni Copemys

x

x

Prodipodomys sp.

Dipoides stirtoni

x x

Antecalomys vasquezi

x

Bensonomys coffeyi

x

Bensonomys gidleyi

x

Bensonomys yahzi

x

Bensonomys

x

Baiomys

x

Paronychomys

x

Paronychomys alticuspis

x

x

Paronychomys lemredfieldi

x

Paronychomys tuttlei

x

x

x

Onychomys larabeei

cf

Galushamys redingtonensis

x

Repomys gustleyi

x

Repomys panacaensis

x

Repomys

cf

sp.

Neotoma sawrockensis

x

Paramicrotoscoptes hibbardi

cf

Goniodontomys disjunctus

x

Promimomys mimus Protopliophenacomys parkeri

Panaca

x

Perognathoides bidahochiensis

Pliosaccomys

Mailbox

x

x

sp.

x x

Ophiomys panacaensis

x

Ondatra zibethicus /meadensis

cf

20

Local faunas

MARTIN—PROMIMOMYS IMMIGRATION

Cambridge

White Cone

Redington

Coffee Ranch

Pinole

Pliozapus solus Osteoborus cyonoides Osteoborus

McKay Reservoir

Mailbox

x

cf

x

x

Santee

Panaca

x x

Borophagus parvus

x

x

Borophagus diversidens

cf

Epicyon validus Eucyon davisi

cf

cf

Canis lepophagus Canis

x

x

Vulpes shermanensis Vulpes stenognathus

cf x

x cf

Vulpes Indarctos oregonensis

x x

x

Agriotherium

x

Plesiogulo lindsayi

sp.

x

x

x

sp.

cf

Trigonictis

x

Pliotaxidea nevadensis

sp.

sp.

x

Pliotaxidea garbari

x

Buisnictis schoffi

x

Brachyopsigale dubius

cf

Martes

x

x

Martinogale alveodens

x

?

Taxidea

x

Bassariscus

x

Sthenictis

x

Lutra

x

Mionictis

cf

Machairodus coloradensis

sp.

Barbourofelis fricki

x

Nimravides

x

Felis proterolyncis

cf

sp.

sp.

x

x

Felis longignathus

x

Felis rexroadensis

cf

Felis

cf

x

Cernictis hesperus

?

Pliomastodon Amebelodon fricki

sp.

x

x

x

x

x

cf

cf

x

sp.

x

x

Tapirus Sphenophalos

x

Hipparion forcei

cf

Neohipparion eurystyle

x

x x

Pseudhipparion

?

cf

Dinohippus leidyanus

sp.

Nannippus lenticularis

sp.

Calippus

x

Pliohippus nobilis

x

x

x

sp.

PALUDICOAL, VOL. 8, NO. 1, 2010

Local faunas

Cambridge

White Cone

Redington

Coffee Ranch

Pinole

Pliohippus interpolatus

x

x

Astrohippus ansae

x

21

McKay Reservoir

Mailbox

Santee

Equus simplicidens

cf

Equus (Hemionus) Aphelops kimballensis

x x

cf

Aphelops mutilis Teleoceras schultzi

x x

Teleoceras

x

x

Prosthenops brachirostris Prosthenops graffhami

x

x x

Platygonus Peudoceras

x

Hemiauchenia vera

cf

x

Megatylopus matthewi

cf

cf

cf

cf

sp. sp.

cf x x

Pediomeryx hemphillensis

cf

Pediomeryx

x

Sinocapra willdownsi

x x

Capromeryx Texoceros altidens

x

x

Alforjas taylori

Texoceras guymonensis

x

x

Megacamelus merriami Pediomeryx figginsi

x

x

Prosthenops

Megatylopus gigas

Panaca

x x sp.

cf

x

Paludicola 8(1):22-36 September 2010 © by the Rochester Institute of Vertebrate Paleontology

AN ICHTHYOSAURUS (REPTILIA, ICHTHYOSAURIA) WITH GASTRIC CONTENTS FROM CHARMOUTH, ENGLAND: FIRST REPORT OF THE GENUS FROM THE PLIENSBACHIAN

Dean R. Lomax Doncaster Museum & Art Gallery, Chequer Rd, Doncaster, DN1, United Kingdom, [email protected]

ABSTRACT A well preserved ichthyosaur specimen from the paleontology collection of Doncaster Museum and Art Gallery, England is described with a focus on the gastric material scattered around the ribs and throughout the matrix. The gastric material comprises coleoid cephalopod hooklets. The ichthyosaur has been dated from the Pliensbachian Stage of the Lower Lias rocks of the Charmouth coastline, Dorset, England. The specimen is identified as Ichthyosaurus sp. and thus extends the geologic range of the genus into the Pliensbachian. The specimen comprises a complete skull, articulated vertebrae, ribs and a fully articulated forefin. The specimen may also contain coprolites preserved within the posterior end of the matrix.

fairly complete skull and anterior post-cranial portion, along with scattered bones from the posterior region where the matrix is uneven (Figure 1). The specimen includes gastric material throughout the ventral region and lowest portion of the ribs, as well as isolated material scattered across the posterior on the lightly colored matrix. The matrix also contains numerous bivalves and possible coprolites. The focus of this paper will be on the gastric material which has not been previously recognized. The accession registers, archives, and additional information indicate that the ichthyosaur allegedly derived from the Kimmeridgian (Upper Jurassic) of Dorset. This paper will present evidence that the specimen was more likely from the Lower Lias deposits near Charmouth, Dorset, England.

INTRODUCTION The Doncaster Museum and Art Gallery was originally opened on March 23rd 1910 as The Beechfield House Museum. It officially reopened in 1964 as the Doncaster Museum and Art Gallery for the purpose of displaying the artifacts in the museum’s collections, including several geological specimens. The ichthyosaur that is the focus of this study (DONMG:1983.98) was excavated in the 1970s and originally owned by Hilary Corke Minerals of Surrey, England. The Doncaster Museum purchased it with the aid of a Science Museum grant in 1983. As the specimen was accessioned into the paleontology collections, it was placed on display in the main galleries of the museum. The ichthyosaur was taken off display during the mid to late 1980s due to retirement of the geologist on staff and subsequent changes in personnel. The ichthyosaur (DONMG:1983.98) was erroneously identified as a very good cast but further analysis and comparisons with additional specimens of real and replicated ichthyosaurs and other fossils revealed that it is a well preserved, genuine ichthyosaur. DONMG:1983.98 was identified as Ichthyosaurus communis (Conybeare, 1822), to which it bears a resemblance in the skull, rostrum and paddle morphology (McGowan and Motani, 2003). The specimen consists of a

DETERMINATION OF THE PROVENANCE As the collection information associated with the fossil is unreliable, research was carried out to determine the correct provenance of the specimen. Information obtained from reference collections and catalogued specimens in other museums suggested that the ichthyosaur was not discovered in the Kimmeridgian (Upper Jurassic) but rather it was more likely from the Hettangian, Sinemurian or Pliensbachian stages of the Lower Lias region around Lyme Regis and Charmouth because of the 22

23

LOMAX—ICHTHYOSAUR GASTRIC CONTENTS

FIGURE 1. DONMG:1983.98. Note the dark area of gastric content between the ribs. Scale bar = 20cm .

FIGURE 2. Stratigraphy of the Lower Jurassic Charmouth Mudstone and Blue Lias along the Dorset coast. The last column indicates the geological age in millions of years. ________________________________________________________________________________________________________________________

sheer numbers of ichthyosaurs discovered and described from those strata (Dineley and Metcalf, 1999). The Lower Lias consists of layers of blue and gray argillaceous limestone in the very lowest portion. Woodward (1893) remarked that these layers occur in even and irregular bands, often nodular and interbedded with blue and brown marls, clays and shales. Dineley and Metcalf

(1999) described the layers as consisting of dark and light shales, mudstones and marls. The matrix of DONMG:1983.98 matches the descriptions of both Woodward (1893) and Dineley and Metcalf (1999) suggestive of the Lyme Regis-Charmouth location. However, the extensive repetition of similar lithologies throughout the Lower Lias of this area makes pinpointing the exact horizon difficult. Black Ven, part of the Charmouth

PALUDICOLA, VOL. 8, NO. 1, 2010

24

FIGURE 3. Belemnite specimen in the matrix of DONMG:1983.98, identified as Bairstowius junceus. Scale bar = 6cm.

FIGURE 4. Distal end of DONMG:1983.98. 1. Putative coprolites circled in black, 2. Femur circled in white, 3. The probable ilium circled in white, 4. Several isolated phalanges (estimated 24 in total) circled in white, and 5. Bivalve specimens circled in black. Scale bar = 10cm.

25

LOMAX—ICHTHYOSAUR GASTRIC CONTENTS

coastline, is an illustrious collection site that has yielded thousands of fossil remains and is thus a potential source of the Doncaster Museum specimen. DONMG:1983.98 lies in a calcareous light grayish blue, mudstone matrix, similar to the lithologies in the Black Ven area. There are several formations within the Lower Lias of Charmouth and Black Ven. It is possible that the ichthyosaur originated from one of three layers: from one of two members of the Charmouth Mudstone Formation or from the ‘Blue Lias Formation’ (Figure 2). The Charmouth Mudstone Formation is subdivided into four members (Gallois and Davis 2001; Gallois, 2008a), two of which are discussed here. The Shales-withBeef Member (Sinemurian) consists of thinly interbedded organic-rich mudstones and calcareous mudstones with thin beds of fibrous calcite (‘beef’) and several beds of tabular and nodular limestone. The member crops out in cliffs below Black Ven, along the Charmouth coastline and continues past Lyme Regis (Gallois, 2008b). A second member within the Charmouth Mudstone Formation is the Stonebarrow Marl Member (Pliensbachian), previously known as the ‘Belemnite Marls’. The Stonebarrow Marl Member is exposed in full thickness above The Spittles and Black Ven (Gallois and Davis, 2001). This member consists of bluish-grey mudstones that alternate in darker and lighter bands. It is capped by a thin limestone called the Belemnite Stone, which yields many belemnites (Lang et al., 1928). Reptilian remains are much rarer in this member than the others (House, 1993), but include the only known specimen of Leptonectes moorei (McGowan and Milner 1999). Another possible layer is the ‘Blue Lias Formation’ (Hettangian). The majority of the marine vertebrates for which Lyme Regis is noted have been discovered in the Blue Lias series of the West Cliff and Church Cliffs and in the overlying dark shales and cement-stones of Black Ven (Woodward, 1893). The Blue Lias Formation consists of a sequence of laterally extensive, alternating thin-bedded (and nodular) limestones and shales (Dineley and Metcalf, 1999). The calcareous beds and the nodular and tabular limestones that occur within the beds of the Blue Lias Formation are richly fossiliferous. (Gallois and Paul, 2009; Simms et al, 2004). Its outcrops are exposed at a number of coastal and inland areas across southwest England. The strong similarities between the three layers described above posed a difficult task in

identifying the exact provenance of DONMG:1983.98 based on lithology. Fortunately a small complete belemnite lay in the matrix close to the dentary of the ichthyosaur. This was identified as Bairstowius junceus (P. Doyle, pers. comm., 2009) and is shown in Figure 3. The belemnite species is known only from the Lower Pliensbachian of the Stonebarrow Marls Member, specifically Bed 110, the polymorphus subzone of the jamesoni Zone (Lang et al., 1928; Doyle, 2010). Although the correct age of DONMG:1983.98 has now been determined, the exact locale is unknown. The Stonebarrow Marls are 23 meters thick and extend throughout the coastline between Charmouth and Lyme Regis including an outcrop on the precipices of Black Ven (Lang et al., 1928). Lang et al (1928) described 19 subdivisions, denoted as Beds 103-121. Bed 110, from which DONMG:1983.98 originated, is estimated at 9 m thick. Although the matrix of the specimen is light, the bed is described as a darker marl; however it does include lighter layers (Lang et al., 1928). It is difficult to collect fossils from the Stonebarrow Marls of Black Ven because they are either covered in talus or the bed is high in the precipice, making it more difficult to remove the fossils, although the specimen may have fallen from the precipice and been discovered as a large slab at the base of the cliff. However, there are two other possibilities. Bed 110 of the Stonebarrow Marls Member is found at beach level below Westhay Cliff on the Charmouth coastline east of Stonebarrow Hill. The Stonebarrow Marls are neither as well developed nor as accessible at Black Ven as they are here (Lang et al ., 1928). Another possibility is about two miles to the east of Westhay Cliff on the foreshore west of Seatown, Dorset where Leptonectes moorei was discovered. L. moorei was described from the uppermost part of the Stonebarrow Marls, 1 m below the thin limestone, namely the Belemnite Stone band of the Lower Pliensbachian (McGowan and Milner, 1999). The belemnite genera preserved with L. moorei were identified as Passaloteuthis sp. and Pseudohastites sp. Both genera are only found in the Stonebarrow Marls in several Beds including Bed 110 (Lang et al, 1928; Doyle, 2010). GENERAL DESCRIPTION OF SPECIMEN DONMG:1983.98 has been partially prepared by the removal of excess mudstone matrix, although it may be in need of further preparation. The

PALUDICOLA, VOL. 8, NO. 1, 2010

26

FIGURE 5. Anterior portion of DONMG:1983.98 showing the skull and pectoral girdle. Scale bar = 20cm.

FIGURE 6. Close up of the dentary of DONMG:1983.98. Note that several of the teeth are pointed but have smooth surfaces. Scale bar = 8cm.

mudstone matrix is weak and two structural cracks run across the pelvic area and another above the orbit. The specimen is in a stable condition

however, having been reinforced with plaster prior to its addition to the collections in 1983. This process of stabilizing fossils is now considered

27

LOMAX—ICHTHYOSAUR GASTRIC CONTENTS

dangerous because it can destroy or harm certain specimens (Cifelli, 1996). The specimen has undergone only a minimal amount of damage, although it is unclear whether this occurred after or before the reinforcement. Further removal of matrix could change the structural integrity of the specimen but it may prove necessary to identify the species.

2006), and (3) Temnodontosaurus nuertingensis, a large, fragmentary skull from Germany (von Huene, 1931). The latter is poorly known, and its generic assignment has been questioned (McGowan and Motani, 2003). Isolated bones and fragmentary material have also been reported from the Pliensbachian of Switzerland, United Kingdom, Germany, Belgium, and Canada (Maisch and Reisdorf, 2006).

FIGURE 8. Articulated coracoids and scapulae of DONMG:1983.98. Scale bar = 7cm. _______________________________________________________

FIGURE 7. The right forefin of DONMG:1983.98. Note the short, wide humerus, without a constriction in the shaft. Scale bar = 10cm. _________________________________________________

Pliensbachian ichthyosaurs are rare and poorly known (Maisch and Reisdorf, 2006), with only three species recognized: (1) Leptonectes moorei, a partial skeleton including skull and some postcranial material from England (McGowan and Milner, 1999), (2) Leptonectes tenuirostris, a three dimensional skull and parts of the postcranial skeleton from Switzerland (Maisch and Reisdorf,

DONMG:1983.98 is covered by a synthetic resin that coats the bones and gastric material but not the matrix. This has resulted in the preservation of fine morphological details but also allowed for examination of the matrix itself. The matrix is very fossiliferous and includes belemnites and bivalves, some of which have already proven useful in determining the provenance. Also scattered throughout the matrix are isolated limb material and teeth (Figure 4). The specimen measures 1.1 meters long, a third of which is the skull (Figure 1). The post cranial elements consist of an almost complete string of 15 articulated pre-sacral vertebrae including the atlas-axis, cervicals and dorsals, and 7 associated vertebrae scattered across the specimen. Five articulated preflexural caudal vertebrae are present in an isolated section lying dorsal to the main skeleton, in the upper section of the slab. Some posterior elements including disarticulated femora and a probable ilium, are present in the pelvic region of the specimen. Two isolated articulated dorsal vertebrae occur at the most posterior end of the

PALUDICOLA, VOL. 8, NO. 1, 2010

28

FIGURE 9. DONMG:1983.98. Close up image of the entire mass of gastric contents. Note the large structural crack. Scale bar = 10cm.

FIGURE 10. Comparison of the varieties of hooklets discovered within the gastric contents of several ichthyosaurs reprinted from Pollard, 1968 with permission of the author.

matrix suggesting that there may have been more vertebrae associated with the specimen but not collected. The skull material articulates with a string of associated vertebrae described above. The skull and rostrum exhibit great details. The maxilla is preserved along with an almost complete premaxilla; however, the end of the premaxilla is broken and it is unclear whether or not the rostrum was complete when discovered. Even though the rostrum is incomplete, it appears to be longer than that of Leptonectes moorei. The orbit is visible and some of the sclerotic ring is present, as well as the lachrymal and prefrontal. The posterior section of the skull seems to have undergone slight changes with several elements dislodged or damaged during taphonomic processes, including the anterodorsal edge of the surangular (Figure 5). The rostrum includes eighteen teeth, five of which have dislodged, perhaps during diagenesis (Figure 6). The base of the teeth are quite robust with a conical shape. The teeth of DONMG:1983.98 are pointed, but some have

29

LOMAX—ICHTHYOSAUR GASTRIC CONTENTS

rounded tips because of wear. Worn teeth of ichthyosaurs usually have a polished or pitted apex rather than the rough surface, suggesting abrasion with hard prey (Massare, 1987). The blunt apex and tooth wear of the ichthyosaur suggest that this type of tooth shape was used for grasping prey with a somewhat hard exterior such as the internal shell of cephalopods or the thick scales of fish (Massare, 1987). During the taphonomic processes the right forefin of the specimen has been partially disarticulated and turned over; it is completely associated and includes the humerus, radius, ulna and carpals (Figure 7). The forefin is similar to that of the genus Ichthyosaurus (McGowan and Motani, 2003, figs. 26, 29, 49; Motani 1999a, figs. 3, 6). The humerus of Leptonectes has a constricted shaft and a more expanded distal end then that of DONMG:1983.98 (McGowan and Milner, 1999; Motani, 1999a). The distal end of the forefin is disarticulated but the post axial and accessory digits are still associated. A number of phalanges are spread throughout the specimen, including some anterior to the forefin. The orientation of the ichthyosaur exposes only the ventral side of the humerus of the left forefin. Further preparation may reveal the rest of the left forefin. The pectoral girdle of the ichthyosaur presents articulated coracoids. Both scapulae are present, however only the left is associated with the coracoids by the slight contribution to the glenoid. The clavicle is articulated with the left scapula and the coracoids. The scapulae and coracoids are comparable to the genus Ichthyosaurus in size and structure, although the peripheral edges of the coracoids are more squared than curved (McGowan and Motani, 2003; Motani, 1999b). However, the differences in shape may be due to flattening (Figure 8). The entire post-cranial portion of the specimen is covered by numerous ribs extending towards the rear of the specimen. At the rear, ribs become disarticulated and begin to split into sections. DESCRIPTION OF GASTRIC MATERIAL Preserved stomach contents from several genera of ichthyosaurs were first noted over 150 years ago (Pollard, 1968). Most of the specimens with stomach contents are described from the Lower Jurassic of southern England and southern Germany (Pollard, 1968; Keller, 1976; Böttcher, 1989; Bürgin, 2000), although several ichthyosaur specimens containing gut contents have been discovered in other areas such as Switzerland

(Reiber, 1970), Wyoming (Massare and Young, 2005) and Australia (Kear, et al., 2003). The first description of small cephalopod hooklets associated with ichthyosaur bones was by Coles in 1853, though he misinterpreted the gut contents as ‘setiform or bristly scales’ (Pollard, 1968). In 1856, Moore was able to accurately identify stomach contents of ichthyosaurs, which contained small cephalopod hooklets (Pollard, 1968). Pollard (1968) described stomach contents in three ichthyosaur individuals from the Oxford University Museum and six from the Natural History Museum, London, and reported on a new specimen from the Lower Lias of Lyme Regis. The described gastric material contain fish remains and cephalopod hooklets. Most gastric contents found in ichthyosaurs are cephalopods or fish remains indicating a low variety in their diets (Massare 1987; Massare and Young, 2005). The exception is the described gastric contents of a Cretaceous ichthyosaur that included the remains of a protostegid sea turtle hatchling and an enantiornithine bird, along with fish (Kear et al, 2003). The gastric contents of DONMG:1983.98 is a large dark mass, extending between the 9th and 22nd ribs. The mass measures 19.6cm in length and 11.9cm at its widest point. The whole gastric mass covers a large section around the ribs. Some hooklets are complete and others are compacted as fragments within the mass. Still other sections of the mass are just dark with no identifiable content (Figure 9). Hooklets are widely scattered throughout the specimen and are most visible scattered on the light coloured matrix near the most ventral point of the ribs. The majority of the complete hooklets appear in the vicinity of the ventral region of the mass, between the 14th and 16th ribs, where a high number of small hooklets are evident. The hooklets are 1.7mm or less in size, as estimated from measurements of ten hooklets. Jeletzky (1966) noted that all belemnites of the order Belemnitida possessed eight or ten arms with hooklets attached, and Pollard (1968) concluded that there must have been at least 300 hooklets on a simple ten-armed belemnite from the Lias. The number of hooklets within the gastric material of DONMG:1983.98 is difficult to assess. Pollard (1968) identified four types of hooklets (Figure 10). There are three different types of hooklets in the gastric contents of DONMG:1983.98, here referred to as Type 1, Type 2 and Type 3. Type 1 hooks maintain a very pronounced and tightly curved uncini that has an aciculate point and tight curve that is swept

PALUDICOLA, VOL. 8, NO. 1, 2010

A

B

C

D

30

31

LOMAX—ICHTHYOSAUR GASTRIC CONTENTS

E

F

G

PALUDICOLA, VOL. 8, NO. 1, 2010

32

H

I

FIGURE 11. Cephalopods hooklets discovered on DONMG:1983.98. A. Microscopic view of the most complete Type 1 hooklet identified. Scale bar = 2mm. B. An illustration of a general belemnite hooklet. Note the sharp backward curve. Modified from Engeser, 1988. C. Type 2 hooklet at microscopic view. Note that it has begun to separate into sections, but still maintains the ontogenic shape and structure. Scale bar = 2.5mm. D. A patch of hooklets outside the large dark mass, made visible on the lightly colored matrix, Type 1 and the possible Type 3 (circled in black) hooklets visible. Scale = 2cm. E. Section of gastric content from the center of the mass measured at 5.5cm in length with the patches of hooklets numbered throughout the section. F. A section of the gastric content showing compaction of hooks (circled in white) as well as complete hooklets (circled in black). Scale bar = 10mm. G. Two hooklets (circled in white) fossilized together between the ribs with disassociated fragments of hooklets. Scale bar = 8mm. H. A section of gastric content visible on the distal end of the matrix away from the large tightly packed mass. Only one complete hook identified (Type 1). Scale bar = 4cm. I. A well-preserved Type 1 hooklet with the bivalve remains. Scale bar = 1.5mm.

33

LOMAX—ICHTHYOSAUR GASTRIC CONTENTS

backwards posteriorly from the base, which is flat and straight (Figure 11A). Several of the hooks appear to be crushed and cracks run throughout. Pollard (1968) referred to a similar type of hook shape, his type D hooklet. He indicated that it was extremely rare in the gastric contents of the ichthyosaur specimen (OUM.J.14800) from the Upper Lias mudstone of Dumbleton, Gloucestershire. However the shape of the hook may be ontogenic. Although similar to Pollard’s (1968) ‘type D’, the Type 1 hooklets from DONMG:1983.98 have a much tighter inward curve. However, they have the same sharp backward curved point and a flat, straight shaft and base (Figure 11B). The Type 2 hooklet, of which only two have been identified, have a similar structure to the Type 1 hooks but has a less tightly curved uncini and an irregularly shaped base (Figure 11C). The Type 3 hooklets are very thin and resemble Pollards (1968) Type C hooklet shape. They have a very straight shaft with the uncini (if present) bearing a slight curve. (Figure 11D), although the majority of the Type 3 hooklets are very fragmented. One notable point is that the complete cephalopod hooklets seem to fossilize together in small pockets (Figure 11E). This may reflect the morphology of the gastrointestinal tract. For every one isolated hooklet, there are at least two or three complete hooklets fossilized together in a small assemblage, although compaction of many fragmented sections of hooklets are prominent throughout the specimen (Figure 11F and 11G). Patches of hooklets are spread across the matrix and hooks are visible in the posterior, including the area with the bivalves (Figure 11H, 11I). The variety of different shapes of hooklets may suggest an ontogenetic range of belemnites that were fed upon. The hooklet shape may have changed with growth of the belemnite or the hooklets may have changed shape along the length of the arm. On the other hand, the hooklets shape may reflect a taxonomic variation. An unusual hooklet-bearing cephalopod, Belemnoteuthis mayri, was described from a Solnhofen Limestone in Bayern, Germany (Engeser and Reitner, 1992). The hooklets of Belemnoteuthis are similar to the Type 3 hooklet shape of DONMG:1983.98 in respect to their elongate shaft. The Type 3 hooks are thinner, however, and could be fragmented sections of the Type 1 hooklets. A specimen of Phragmoteuthis from the Austrian Alps was described with arm hooklets that were small (less than 1 mm), slender, and almost straight (Doguzhaeva, et al., 2007). They compare in shape with the Type 3 hooklets.

Pollard (1968) identified the hooklets as belonging to dibranchiate cephalopods and concluded that the hooklets belonged to the family Belemnitidae. Since then, however the hooklets have been reassessed as belonging to belemnoteuthid coleoids (Valente, et al., in press). The taxonomy of the hooklets of DONMG:1983.98 is unknown but the low variety hooklet shapes suggests that they are likely to be from one particular species. It should also be noted that an unidentified fish scale is present in the matrix of DONMG:1983.98, lying 0.5cm ventral to the gastric mass and measuring 1.2cm in length and 0.5cm at its widest point. The scale is probably part of the gastric contents and has been displaced from the general mass in a way similar to the many patches of hooklets which are dispersed throughout the ventral and distal regions of the matrix. (Figure 12). FURTHER DESCRIPTION OF DISPOSITION The matrix at the posterior end of the ichthyosaur exhibits an uneven surface with disassociated bones including small phalanges, hindfin material, an abundance of bivalve remains and possibly coprolites. Coprolites have been described from the Lias of Lyme Regis (Pollard, 1968). The shape and size of coprolites are an important characteristic in identifying the kind of animal that formed them (Chin, 2002). Spiral shaped coprolites were produced by fish with spiral intestinal valves such as sharks (Chin, 2002). Possible coprolites at the distal end of DONMG:1983.98 are elongated, irregular, and have a rounded shape with a rough surface. However, they show a circular pattern, not a spiral. Firtion (1938, as cited in Pollard, 1968) analyzed the contents of coprolites from the Lower Lias of Alsace, France. He found that the undigested contents included a large diversity of marine taxa such as crinoids, gastropods or bivalves, with less abundant foraminifera, ostracods, fish remains and brachiopod shells (Pollard, 1968). Furthermore, the coprolites Firtion studied rarely had spiral folds and were suggested to belong to marine reptiles such as ichthyosaurs (Pollard, 1968). Pollard (1968) disagreed with this, however, because gastric contents such as those described had never been discovered in ichthyosaurs. The possible coprolites on the matrix of DONMG:1983.98 are identified based solely on shapes similar to those described by Firtion (1938, as cited in Pollard, 1968). However no hooklets were discovered in the putative coprolites (Fig 4).

PALUDICOLA, VOL. 8, NO. 1, 2010

34

FIGURE 12. Fish scale directly beneath large gastric mass. Scale bar = 2cm. ________________________________________________________________________________________________________________________

CONCLUSION The identification of one belemnite within the matrix of DONMG:1983.98 enabled the determination of its geologic age as Lower Pliensbachian. However, DONMG:19.83.98 does not appear to belong to any of the ichthyosaur genera described from the Pliensbachian. The humerus, scapula, and coracoid are quite different from the respective elements of Leptonectes and Temnodontosaurus (McGowan and Motani, 2003). The Doncaster specimen is similar to Ichthyosaurus in the paddle and pectoral girdle morphologies (McGowan, 1974; Motani, 1999a), and is herein identified as Ichthyosaurus sp., pending a more complete taxonomic analysis. This is the first report of Ichthyosaurus from the Pliensbachian, and extends the range of the genus. The Doncaster specimen may be a new species. Additional study is needed before the specific taxonomic assignment of the specimen can be determined. The rediscovery of the ichthyosaur has enabled the specimen to be an asset to the paleontology collections of Doncaster Museum and Art Gallery. The discovery demonstrates that small regional museums may house important specimens that can warrant further work. Notable discoveries on DONMG:1983.98 include the gastric mass and the putative coprolites and possibly a new taxon. The low diversity of forms of the hooklets preserved in the gastric contents, suggests that this ichthyosaur

could have fed on a single species of coleoid at a time. The uneven surface at the posterior end of the specimen has thin, rounded coprolite-like shapes that warrant further study. The general preservation of the ichthyosaur is excellent, which will permit a more detailed taxonomic study. Further study of DONMG: 1983.98 may unveil new discoveries.

ACKNOWLEDGEMENTS I am very grateful to the Doncaster Museum and Art Gallery for allowing me to create the Fabulous Fossils exhibit, which led to the rediscovery of the ichthyosaur specimen, and for permission to describe the gastric contents. I would like to thank a list of people for reviews, encouragement and help without which this paper would not have been possible: J. Massare, W. Wahl, E. Maxwell, P. Robinson, M. Bedell Jr, N. Tamura, J. Botting, T. Birkemeier, C. Racay, M. McGrane, R. Motani, E. Veitch, T. Bolton, A. Milner, B. Blessed and K. Chin. Thanks also to A. Hyde for help with photos, J. Pollard for permission to reprint the varied hooklets image and to P. Doyle for review and identification of the belemnite.

35

LOMAX—ICHTHYOSAUR GASTRIC CONTENTS

LITERATURE CITED Böttcher, R. 1989. Über die Nahrung eines Leptoptergius (Ichthyosauris, Reptilia) aus dem süddeutschen Posidonienschifer (Unterer Jura) mit Bemerkungen über den Magen der Ichthyosaurier. Stuttgarter Beiträge zur Naturkunde Ser. B 155:1-19. Bürgin, T. 2000. Euthynotus cf. incognitus (Actinopterygii, Pachycormidae) als Mageninhalt eines Fischsauriers aus dem Posidonienschiefer Süddeutschlands (Unerer Juras, Lias epsilon). Eclogae Geologicae Helvetiae 93:491-496. Chin, K. 2002. Analyses of coprolites produced by carnivorous vertebrates. Pp. 43-49 in M. Kowalewski and P. H. Kelley (eds.), Predation in the Fossil Record. Paleontological Society, Special Paper 8. Cifelli, R. L. 1996. Preparation techniques in vertebrate paleontology. Pp. 77-80, in K. S. Johnson and N. H. Suneson (eds.), Rockhounding and Earth Science Activities in Oklahoma, 1995 Workshop, Oklahoma Geological Survey Special Publication 96-5. Conybeare, W. D. 1822. Additional notices on the fossil genera Ichthyosaurus and Plesiosaurus. Transactions of the Geological Society of London, Second Series 1: 103-123. Dineley, D. and S. Metcalf. 1999. Fossil Fishes of Great Britain, Geological Conservation Review Series, vol. 16, Peterborough, U.K., 675 pp. Doguzhaeva, L. A., H. Summesberger., H. Mutvei and F. Brandstaetter. 2007. The mantle, ink sac, ink, arm hooks and soft body debris associated with the shells in Late Triassic coleoid cephalopod Phragmoteuthis from the Austrian Alps. Nanjing Institute of Geology and Palaeontology, Palaeoworld 16: 272–284 Doyle, P. 2010. Belemnites Pp. 262-275 in A. R. Lord and P. G. Davis (eds.) Fossils from the Lower Lias of the Dorset Coast, Palaeontological Association Field Guides to Fossils No. 13. 436 pp., 78 pls, 78 text-figs. Engeser, T. S. and M. R. Clarke. 1988. Cephalopod hooklets, both recent and fossil. Pp. 332-338 in M. R. Clarke and E. R. Truman (eds.), The Mollusca Vol.12: Paleontology and Neontology of Cephalopods. Academic Press, NY. Engeser, T.S. and J. Reitner. 1992. Ein neues Exemplar von Belemnoteuthis mayri Engeser & Reitner, 1981 (Coleoidea, Cephalopoda) aus dem Solnhofener Plattenkalk (Untertithonium) von Wintershof, Bayern. Archaeopteryx 10:1317.

Firtion, F. 1938. Coprolithes du Lias Inferieur d’Alsace et de Lorraine. Bulletin du Service Carte Geologique d’Alsace et de Lorraine 5. 2743, pls, 4-8. Gallois, R. W. 2008a. Geological controls on the failure mechanisms within the Black VenSpittles landslip complex, Lyme Regis, Dorset. Geoscience in South-West England, 12, 9-14. Gallois, R. W. 2008b. The lithostratigraphy of the Shales-with-Beef Member of the Charmouth Mudstone Formation, Lower Jurassic. Geoscience in South-West England 12: 32-40. Gallois R. W. and G. M. Davis. 2001. Saving Lyme Regis from the sea: Recent geological investigations at Lyme Regis, Dorset. Geoscience in South-west England 10: 183-194. Gallois. R. W. and C.R.C. Paul. 2009. Lateral variations in the topmost part of the Blue Lias and basal Charmouth Mudstone formations (Lower Jurassic) on the Devon and Dorset coast. Geoscience in South-West England 12, 125-133. House, M. R., 1993. Geology of the Dorset Coast (2nd Edition). Geologists' Association, London. No.22. 164 pp. Huene, F. von 1931. Neue Ichthyosaurier aus Württemberg. Neues Jahrbuch für Mineralogie, Geologie und Paläontologie, Abteilung B 65: 305–320. Jeletzky, J. A. 1966. Comparitive morphology, phylogeny and classification of fossil Coleoidea. Paleontological Contributions, University of Kansas, Mollusca 7: 1-162, pls. 1-25. Kear, B. P., W. E. Boles, and E. T. Smith. 2003. Unusual gut contents in a Cretaceous ichthyosaur. Proceedings of the Royal Society, Biological Sciences 270: S206–S208. Keller, T. 1976. Magen- und Darminhalte von Ichthyosaurien des Süddeutschen Posionienschiefers. Neues Jahrbuch für Geologie und Paläontologie, Monatshefte 5: 266-283. Lang, W.D., L. F. Spath, L. R. Cox, and H. M. MuirWood. 1928. The Belemnite Marls of Dorset. Quarterly Journal of the Geological Society 84: 179-222. Maisch, M. W. and A. G. Reisdorf. 2006. Evidence for the longest stratigraphic range of a postTriassic ichthyosaur: a Leptonectes tenuirostris from the Pliensbachian (Lower Jurassic) of Switzerland. Geobios 39:491-505. Massare, J. A. 1987. Tooth morphology and prey preference of Mesozoic marine reptiles. Journal of Vertebrate Paleontology 7: 121-137. Massare, J. A. and H. A. Young. 2005. Gastric contents of an ichthyosaur from the Sundance Formation (Jurassic) of central Wyoming. Paludicola 5: 20-57.

PALUDICOLA, VOL. 8, NO. 1, 2010

McGowan, C. and A. C. Milner. 1999. A new Pliensbachian ichthyosaur from Dorset, England. Palaeontology 42: 761-768. McGowan, C. and R. Motani. 2003. Handbook of Paleoherpetology, Part 8 Ichthyopterygia. Verlag Dr. Friedrich Pfeil, München, 175 pp. McGowan, C. 1974. A revision of the latipinnate ichthyosaurs of the Lower Jurassic of England (Reptilia: Ichthyosauria). Life Sciences Contributions, Royal Ontario Museum 100: 130. Motani R. 1999a. On the evolution and homologies of ichthyopterygian forefins. Journal of Vertebrate Paleontolog 19: 28-41. Motani, R. 1999b. Phylogeny of the Ichthyopterygia. Journal of Vertebrate Paleontology 19: 473-496. Pollard, J. E. 1968. The gastric contents of an ichthyosaur from the lower Lias of Lyme Regis. Palaeontology 11: 376-388, pls. 72-73.

36

Reiber, H. 1970. Phragmoteuthis ?ticinensis n. sp., ein Coleoidea-Rest aus der Grenzbitumenzone (Mittlere Trias) des Monte San Giorgio (Kt. Tessin, Schweiz). Paläontologische Zeitschrift 44: 32-40. Simms, M.J., N. Chidlaw, N. Morton, and K. N. Page. 2004. British Lower Jurassic Stratigraphy, Geological Conservation Review Series, No. 30, Joint Nature Conservation Committee, Peterborough, 458 pp. Valente, D. E., A. L. Edwards and J. E. Pollard. In Press. Reappraisal of the gastric contents of a Lower Jurassic ichthyosaur. Geological Curator 9, part 3. Woodward, H. B. 1893. The Jurassic Rocks of Britain, Vol. 3, The Lias of England and Wales (Yorkshire excepted). Memoirs of the Geological Survey of Great Britain, London. 399 pp.

Paludicola 8(1):37-48 September 2010 © by the Rochester Institute of Vertebrate Paleontology

FOSSIL CHIMAEROID REMAINS (CHONDRICHTHYES: HOLOCEPHALI) FROM WILLIAMSBURG COUNTY, SOUTH CAROLINA, USA

David J. Cicimurri Campbell Geology Museum, Clemson University, Clemson, South Carolina 29634

ABSTRACT Three fossil holocephalian tooth plates have been recovered in Kingstree, Williamsburg County, South Carolina. All of the fossils were collected from a lag deposit containing a temporally mixed vertebrate assemblage. Two specimens, an incomplete left mandibular tooth plate and an incomplete left palatine tooth plate, are Edaphodon and compare favorably to E. mirificus. The third specimen is an incomplete and highly abraded right mandibular tooth plate from a very young individual that is questionably referred to Edaphodon. The tooth plates were associated with Cretaceous shark and dinosaur teeth, Paleocene shark and crocodilian teeth and turtle bones, and Plio-Pleistocene shark teeth and terrestrial mammal remains. The source of the Cretaceous fossils is arguably from Maastrichtian (late Cretaceous) strata (i.e., Peedee or Steel Creek formations), whereas Paleocene fossils are likely derived from the Danian (lower Paleocene) Rhems Formation. These fossils were probably concentrated together during PlioPleistocene sea level highstand, at which time the younger vertebrate material was deposited.

Fossils from Clapp Creek came to the attention of Rudy Mancke (then Curator of Natural History at the South Carolina State Museum) in the mid 1980s, and soon thereafter he alerted Bruce Lampright (then of Coastal Carolina University) to the fossil deposits to be found there. Lampright told Aura Baker (former president of the Myrtle Beach Fossil Club) about the deposit and she encouraged club members to collect at the site. Both she and Lampright ultimately contributed significant collections to the SC State Museum. Some of the fossil species occurring at the Kingstree site have been discussed in the literature (i.e., Briedis and Knight, 1996; Erickson, 1998; Hutchison and Weems, 1998; Knight et al., 2007) and the Baker and Lampright collections are receiving renewed interest by the present author. This collection has proven to be paleontologically significant, as it contained one of the few records of Schizorhiza stromeri from North America (Knight et al., 2007), as well as some of the few dinosaur remains (hadrosaurian teeth) from the state (Erickson et al., in press). In a review of the State Museum collection, several more unexpected vertebrate occurrences were noted (currently under study), including the three holocephalian remains that form the basis of this report. The purpose of this paper is to provide a detailed description of the fossils, which represent the first record of Edaphodon and only the third, fourth and fifth holocephalian remains reported from the state.

INTRODUCTION Some of the geologic history of the South Carolina Coastal Plain is preserved as a complex stratigraphic sequence that is far from completely understood. There were many transgressions and regressions of the Atlantic Ocean over the last 75 million years, and sediment composition varies greatly both vertically (time) and laterally (geography). This lithologic variation makes correlation of widely separated areas difficult, especially when units were eroded during marine transgressive/regressive events or terrestrial fluvial processes. The local presence of a particular, generally widely distributed formation is therefore not necessarily predictable. Often, contacts between lithologic units are marked by a conspicuous lag deposit, and in South Carolina fossils of greatly different ages can be found mixed together. One such example is found along the banks of Clapp Creek within the city limits of Kingstree, Williamsburg County, South Carolina (Figure 1). Fossils recovered from a thick lag at the site include a mixture of late Cretaceous, early Paleocene, and Plio-Pleistocene taxa. The Cretaceous component contains a variety of terrestrial (dinosaur) and marine (elasmobranchs) animals, as do the Paleocene (crocodilians, trionychid and chelonioid turtles, elasmobranchs) and PlioPleistocene components (equids and proboscideans, elasmobranchs and cetaceans).

37

38

PALUDICOLA, VOL. 8, NO. 1, 2010

I also discuss the potential ages and stratigraphic provenances of these chimaeroid fossils. __________________________________________

terminology for tooth plate morphology follows Stahl and Parris (2004). Measurements of the fossils were taken with Marathon digital calipers. Stahl (1999) included Callorhynchidae, Chimaeridae, and Rhinochimaeridae within Chimaeriformes, placing extinct taxa like Edaphodon, Ischyodus and others into two subfamilies (Callorhynchinae and Edaphodontinae) within Callorhynchidae (see also Stahl and Chatterjee, 2002; Parmley and Cicimurri, 2005; Takeuchi and Huddleston, 2006; Shin, 2010). However, these taxa are more often placed in the extinct Edaphodontidae (i.e., Ward and Grande, 1991; Nessov and Averianov, 1996; Popov and Beznosov, 2006; Popov and Shapovalov, 2007; Popov, 2008) and this classification is followed here. SYSTEMATIC PALEONTOLOGY Elasmobranchii Bonaparte, 1838 Chimaeroidei Patterson, 1965 Edaphodontidae Garman, 1901 Edaphodon Buckland, 1838 Edaphodon sp. cf. E. mirificus Leidy, 1856 Figures 2 and 3

FIGURE 1. A. Outline map of contiguous USA showing some southeastern states. B. View of central South Carolina coastal region showing outcrop belts of coastal plain strata. A modified from Case (1994) and B adapted from Weems and Bybell (1998). _______________________________________________________

MATERIALS AND METHODS The three specimens described herein are housed in the South Carolina State Museum (SC), and all were collected from Clapp Creek, approx. 75 m downstream from a bridge on Lawrence Street, Kingstree, Williamsburg County, South Carolina. Specimens (or photographs of specimens) from the following institutions have been examined: Academy of Natural Sciences (ANSP), Philadelphia, Pennsylvania; New Jersey State Museum (NJSM), Trenton, New Jersey; Sternberg Museum of Natural History (FHSM), Hays, Kansas; American Museum of Natural History (AMNH), New York; and Campbell Geology Museum (BCGM), Clemson, South Carolina. Descriptive

Material Examined—SC83.89.19, incomplete left mandibular tooth plate (Figure 2); SC87.158.150, incomplete left palatine tooth plate (Figure 3). Description—The mandibular tooth plate (SC83.89.19) is missing much of the mesial beak, at least one third of the distal end, and all of the tritor pads. However, the remaining portion is well preserved, with the mesodistal length measuring 67.90 mm and maximum labiolingual width measuring 23.79 mm. The plate is laterally compressed, with the area immediately mesial to the anterior outer tritor forming the beginning of the beak (Figure 2). The left and right mandibular plates articulated along a symphyseal surface, of which only a length of 32.70 mm is preserved (Figure 2A-B). The distal end of the plate appears to diverge at the point immediately distal to the end of the symphyseal surface (Figure 2E-F). The labial face is weakly convex flat dorsally, concave medially, and apparently convex ventrally, with the entire surface finely striated parallel to the plate’s length (Figure 2C). As preserved, the ventral aspect of the lingual face (below the middle tritor) is flat and relatively smooth, but it appears that the middle tritor constituted the upper half of the face distal to the beak (Figure 2A-B). Although no tritoral tissue remains, traces of four tritors are preserved as cancellous-textured and laminar-textured attachment surfaces on the mandibular dentine. The attachment surface of the symphyseal tritor spans the entire length of the labioventral face of the tooth plate (Figure 2C-D), and the

CICIMURRI—CHIMAEROID FROM SOUTH CAROLINA

39

FIGURE 2. SC83.89.19, Edaphodon sp. cf. E. mirificus left mandibular tooth plate. A-B, specimen (A) and interpretive drawing (B) in lingual view (mesial at right). C-D, specimen (C) and interpretive drawing (D) in labial view. E-F, specimen (E) and interpretive drawing (F) in oral view (mesial at left). In B and D, dotted regions indicate locations of tritors and hatched regions indicate broken surfaces. Abbreviations: aot, anterior outer tritor; mt, middle tritor; pot, posterior outer tritor; S, symphysis; st, symphyseal tritor. Scale bars = 10 mm.

40

PALUDICOLA, VOL. 8, NO. 1, 2010

FIGURE 3. SC87.158. 150, Edaphodon sp. cf. E. mirificus left palatine tooth plate. A-B, specimen (A) and interpretive drawing (B) in oral view (mesial at right, labial at bottom). C, distal view (labial at left). D, aboral view (labial at top). E, mesial view (labial at right). Dotted areas in B indicate locations of tritors. Abbreviations: ait, anterior inner tritor; ot, outer tritor; pit, posterior inner tritor. Scale bars = 10 mm. ________________________________________________________________________________________________________________________

dorsal aspect of this surface is cancellous, whereas the ventral aspect bears numerous closely spaced (1 mm apart) vertical laminae. An ovoid anterior outer tritor measured 10.05 mm long x 5.26 mm wide and was situated on a prominence located labially and at the mesial end of the middle tritor (Figure 2B, D, F). The tritor tissue immediately distal to the exposed portion was covered by dentine, but is now preserved as a hollow circular tube located lateral to the middle tritor (separated by as little as 2.5 mm of dentine) and seen at the broken distal surface (29.48 mm from the distal end of the exposed portion of the tritor, 22.90 mm below the dorsal margin). The exposed portion of an ovoid, elongated posterior outer tritor measured 18.89 mm long x 5.91 mm wide. This tritor was situated on a higher prominence on the labial margin, adjacent to the middle tritor (Figure 2B, D, F). Immediately distal to the exposed portion, the tritor tissue was covered by a thin layer of dentine (up to 3 mm thick), but as preserved the location of the tissue is indicated by a hollow circular tube exposed at the broken distal end of the plate. Based on the preserved attachment surface, a very large middle tritor was pointed mesially and separated from the outer tritors by 3 to 4 mm of dentine, and this surface measures 49.26 mm long and 21.43 mm wide (Figure 2B, F). At the distal one quarter of the posterior outer tritor, the dentine

overhangs the middle tritor attachment surface (Figure 2A), indicating that the middle tritor pad was covered by dentine in this area (i.e., only the mesial 38.67 mm of the middle tritor pad was exposed in life). In labial view the oral margin of the labial face has a sigmoidal outline due to the prominences (Figure 2C-D). The palatine plate (SC87.158.150) is incomplete, missing an unknown portion of the mesial end and the entire postoral surface (some of the distal oral surface is also absent). As preserved, the specimen has a rectangular appearance in oral/aboral view (Figure 3AB, D) and measures 42.18 mm long x 26.27 mm at its widest (the plate is slightly narrower mesially at 22.37 mm). Only 27.36 mm of the symphyseal border is preserved, and this surface is flat and nearly vertical. Although the labial margin is damaged, it appears to flare outward slightly at approximately the middle of the preserved portion. As is the case with the mandibular plate, no tritoral tissue is preserved on the palatine plate, but the attachment surfaces of three tritors occupy nearly the entire oral surface (Figure 3AB). The attachment surface of the outer tritor is located along the labial margin and measures 37.72 mm long x 10.86 mm at its widest. However, it appears that only the mesial 22.22 mm of tritor was exposed as a triangular pad (dentine is broken away where it would have covered part of the tritor tissue). Attachment

CICIMURRI—CHIMAEROID FROM SOUTH CAROLINA

surfaces of the anterior and posterior inner tritors are found along the symphyseal margin. As preserved, 18.25 mm of the anterior tritor was exposed, but at least 21.84 mm of tissue extended into the palatine dentine (only an ovate hollow tube remains above the posterior inner tritor, visible at the broken mesial and distal ends of the plate; Figure 3C, E). The posterior inner tritor was located immediately distal to the exposed anterior inner tritor, and 30.23 mm of the attachment surface is preserved. The outer tritor and anterior inner tritor appear to have been relatively narrow, but because the dentine is incomplete it is unclear if the posterior inner tritor was broader than preserved. In aboral view, unabraded external surfaces are smooth (Figure 3D). A 6.13 mm high x 7.58 mm wide convex ridge is located above the outer tritor and a 5.06 mm high x 5.43 mm wide sharp ridge is located over the anterior outer tritor, and these ridges are separated by a U-shaped furrow (Figure 3C, E). Remarks—Within Chimaeriformes, only Callorhynchidae and Edaphodontidae have a fossil record in North America, but comparison of the South Carolina material was restricted to Edaphodontidae because Callorhynchidae is only known from egg cases (Hussakof, 1912; Brown, 1946; Obruchev, 1967; Stahl, 1999). Several edaphodontid genera have been reported from Cretaceous deposits of North America, including Edaphodon, Eumylodus, Ischyodus, and Leptomylus (i.e., Cope, 1869; Case, 1978; Stahl and Parris, 2004; Cicimurri et al., 2008), but only Edaphodon and Ischyodus have a Paleocene or younger record (i.e., Cope, 1875; Case, 1996; Hoganson and Erickson, 2005; Parmley and Cicimurri, 2005). The mandibular plates of Eumylodus are unknown, but those of Edaphodon, Ischyodus and Leptomylus are similar in overall morphology. Mandibular plates of Leptomylus (ANSP 9440) are very unusual in that outer tritors are lacking and only a very narrow middle tritor is generally developed (see also Hussakof, 1912; Stahl, 1999). Mandibular plates of Ischyodus and Edaphodon can be difficult to distinguish from one another (Stahl, 1999) if the remains are incomplete, but Ischyodus mandibular plates have four or more tritors (i.e., Case, 1978; Ward and Grande, 1991; Popov, 1999a; Hoganson and Erickson, 2005). Although SC83.89.19 is incomplete, there is no indication of large crushing pads other than the anterior and posterior outer and middle tritors. The symphysis of SC83.89.19 is elongated and flat like Edaphodon mandibular plates I examined from New Jersey (i.e., NJSM 11362; see also Stahl and Parris, 2004) and Georgia (SC2004.34.1; see also Parmley and Cicimurri, 2005), a feature not observed on mandibular plates of Ischyodus or Leptomylus that I examined (also Hussakof, 1912; Case, 1978; Ward and Grande, 1991; Stahl, 1999; Hoganson and Erickson, 2005). I

41

conclude that SC 83.89.19 is referable to the genus Edaphodon. Although incomplete, SC87.158.150 (palatine tooth plate) was compared to the palatine plate of Eumylodus (FHSM VP-16685), which has four diminutive and widely separated tritors, wide oral region, and lacks a deep longitudinal sulcus on the aboral surface (Cicimurri et al., 2008). The aboral surface of a Leptomylus densus palatine plate (ANSP 9441) is relatively flat and smooth, and there are only two narrow and elongated tritors on the oral surface (inner and outer; see also Hussakof, 1912; Stahl, 1999). Palatine plates of Ischyodus and Edaphodon are morphologically similar, but Ischyodus plates bear four large tritors on the oral surface, and there may be accessory tritors along the labial margin (Ward and Grande, 1991; Popov, 1999a; Stahl, 1999; Hoganson and Erickson, 2005). In contrast, Edaphodon palatines contain three large tritors (i.e., Stahl and Parris, 2004; Shin, 2010), a fourth is only occasionally present (Ward, 1973), and there are no accessory tritors (Stahl, 1999). SC87.158.150 is incomplete, but there are no indications of tritors other than the anterior and posterior inner tritors and outer tritor. Whereas the aboral surface of Ischyodus is generally flat and exhibits a descending lamina, SC87.158.150 and palatine plates of Edaphodon exhibit a deep longitudinal sulcus and lack a descending lamina (see Patterson, 1992; Popov, 1999a; Stahl, 1999). These features lead me to assign SC87.158.150 to Edaphodon. The small sample size, incompleteness of the material, and lack of stratigraphic control limits the accuracy of species determination for SC83.89.19 and SC87.158.150. Edaphodon has a long temporal distribution (Cretaceous to Pliocene) and was widely distributed geographically (Stahl, 1999). Complete dentitions are known for only a few species (i.e., E. hesperis, Edaphodon mirificus, E. sedgwicki), and many fossil edaphodontids are known only from isolated tooth plates (Stahl, 1999; Duffin, 2001). Tooth plate morphology varies greatly between species, with age (ontogeny; see Hussakof , 1912 and Ward and Grande, 1991), and even within the mouth of a single individual (Stahl and Parris, 2004). This inter- and intra-specific variation adds to the difficulty of identifying the Kingstree Edaphodon species. The post-Cretaceous Edaphodon record of the USA very poorly known. However, SC83.89.19 (left mandibular) differs from the Late Eocene Edaphodon sp. reported by Parmley and Cicimurri (2005), as well as from lower Paleogene European taxa like E. leptognathus (Agassiz, 1843) and E. bucklandi (Agassiz, 1843), in that the middle tritor does not appear to have been bifurcated (see also Kemp et al., 1990; Consoli, 2006). The labial face of SC83.89.19 is

42

PALUDICOLA, VOL. 8, NO. 1, 2010

convex dorsally and concave ventrally, whereas the incomplete mandibular of E. eocaenus (AMNH 7205; see also Cope, 1875) from New Jersey exhibits a more evenly convex labial face. Considering those Edaphodon species reported from US Cretaceous strata, a large number, primarily from Maastrichthian (late Cretaceous) greensands of New Jersey, have been described. Hussakof (1912) synonymized numerous species erected by Cope (1869, 1875) with E. mirificus (i.e., fecundus, incrassatus, longirostris), and Stahl (1999) recognized an additional four species: E. agassizi (Buckland, 1835), E. latigerus (Cope, 1869), E. sedgwicki (Agassiz, 1843), and E. stenobyrus (Cope, 1875). All of these species occur in Maastrichtian greensands of New Jersey, but unfortunately the stratigraphic position of much of the original material reported by Cope (1869, 1875) is imprecisely known. Although the tooth plates have varied morphologies, the species may someday prove to be conspecific; the possibility of multiple coeval species cannot be ruled out, but this may have been unlikely (see also Popov, 2008) because the great similarities in tooth plate morphologies indicates similar prey preferences, and therefore intra- and interspecific competition for food resources. Hussakof (1912) synonymized Cope’s (1869) Ischyodus smocki (see also Fowler, 1911) with the European species Edaphodon agassizi, and SC83.89.19 differs from Cope’s type mandibular (AMNH 7192) in having a smaller anterior outer tritor, sloping as opposed to horizontal oral margin from this tritor to the posterior outer tritor, and the dorsal margin of the middle tritor is convex, not concave. However, European E. sedgwicki occurs in “middle” Cretaceous (Cenomanian and Turonian) strata, and these mandibulars have a narrower but more elongated middle tritor than SC83.89.19, and the posterior outer tritor is also more elongated (see also Patterson, 1992; Duffin and Reynders, 1995; Popov, 2008). In labial view, the type of E. latigerus (AMNH 2238) appears very elongated compared to SC83.89.19 due to the much greater length of the anterior prominence (for the anterior outer tritor). In addition, the posterior outer tritor is much more elongated, and the middle tritor is narrower and more elongated than SC83.89.19 (see also Hussakof, 1912; Duffin and Reynders, 1995). Although it is broken, the location of the symphyseal tritor of SC83.89.19 indicates that the tooth plate was not nearly as deep as the type mandibular of E. stenobyrus (AMNH 7204). Additionally, the middle tritor of E. stenobyrus is much smaller than that of SC83.89.19 (see also Hussakof, 1912; Duffin and Reynders, 1995). Hussakof (1912) synonymized Cope’s (1875) Ischyodus tripartitus (see also Fowler, 1911) with the European species E. sedgwicki. This species has a very elongated beak, a feature not

preserved in SC83.89.19. However, both the US and European mandibulars assigned to E. sedgwicki bear a median tritor with two or more divisions (Cope, 1875; Hussakof, 1912; Stahl, 1999). In contrast, the middle tritor of SC SC83.89.19 appears to have been large and without division. The mandible of E. barberi Applegate 1970 from the Campanian of Alabama differs significantly from SC83.89.19 in the morphology and locations of the tritors, the labial margin from the beak to the posterior outer tritor is arcuate, and the symphyseal surface is indistinct. The South Carolina specimen also differs from the large mandibular plates of E. hesperis (Campanian of Vancouver Is., Canada) in that the anterior and posterior outer tritors are situated more mesially with respect to the middle tritor (see Shin, 2010). With respect to the depth of the mandibular plate, length of the anterior and posterior prominences, location and morphology of the attachment surfaces for the tritor tissues, and the shape of the symphyseal surface, SC83.89.19 is more similar to mandibular plates of E. mirificus (see also Fowler, 1911; Hussakof, 1912; Stahl, 1999; Stahl and Parris, 2004) than to the other species mentioned above. In addition, SC87.158.150 is morphologically inseparable from palatine plates of E. mirificus that I examined (i.e., ANSP 5481 and 5825, NJSM 11362). ? Edaphodon Figure 4 Material Examined--SC.158.151, incomplete right mandibular tooth plate. Description—The specimen is the smallest fossil chimaeroid mandibular tooth plate I have encountered. Some of the mesial end and an unknown portion of the distal end are missing, and the tritor tissue is not preserved. The remaining section is highly abraded (post-mortem taphonomic processes). As preserved, the mesodistal length measures only 13.4 mm, with maximum labiolingual width measuring 4.85 mm. The plate is laterally compressed, with the mesial-most area appearing to form the beginning of a beak (Figure 4A, C, D). The labial face is smooth and featureless, and the distal end does not appear to diverge as is does on SC83.89.19. Two tritors form the dorsal aspect of the lingual face, whereas the ventral aspect exhibits a mesially located circular fossa and a more distally located longitudinal furrow below the lower tritor. As is the case with SC83.89.19, no tritoral tissue is preserved on SC.158.151, but four tritors are indicated by cancellous-textured and laminar-textured surfaces on the mandibular dentine. A symphyseal tritor spans the entire length of the tooth plate (Figure 4A-B) but is best seen linguo-ventrally along the mesial half of the tooth plate as closely spaced (0.5 mm

CICIMURRI—CHIMAEROID FROM SOUTH CAROLINA

43

FIGURE 4. SC.158.151, ? Edaphodon sp. right mandibular tooth plate. A-B, specimen (A) and interpretive drawing (B) in lingual view (mesial at left). C, oblique lingual view showing anterior outer and middle tritor. D-E, specimen (D) and interpretive drawing (E) in oral view (mesial at left). Abbreviations as in Figure 2, with the addition of vmt, ventral pad of middle tritor. Scale bars = 2 mm. ________________________________________________________________________________________________________________________

apart) vertical laminae separated by matrix-filled spaces. As preserved, the surface of an anterior outer tritor measures 6.62 mm long x 1.63 mm wide, but a slight medial constriction indicates that only the mesial half of the tritor tissue functioned as a crushing pad. This tritor is situated on an indistinct prominence located along the labial margin, directly adjacent to and separated from a middle tritor by 1.02 to 1.68 mm of dentine (Figure 4D-E). The surface for the middle tritor is sinuous and extends along the entire dorsal margin of the lingual face, measuring 8.51 mm long x 2.89 mm wide (Figure 4A-B). It is not clear how much of the tritor tissue was exposed in life. An additional tritor is located below the middle tritor and separated from it by only 0.89 mm of dentine (Figure 4A-C). The surface for this lower tritor measures up to 7.99 mm long and 1.32 mm wide, but broken dentine indicates that only the mesial 3.22 mm of tritor tissue was exposed in life.

Remarks—SC.158.151 is highly abraded from post-mortem taphonomic processes, and I have not previously encountered a mandibular plate this small. The laminated symphyseal tritor (Figure 4B) leads me to assign SC.158.151 to Edaphodontidae (see Ward and Grande, 1991). The tritor below the middle tritor (Figure 4A-B) could be interpreted as a posterior inner tritor as seen in Ischyodus bifurcatus. The large tritor pad seen on I. bifurcatus mandibular tooth plates has been interpreted as being formed from the fusion of the middle tritor with the posterior inner tritor (Case, 1978; and Case and Schwimmer, 1992), and SC.158.151 could therefore represent an ontogenetic stage within the species where the two tritors have not yet fused. However, there is no indication of an anterior inner tritor near the ventral margin as in I. bifurcatus, I. dolloi, and Ischyodus rayhaasi (see Case and Schwimmer, 1992; Stahl, 1999; Hoganson and Erickson, 2005), and no descending lamina is visible, suggesting that the plate is not Ischyodus.

44

PALUDICOLA, VOL. 8, NO. 1, 2010

This tritor may simply have been exposed as an accessory pad below the middle tritor, and this feature is also seen along the ventral margin of the middle tritor of SC83.89.19 (Edaphodon). In Figure 2B, note that there is a small projection of pleromin (not preserved) into the symphyseal surface, and a thin layer of dentine would separate this pleromin from the main body of the middle tritor. A similar thin pleromin body can be seen on mandibular plates of E. leptognathus and E. bucklandi (see also Kemp et al., 1990), but it is unclear if these pleromin bodies on SC83.89.19 and SC.158.151occurred as an elonageted, functional tritor or were covered by dentine. These do not appear to be equivalent to the inner tritor of E. mantelli as identified by Popov (1999b). The apparently large area occupied by the tritoral tissue on SC.158.151 is likely related to the very young age of the individual, as Ward and Grande (1991) noted that the area decreases with age (as tooth plate size increases). An apparent symphysis, that is, the lingual margin extending from the symphyeal tritor to the middle tritor, is slightly concave and somewhat flattened (similar to Edaphodon), but it is difficult to ascertain if this latter feature is natural or the result of post-mortem abrasion. Even though SC.158.151 is incomplete and highly abraded, its morphology is more consistent with Edaphodon than Ischyodus, and this specimen is therefore tentatively attributed to a very young Edaphodon sp. DISCUSSION Other Chimaeroid Fossils from South Carolina—Unfortunately, of five South Carolina chimaeroid fossils known to us, none are stratigraphically well constrained. The ablated tritor noted by Cicimurri (2007; BCGM 7007) has a rectangular outline in occlusal view, being longer (27.42 mm) than wide (13.64 mm), and I believe that it was located on a palatine tooth plate of Edaphodon or Ischyodus. The specimen was recovered from a lag deposit at the base of the Peedee Formation (middle Maastrichtian), but Cicimurri (2007) believed that the tritor was reworked from the underlying late Campanian Donoho Creek Formation (Black Creek Group). The only chimaeroid remains previously reported from Black Creek Group deposits were attributed to Ischyodus bifurcatus (Robb, 1989), but BCGM 7007 can only be referred to Edaphodontidae indeterminate. The mandibular tooth plate fragment attributed to Ischyodus by Purdy (1998) was collected at St. Stephen, Berkeley County, 30 km SSW of Kingstree. This specimen, ChM PV3899, may have been derived from the upper Paleocene Williamsburg Formation (calcareous nannofossil Zones NP 4 – NP 9), but it was collected as float and its precise stratigraphic position

is uncertain; it is not out of the realm of possibility that it originated from the underlying Rhems Formation, which was also exposed at the site. As illustrated by Purdy (1998; Figure 6C), the specimen lacks tritor tissue, but the preserved attachment surface clearly exhibits laminated texture (at the top of the photograph), indicating a laminated tritor. In Edaphodontidae, only symphyseal tritors are laminated (Ward and Grande, 1991), and ChM VP3899 could therefore represent a taxon other than Ischyodus (i.e., Edaphodon). I have examined this specimen and consider it best to identify it as Edaphodontidae indeterminate because it is too fragmentary for even a generic assignment. Age of the Kingstree Chimaeroid Fossils and Stratigraphic Provenance—Although SC87.158.150 (palatine tooth plate) and SC83.89.19 (mandibular tooth plate) are incomplete, they may not have been subjected to extensive transport because they are relatively unabraded. As noted earlier, the fossils recovered from the Kingstree lag represent a mixture of Cretaceous, Paleocene, and Pliocene taxa. If SC87.158.150 and SC83.89.19 do represent Edaphodon mirificus, Maastrichtian-aged source strata are indicated, as the species is well known from Maastrichtian greensands of New Jersey (i.e., Hussakof, 1912; Stahl and Parris, 2004). This age is supported by the association of Schizorhiza stromeri (see Knight et al., 2007), Rhombodus binkhorsti, and Serratolamna serrata, typically Maastrictian species. The latter two elasmobranch taxa are thus far only known to occur within the Maastrichtian Peedee Formation of South Carolina (DJC unpublished data; see also Case, 1979 for North Carolina records). I conclude that the Cretaceous-aged fossils were derived from strata of Maastrichtian age, and the Peedee Formation has been reported as occurring in the Kingstree area (Weems and Bybell, 1998). The Peedee Formation was deposited in a shallowing upward, outer to inner neritic environment within calcareous nannofossil Zones CC25 and CC26 (69.5-65.6 Ma; see Edwards et al., 2000; Thibault and Gardin, 2006). The occurrence of dinosaur teeth in the Kingstree lag must also be considered, as dinosaur remains are unknown from the Peedee Formation. Dinosaur and shark teeth are found mixed together in deltaic deposits of the Campanian Black Creek Group in Florence and Darlington counties, 40 to 60 km NNE of Kingstree (Cicimurri, 2007; DJC unpublished data), and it is possible that some of the Kingstree fossils were derived from strata of that age. However, the Maastrichtian Steel Creek Formation, temporally equivalent to the Peedee Formation and representing deltaic deposition (Fallaw and Price, 1995), has been identified in the Turbeville area of Clarendin Cocunty, only 30 km northwest of Kingstree in Clarendon

CICIMURRI—CHIMAEROID FROM SOUTH CAROLINA

County (Erickson et al., in press). It is most parsimonious to interpret the mixture of Cretaceous terrestrial and marine taxa in the Kingstree lag as having been derived from a Maastrichtian nearshore marine (deltaic) deposit that contains dinosaur and elasmobranch remains, as opposed to having been derived from a Campanian-aged horizon or being a mixture of Campanian and Maastrichtian fossils. The Paleocene fossils occurring at Kingstree were reported as being derived from the lower Paleocene (Danian; calcareous nannofossil zone NP 1; possibly as young as NP 3) Rhems Formation (65.5 to 61 Ma; see Weems and Bybell, 1998; Erickson, 1998; Hutchinson and Weems, 1998; Waga et al., in press), and this interpretation is supported by some of the associated batoid fossils (DJC unpublished data). The Rhems Formation is locally exposed in the Kingstree area, and a nearshore marine, possibly deltaic environment is indicated both lithologically and biologically (Erickson, 1998; Sawyer, 1998; Weems and Bybell, 1998; Hutchinson and Weems, 1998). Although Edaphodon is known from PlioPleistocene strata of Europe and Australia (Stahl, 1999; Consoli, 2006), no material has been reported from the USA. The Kingstree lag deposit probably accumulated during sea level highstand when a fluvial system emptied into the sea in the Kingstree area. Coastal erosional processes resulted in the mixing of the Cretaceous and Paleocene fossils along with PlioPleistocene terrestrial (horse) and marine animals (i.e., whales, sharks and rays). CONCLUSIONS The first records of Edaphodon from South Carolina consist of an incomplete left mandibular tooth plate, incomplete left palatine tooth plate, and incomplete right mandibular of a very young individual. The morphologies of the former two specimens are most similar to equivalent elements of E. mirificus from the Maastrichtian of New Jersey. These chimaeroid fossils were recovered from a lag deposit containing a temporally mixed assortment of fossils, including taxa from the Cretaceous, Paleocene, and Plio-Pleistocene. The Cretaceous fossils are indicative of a Maastrichtian age, and source strata belonged to the Peedee Formation or temporally equivalent Steel Creek Formation. Paleocene fossils indicate a Danian (lower Paleocene) age, and the likely source strata are from the Rhems Formation. These older fossils were probably concentrated together during Plio-Pleistocene sea level high-stand.

45

ACKNOWLEDGMENTS Aura Baker and Bruce Lampright donated parts of their Kingstree collections to SC. Billy Palmer and Lampright provided historical information regarding the discovery of the site, and Chuck Ciampaglio (Wright State University, Celina, Ohio) shared his knowledge of the various species he has collected there. Jim Knight (SC), David Parris (New Jersey State Museum, Trenton), Ned Gilmore and Ted Daeschler (Academy of Natural Sciences, Philadelphia, Pennsylvania), Al Sanders (Charleston Museum, Charleston, South Carolina), Ivy Rutzky and John Maisey (American Museum of Natural History, New York), and Richard Zakrzewski (Sternberg Museum of Natural History, Hays, Kansas) graciously allowed access to their collections through direct visit, loan of material, or providing photographs of specimens. Evgeny Popov (Saratov State University, Russia) was kind enough to comment on photographs of SC.158.151 and he provided photographs of Edaphodon bucklandi and E. leptognathus mandibular plates. LITERATURE CITED Aggasiz, L. 1843. Recherches sur les Poissons Fossiles, vol. 3. Neuchâtel et Soleure, Switzerland, 390 pp. Applegate, S. P. 1970. The vertebrate fauna of the Selma Formation of Alabama. Part 8. The Fishes. Fieldiana Geology Memoirs 3:383-433. Bonaparte, C. 1838. Selachorum tabula analytica. Nuovi Annali della Scienza Naturali (Blogna) 2:195-214. Briedis, D., and J. L. Knight. 1996. The Kingstree Fauna: What does it mean? South Carolina Academy of Sciences. 58:76. Brown, R. W. 1946. Fossil egg capsules of chimaeroid fishes. Journal of Paleontology 20(3):261266. Buckland, W. 1835. A notice on the fossil beaks of four extinct species of fishes, referrible to the genus Chimaera, which occur in the Oolithic and Cretaceous formations of England. Proceedings of the Geological Society of London 2:205-206. Buckland, W. 1838. On the discovery of fossil fishes in the Bagshot Sands at Goldworth Hill, 4 miles north of Guilford. Proceedings of the Geological Society of London 2:687-688. Case, G. R. 1978. Ischyodus bifurcatus, a new species of chimaeroid fish from the upper Cretaceous of New Jersey. Geobios 11:21-29. Case, G. R. 1979. Cretaceous selachians from the Peedee Formation (late Maastrichtian) of

46

PALUDICOLA, VOL. 8, NO. 1, 2010

Duplin County, North Carolina. Brimleyana 2:77-89. Case, G. R. 1996. A new selachian fauna from the lower Hornerstown Formation (early Paleocene/Montian) of Monmouth County, New Jersey. Palaeontographica Abteilung A 242:1-14. Case, G. R., and D. R. Schwimmer. 1992. Occurrence of the chimaeroid Ischyodus bifurcatus Case in the Cusseta Formaiton (Upper Cretaceous, Campanian) of western Georgia and its distribution. Journal of Paleontology 66(2):347350. Cicimurri, D. J. 2007. A late Campanian (Cretaceous) selachian assemblage from a classic locality in Florence County, South Carolina. Southeastern Geology 45(2):59-72. Cicimurri, D. J., D. C. Parris, and M. J. Everhart. 2008. Partial dentition of a chimaeroid fish Chondrichthyes, Holocephali) from the upper Cretaceous Niobrara Chalk of Kansas, USA. Journal of Vertebrate Paleontology 28(1):34-40. Consoli, C. P. 2006. Edaphodon kawai, sp. nov. (Chondrichthyes, Holocephali): A late Cretaceous chimaeroid from the Chatham Islands, Southwest Pacific. Journal of Vertebrate Paleontology 26(4):801-805. Cope, E. D. 1869. Descriptions of some extinct fishes previously unknown. Proceedings of the Boston Society of Natural History 12:310-317. Cope, E. D. 1875. The vertebrata of the Cretaceous formations of the West. Report of the United States Geological Survey of the Territories, 2, 303 pp. Duffin, C. J. 2001. Chimaerid (Holocephali, Chimaeriformes) vomerine toothplate from the upper Cretaceous of Belgium. Palaeontology, 44(6):1179-1188. Duffin, C. J., and J. P. H. Reynders. 1995. A fossil Chimaeroid from the Gronsveld Member (Late Maastrichthian, Late Cretaceous) of northeast Belgium. Belgian Geological Survey, Special Paper 278:111-156. Edwards, L. E., Gohn, G. S., Bybell, L. M., Chirico, P. G., Christopher, R. A., Frederiksen, N. O., Prowell, D. C., Self-Trail, J. M., and Weems, R. E. 2000. Supplement to the preliminary stratigraphic database for subsurface sediments of Dorchester County, South Carolina. U.S. Geological Survey Open-File Report 00- 049-B, 44 pp. Erickson, B .R. 1998. Crocodilians of the Black Mingo Group (Paleocene) of the South Carolina Coastal Plain. Pp. 196-214 in A.E. Sanders (ed.), Paleobiology of the Williamsburg Formation (Black Mingo Group; Paleocene) of South

Carolina, USA. Transactions of the American Philosophical Society 88(4). Erickson, B., A. E. Sanders, and J. L. Knight. In press. Cretaceous reptiles and dinosaurs of South Carolina. Proceedings of the Academy of Natural Sciences, Philadelphia. Fallaw, W. C., and V. Price. 1995. Stratigraphy of the Savannah River Site and vicinity. Southeastern Geology 35(1):21-58. Fowler, H. W. 1911. A description of the fossil fish remains of the Cretaceous, Eocene and Miocene formations of New Jersey. New Jersey Geological Survey, Bulletin 4, 182 pp. Garman, S. 1901. Genera and families of the chimaeroids. Proceedings of the new England Zoological Club 2:75-77. Hoganson, J. W., and J. M. Erickson. 2005. A new species of Ischyodus (Chondrichthyes: Holocephali: Callorhynchidae) from upper Maastrichtian shallow marine facies of the Fox Hills and Hell Creek formations, Williston Basin, North Dakota. Palaeontology 48(4):709721. Hulbert, R. C., Jr. 2001. The Fossil Vertebrates of Florida. University of Florida Press, Gainesville, 384 pp. Hussakof, L. 1912. The Cretaceous chimaeroids of North America. Bulletin of the American Museum of Natural History 31:195-288. Hutchinson, J. H., and R. E. Weems. 1998. Paleocene 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, USA. Transactions of the American Philosophical Society 88(4). Kemp, D., L. Kemp, and D. J Ward. 1990. An Illustrated Guide to the British Middle Eocene Vertebrates. Privately published, London, 59 pp. Knight, J. K., D. J. Cicimurri, and R. W. Purdy. 2007. New western hemisphere occurrences of Schizorhiza Weiler, 1930 and Eotorpedo White, 1934 (Chondrichthyes, Batomorphii). Paludicola 6(2):87-93. Leidy, J. 1856. Notice of the remains of extinct vertebrate animals of New Jersey, collected by Prof. Cook of the State Geological Survey under the direction of Dr. W. Kitchell. Proceedings of the Academy of Natural Sciences of Philadelphia 8:220-221. Nessov, L. A., and A. O. Averianov. 1996. Early Chimaeriformes of Russia, Ukraine, Kazakhstan and Middle Asia II. Description of New Taxa. Vestnik Sankt-Peterburgskogo Universitita, Series 7, Issue 3(21):3-10 [in Russian].

CICIMURRI—CHIMAEROID FROM SOUTH CAROLINA

Obruchev, D. V. 1967. Fossil chimaera egg capsules. International Geology Review 9(4):567–573. Parmley, D., and D. J. Cicimurri. 2005. First record of a chimaeroid fish from the Eocene of the southeastern United States. Journal of Paleontology 79(6):1219-1221. Patterson, C. 1965. The phylogeny of the chimaeroids. Philosophical Transactions of the Royal Society of London (B) 249:101-219. Patterson, C. 1992. Interpretation of the toothplates of chimaeroid fishes. Zoological Journal of the Linnean Society 106:33-61. Popov, E. V. 1999a. New data on the morphology of dental plates of chimaeroid fishes of the genus Ischyodus from the Cretaceous and Paleogene of central Russia and the Volga region. Russian Academy of Sciences, Proceedings of the Zoological Institute 277:67-82 [in Russian with English figure captions]. Popov, E. V. 1999b. On a record of dental plate of the large chimaeroid Edaphodon mantelli Buckland, 1835) from the Lower Santonian of the Saratov Province (Holocephali, Edaphodontidae). Transactions of the Scientific Research Geological Institute, new series 1:137-141 [in Russian with English abstract and figure captions]. Popov. E.V. 2008. Revision of the chimaeroid fishes (Holocephali, Chimaeroidei) from the British Cretaceous. Acta Geologica Polonica, 58(2):243-247. Popov, E. V. and Beznosov, P.A. 2006. Remains of chimaeroid fishes (Holocephali: Chimaeroidei) from the Upper Jurassic deposits of Komi Republic, Russia. Pp. 55-64 in A. Y. Rozanov , A. V. Lopatin, and P. Y. Parkhaev (eds.), Modern Russian paleontology: Classic and newest methods. Russian Academy of Sciences,Paleontological Institute, Moscow [in Russian with English abstract]. Popov, E. V. and Shapovalov, K.M. 2007. New finds of chimaeroid fishes (Holocephali, Chimaeroidei) from the Jurassic of European Russia. Pp. 25-44 in Modern Russian paleontology: Classic and newest methods. Russian Academy of Sciences, Paleontological Institute, Moscow [in Russian with English abstract]. Popov, E. V., and A. A. Yarkov. 2001. A New Giant Species of Edaphodon (Holocephali: Edaphodontidae) from the Beryozovaya Beds (Lower Paleocene) of the Volgograd Volga Region. Paleontological Journal 35(2):183187. Purdy, R. W. 1998. Chondrichthyan fishes from the Paleocene of South Carolina. Pp. 122-146 in A.

47

E. Sanders (ed.), Paleobiology of the Williamsburg Formation (Black Mingo group; Paleocene) of South Carolina, USA. Transactions of the American Philosophical Society 88(4). Robb, A. J., III. 1989. The upper Cretaceous (Campanian, Black Creek Formation) fossil fish fauna of Phoebus Landing, Bladen County, North Carolina. Mosasaur 4:75-92. Sawyer, G. T. 1998. Coprolites from the Black Mingo Group (Paleocene) of South Carolina. Pp. 221228 in A. E. Sanders (ed.), Paleobiology of the Williamsburg Formation (Black Mingo Group; Paleocene) of South Carolina, USA. Transactions of the American Philosophical Society 88(4). Shin, J.-Y. 2010. A new species of Edaphodon (Chondrichthyes: Holocephali) from the upper Cretaceous Haslam Formation, Vancouver Island, British Columbia, Canada. Journal of Vertebrate Paleontology 30(4):1012-1018. Stahl, B. J. 1999. Chondrichthyes III: Holocephali. In H.-P. Schultze (ed.), Handbook of Paleoichthyology, Volume 4. Friedrich Pfeil, Munich, 164 pp. Stahl, B. J., and S. Chatterjee. 2002. A Late Cretaceous callorhynchid (Chondrichthyes, Holocephali) from Seymour Island, Antarctica. Journal of Vertebrate Paleontology 22(4):848-850. Stahl, B. J., and D. C. Parris. 2004. The complete dentition of Edaphodon mirificus (Chondrichthyes: Holocephali) from a single individual. Journal of Paleontology 78(2):388392. Takeuchi, G. T., and R. W. Huddleston. 2006. A Miocene chimaeroid fin spine from Kern County, California. Bulletin of the Southern California Academy of Science 105(2):85-90. Thibault, N., and S. Gardin. 2006. Maastrichtian calcareous nannofossil biostratigraphy and paleoecology in the Equatorial Atlantic (Demerara Rise, ODP Leg 207 Hole 1258A). Revue de Micropaléontologie 49:199-214. Ward, D. J. 1973. The English Paleogene chimaeroid fishes. Proceedings of the Geological Association 84:315-330. Ward, D. J., and L. Grande. 1991. Chimaeroid fish remains from Seymour Island, Antarctic Peninsula. Antarctic Science 3(3):323-330. Waga, D. D., A. S. Andreeva-Grigorovich, and N.V. Maslun. In press. Calcareous nannofossil biostratigraphy of the Paleocene sediments of the Odessa Gas Field (NW Black Sea). Geobios. Weems, R. E., and L. M. Bybell. 1998. Geology of the Black Mingo Group (Paleocene) in the Kingstree and St. Stephen areas of South

48

PALUDICOLA, VOL. 8, NO. 1, 2010

Carolina. Pp. 9-27 in A. E. Sanders (ed.), Paleobiology of the Williamsburg Formation (Black Mingo Group; Paleocene) of South Carolina, USA. Transactions of the American Philosophical Society 88(4).

Woodward, A. S. 1891. Catalogue of the fossil fishes in the British Museum (Natural History), Part 2. British Museum of Natural History, London, 567 pp.

Paludicola 8(1):49-73 September 2010 © by the Rochester Institute of Vertebrate Paleontology

NEW RECORDS OF RODENTIA FROM THE DUCHESNEAN (MIDDLE EOCENE) SIMI VALLEY LANDFILL LOCAL FAUNA, SESPE FORMATION, CALIFORNIA

Thomas S. Kelly Research Associate, Vertebrate Paleontology Section, Natural History Museum of Los Angeles County, 900 Exposition Blvd, Los Angeles, California 90007

ABSTRACT A large number of isolated rodent teeth have been recently recovered from the Duchesnean (middle Eocene) Simi Valley Landfill Local Fauna of the Sespe Formation during a paleontologic mitigation program at the Simi Valley Landfill and Recycling Center, Ventura County, California. Included in these teeth are new samples of Metanoiamys korthi, Paradjidaumo reynoldsi, Simiacritomys whistleri, and Simimys landeri. The discovery of upper and putative lower premolars of Simiacritomys whistleri supports it’s referral to Eomyidae. New occurrences for the Simi Valley Landfill Local Fauna include Eomyidae (one or more species of uncertain affinities) and Pareumys sp.

Within the new specimens from bed 30A are numerous additional rodent teeth, including those of species that were previously known from small sample sizes (Metanoiamys korthi Kelly and Whistler, 1998, Paradjidaumo reynoldsi Kelly, 1992, Simiacritomys whistleri Kelly, 1992, and Simimys landeri Kelly, 1992). Also, included within these specimens are the first records of Pareumys sp. and at least one or more undetermined eomyid species from bed 30A. The purpose of this paper is to document the new rodent specimens from the Simi Valley Landfill Local Fauna.

INTRODUCTION The middle member of the Sespe Formation, which is exposed along the north side of Simi Valley, Ventura County, California, has previously yielded numerous middle Eocene (Uintan and Duchesnean) fossil mammals (e.g., Golz, 1976; Golz and Lillegraven, 1977; Mason, 1988; Kelly, 1990, 1992, 2009, 2010; Kelly et al., 1991; Kelly and Whistler, 1994, 1998). Kelly (1990, 1992) and Kelly et al. (1991) recognized five superposed local faunas from the middle member, the youngest of which is the Duchesnean Simi Valley Landfill Local Fauna from Natural History Museum of Los Angeles County (LACM) locality 5876 within bed 30A of the locally exposed Sespe Formation. Based on paleomagnetic studies, Prothero et al. (1996) placed bed 30A within Chron 17r of the geomagnetic polarity time scale, or about 38.0-37.8 million years before present (Luterbacher et al., 2004). A large number of isolated small mammal teeth have recently been recovered from bed 30A by wet screen sieving of bulk matrix during a paleontologic mitigation program at the Simi Valley Landfill and Recycling Center (Lander, 2008; Kelly, 2009, 2010). Kelly (2009, 2010) previously documented two new species in these specimens, the rodent Heliscomys walshi and the lipotyphlan Batodonoides walshi, along with new records of Marsupialia, additional lipotyphlans, and Primates.

METHODS Measurements of teeth were made with an optical micrometer to 0.001 mm and then rounded off to the nearest 0.01 mm. Teeth that had the enamel partially abraded away were not included in the measurements. Because of minor inaccuracies in the calibration of the optical micrometer used by Kelly (1992) and Kelly and Whistler (1998), which resulted in their dental measurements being slightly larger than the actual measurements (by ~ 2%), all measurements of rodent specimens from the Simi Valley Landfill Local Fauna were recalculated for this paper and are included in the dental statistics for each species. Certain rodent M3 specimens (LACM 153836, 153838, 153843, 153844, 153854, and 153855) in the new sample from the Simi Valley Landfill Local Fauna could not be identified because they are so worn that their occlusal patterns are undeterminable. Cheek teeth cusp/crest terminology 49

 

50

PALUDICOLA, VOL. 8, NO. 1, 2010

  follows Wood and Wilson (1936) with additional minor crest terminology for Metanoiamys following Chiment and Korth (1996) and for Simimys following Lillegraven and Wilson, 1975. Dental formula follows standard usage. Upper and lower teeth are designated by uppercase and lowercase letters, respectively. All specimens were recovered by wet screen sieving of bulk matrix from bed 30A of the middle member of the Sespe Formation (locality LACM 5876) at the Simi Valley Landfill and Recycling Center during a mitigation program directed by Paleo Environmental Associates, Inc., for Waste Management of California, Inc. Walsh (2008) provided preliminary identifications and a cursory summary of the new sample of small mammal teeth recovered during the last phase of the mitigation program for the final (non-peer reviewed) proprietary report prepared by Paleo Environmental Associates, Inc. However, during the detailed study reported on herein, it became clear that a number of Walsh’s identifications were incorrect. All specimens are deposited in the Vertebrate Paleontology Section of the Natural History Museum of Los Angeles County. Detailed locality data are available at this institution and also see Lander (2008). Abbreviations and acronyms are as follows: ap, greatest anteroposterior length; CV, coefficient of variation; L, left; N, number of specimens; OR, observed range; R, right; SD, standard deviation; tr, greatest transverse width; tra, anterior transverse width; trp, posterior transverse width. SYSTEMATIC PALEONTOLOGY Order Rodentia Bowdich, 1821 Family Eomyidae Winge, 1887 Genus Metanoiamys Chiment and Korth, 1996 Metanoiamys korthi Kelly and Whistler, 1998 Figures 1,2 Referred Specimens―dP4, LACM 153865, 153877, 153878; P4, LACM 153769, 153771, 153781, 153783, 153784, 153785, 153874, 153879, 153880, 153883, 153896; M1, LACM 153779, 153884, 153891, 153893, 153894, 153895; M2, 153885, 153892, 153913, 153914, 153915, 153916, 153917; M3, LACM 153929, 153930, 153931, 153932, 153933, 153934, 153935, 153936, 153937, 153938, 153939; dp4, LACM 153866, 154867, 153868, 153869, 153870, 153871, 153872, 153873; p4, LACM 153780, 153789, 153792, 153864; 153875, 153876, 153881, 153882; m1 or m2, LACM 153886, 153887, 153888, 153889, 153897, 153898, 153899, 153901, 153902, 153903, 153904, 153905, 153906, 153907, 153908, 153909, 153910, 153911, 153912, 153918, 153919, 153920, 153921, 153922, 153923, 153924, 153925, 153926, 153927, 153928; m3, LACM 153940, 153941, 153942,

153943, 153944, 153945, 153946, 153948, 153949, 153950, 153951, 153952, 153953. Description―Kelly and Whistler (1998) provided detailed descriptions of P4-M2 and m1-2, so these will not be repeated here. However, the new sample does provide additional information on the individual variation of these teeth, which is included below. The dP4, M3, dp4, p4, and m3 were previously unknown for Metanoiamys korthi, so detailed descriptions of these teeth are also included below. Kelly and Whistler (1998) previously assigned LACM 131059 to m1 or m2, but it actually represents M1. Three teeth are identified as dP4s. They are very similar in occlusal morphology to P4, but are distinctly smaller (Figure 1A, Table 1). The occlusal outline is almost square. A well-developed anterior cingulum is present that extends lingually from the anterior side of the paracone to the anterocone, where it has a slight indentation, and then continues to the anterolingual corner of the tooth. A small, distinct anterocone and adlophule (crest extending posteriorly from the anterior cingulum at the position of the anterocone and connecting to the protoloph or protocone) are present in all specimens. The primary cusps are well-developed with the paracone and metacone taller than the protocone and hypocone. The protoloph and metaloph are moderately high, complete crests that connect the paracone with the protocone and the metacone with the hypocone, respectively. Mesostyles are lacking in all specimens. A mesocone is lacking in one specimen and represented as a slight swelling or small cuspule near the center of the endoloph in the other two specimens. The endoloph is complete, connecting the protocone and hypocone. A mesoloph is lacking in two specimens, but on the third specimen the mesoloph extends labially from the mesocone as a simple spur of moderate length. The posterior cingulum is a distinct crest that extends labially from the posterolabial side of the hypocone to the posterior base of the metacone. The new 11 specimens of P4 exhibit very similar occlusal patterns that agree well morphologically with those described by Kelly and Whistler (1998) (Figure 1B-D). The new 13 specimens of M1 and M2 agree well morphologically with the sample described by Kelly and Whistler (1998). All have a well-developed adlophule, a complete, moderately high endoloph, a distinct anterocone present on the anterior cingulum, and an anterior cingulum that extends lingually from the paracone to the anterocone and then continues to the anterolingual margin of the tooth (Figure 1E-I). Mesolophs are completely lacking in seven specimens, present as very short, simple spurs in five specimens, and in one specimen the mesoloph is moderately short with a slight bifurcation (additional spur) at its labial end. Mesocones are present on all M1-2s that vary

KELLY—DUCHESNEAN RODENTS FROM CALIFORNIA

51

 

FIGURE 1. Upper teeth of Metanoiamys korthi. A, LdP4, LACM 153865; B, RP4, LACM 153878; C, RP4, LACM 153896; D, LP4, LACM 153874; E, LM1, LACM 153893; F, LM1, LACM 153894; G, LM2, LACM 153915; H, RM2, LACM 153916; I, RM2, LACM 153917; J, RM3, LACM 153937; K, RM3, LACM 153936; L, LM3, LACM 153929. All occlusal views, lingual at bottom, scale = 1 mm.

52

PALUDICOLA, VOL. 8, NO. 1, 2010

 

FIGURE 2. Lower teeth of Metanoiamys korthi. A, Rdp4, LACM 153872; B, Rdp4, LACM 153871; C, Ldp4, LACM 153868; D, Rp4, LACM 153881; E, Rp4, 153882; F, Lm1 or m2, LACM 153898; G, Lm1 or m2, LACM 153899; H, Rm1 or m2, LACM 153907; I, Lm1 or m2, LACM 153920; J, Lm1 or m2, LACM 153921; K, Rm3, LACM 153952; L, Rm3, LACM 153946. All occlusal views, lingual at top, scale = 1 mm.

KELLY—DUCHESNEAN RODENTS FROM CALIFORNIA

53

  from a small distinct cusp (seven specimens) to a small, swelling or bulge on the endoloph (six specimens). Mesostyles are lacking in six specimens and represented as a small bump or swelling along the labial margin of the tooth between the paracone and metacone on seven specimens. In six specimens, a short, simple, crest or spur (= posterior arm of the protocone of Kelly and Whistler, 1998) is present that extends from the anterior side of the endoloph, near its connection with the protocone, towards the mesoloph (two specimens), towards the protocone (two specimens), or directly straight labially (two specimens). This crest is lacking in seven specimens. The M1 is differentiated from the M2 by having the tra and trp more nearly equal in width (less transversely expanded anteriorly), whereas the M2 has the tra relatively wider than the trp (more transversely expanded anteriorly). Eleven teeth are identified as M3 in the new sample. The occlusal outline is subcircular (Figure 1JL). The paracone is the largest and tallest primary cusp. The protocone is slightly smaller than paracone and positioned medially along the lingual margin of the tooth. The entoconid is a distinct, transversely compressed cusp that is lower in height than the paracone. The hypocone varies from a relatively distinct, anteroposteriorly compressed cusp to a moderate swelling along the posterolingual edge of the tooth at about the level of the labial border of the protocone. In two specimens, the protocone and hypocone are almost joined, whereas in all the other teeth they are separated by a distinct notch. The anterior cingulum is a well-developed, moderately high crest extending from the anterolabial side of the paracone to the anterolingual corner of the tooth. A small anterocone is present on the anterior cingulum in all specimens. A well-developed adlophule is present in all specimens and exhibits a strong connection with the protoloph in all but two specimens, where the adlophule is slightly constricted at the junction with the protoloph. A well-developed protoloph, connecting the protocone and paracone is present in all specimens. A very low crest (= incipient metaloph) connecting the hypocone and metacone is present in four specimens, whereas it is lacking on all the other teeth. The posterior cingulum, which is only slightly lower in height than the metacone and hypocone, extends in an arc from the labial side of the hypocone to the posterior labial side of the metacone. The small crests and cuspules within the central valley exhibit the most variability. In six specimens, a small, short, low crest is present that extends anterolabially from either the posterolabial corner of the protocone (three specimens) or the anterolingual corner of the hypocone (three specimens) into the central valley and then turns either

slightly more anteriorly or slightly more labially. Of the six specimens exhibiting this morphology, one has a minute mesocone present on the crest. The crest is completely absent in one tooth, but a small, though distinct, cuspule is present in the central valley that appears to represent a mesocone. In the other four teeth, the crest is only represented by a minute, short spur with a small mesocone occurring at the labial end of the spur (3 specimens) or isolated from the spur (one specimen). The M3 of Metanoiamys korthi is similar in size to those of Paradjidaumo reynoldsi. However, it can be easily distinguished from the M3 of P. reynoldsi by having the following: 1) an adlophule is present; 2) an anterocone is present; and 3) the anterior cingulum extends to the lingual margin of tooth. Lacking are a long, well-developed mesoloph and a distinct hypoloph, which are present in P. reynoldsi and result in a five crested occlusal pattern. Walsh (1997) clarified the characteristics that allow dp4 and p4 of Metanoiamys to be differentiated from each other, wherein dp4 is smaller, relatively longer anteroposteriorly (more elongated occlusal outline) with the anterior transverse width relatively narrower, and the anteroconid usually relatively larger. Eight teeth are identified as dp4, which are smaller than p4 (Figure 2A-C). The anteroconid is a moderately well-developed cusp that is positioned anteriorly and medially relative to the metaconid and protoconid. It is distinctly lower than the metaconid and protoconid. No anterior cingulids are present on any of the dp4s. The metaconid and protoconid are distinct cusps that are separated by a shallow valley and, with wear, this valley disappears to allow formation of a short metalophid connection between the cusps. The entoconid and hypoconid are welldeveloped cusps with the entoconid higher than the hypoconid. The hypolophid is a distinct crest. A complete, but low, ectolophid extends from the hypoconid to the protoconid on all of the dp4s. A distinct, small mesoconid is present on three dp4s, whereas on the others it varies from incipient to weakly expressed, that is from a slight to moderate swelling on the ectolophid. On two teeth a very short mesolophid extends lingually from the mesoconid into the central valley, and on one tooth a low, thin mesolophid extends about three-quarters of the way across the central valley, whereas on the other dp4s a mesolophid is lacking. A mesostylid is lacking on all dp4s. The posterior cingulid is moderately well-developed, extending from the posterolingual side of the hypoconid to the posterior medial base of the entoconid. All dp4s exhibit distinctly narrower anterior transverse widths relative to the posterior transverse widths (Table 2).

54

PALUDICOLA, VOL. 8, NO. 1, 2010

  TABLE 1. Dental statistics (in mm) of all upper teeth of Metanoiamys korthi from Simi Valley Landfill Local Fauna. Includes new sample provided herein plus specimens described by Kelly (1992) and Kelly and Whistler (1998). Tooth/ Dimension N Mean OR SD CV dP4 ap 3 0.84 0.81-0.90 0.05 ― dP4 tra 3 0.80 0.76-0.85 0.05 ― dP4 trp 4 0.83 0.81-0.84 0.02 3.0 P4 ap 14 1.00 0.95-1.08 0.05 5.0 P4 tra 14 1.01 0.95-1.06 0.04 4.0 P4 trp 14 1.00 0.96-1.06 0.03 3.0 M1 ap 8 1.06 1.03-1.13 0.04 3.8 M1 tra 8 1.13 1.05-1.19 0.06 5.3 M1 trp 8 1.10 1.03-1.16 0.05 4.6 M2 ap 11 1.04 0.95-1.09 0.04 3.9 M2 tra 10 1.15 1.10-1.19 0.04 3.5 M2 trp 9 1.08 0.98-1.12 0.04 3.7 M3 ap 11 0.85 0.80-0.95 0.04 4.7 M3 tra 10 0.94 0.86-0.95 0.04 4.3 M3 trp 10 0.76 0.70-0.84 0.05 6.6 _______________________________________________________ TABLE 2. Dental statistics (in mm) of all lower teeth of Metanoiamys korthi from Simi Valley Landfill Local Fauna. Includes new sample provided herein plus specimens described by Kelly (1992) and Kelly and Whistler (1998). Tooth/ Dimension N Mean OR SD CV dp4 ap 7 0.86 0.82-0.90 0.03 3.5 dp4 tra 7 0.50 0.47-0.53 0.03 6.0 dp4 trp 8 0.71 0.67-0.75 0.03 4.2 p4 ap 8 1.01 0.95-1.08 0.04 4.0 p4 tra 8 0.67 0.62-0.72 0.03 4.5 p4 trp 7 0.85 0.81-0.90 0.04 4.7 m1 or m2 ap 33 1.10 1.00-1.18 0.05 4.5 m1 or m2 tra 30 1.01 0.90-1.09 0.05 5.0 m1 or m2 trp 31 1.06 0.98-1.13 0.04 3.8 m3 ap 13 1.06 1.03-1.10 0.02 1.9 m3 tra 10 0.97 0.90-1.03 0.03 3.1 m3 trp 11 0.85 0.80-0.93 0.04 4.7 ______________________________________________________

The eight teeth identified as p4 differ from dp4 by having a relatively wider trigonid and less anteroposteriorly elongated occlusal outline. The primary cusps (metaconid, protoconid, entoconid, and hypoconid) are well-developed cusps with the entoconid being the largest and highest (Figure 2D-E). A small, distinct anteroconid is present on seven p4s and lacking on one. However on one p4, the anteroconid is very small and connected to the base of the protoconid by a low, short cristid. A similar cristid is present on the p4 that lacks a distinct anteroconid. The metalophid extends from the protoconid to the metaconid and is better developed than that of dp4. The ectolophid is a well-developed, complete cristid that connects the protoconid to the hypoconid and is relatively higher than that of the dp4. A mesoconid, which varies from a distinct triangular shaped cusp to a moderate swelling on the ectolophid, is present on seven of the p4s. On one p4, the mesoconid is incipient; it is only a very slight swelling on the

ectolophid. A mesostylid is present between the metaconid and entoconid on three p4s, which varies from a small distinct cusp on one to a small bulge on the other two. The other five p4s lack a mesostylid. A mesolophid is present on five p4s, varying from a very short lingually directed spur from the mesoconid (four p4s) to a very low cristid (one p4) extending lingually about half way across the central valley. On the other three p4s, a mesolophid is lacking. The hypolophid is well-developed on all p4s, connecting the entoconid to the hypoconid. The posterior cingulid is a distinct cristid, slightly lower than the hypolophid that extends lingually from the posterolabial side of the hypoconid to the posterior side of the entoconid. Confident differentiation of Metanoiamys first and second lower molars from samples consisting only of isolated teeth is not possible. The new sample includes 30 m1 or m2s, which agree well morphologically with those described by Kelly and Whistler (1998). A distinct anteroconid, complete ectolophid, and well-developed adlophulid (crest extending posteriorly from the anterior cingulid at the position of the anterocone and connecting with the metalophid or protoconid) are present on all the m1-2s (Figure 2F-J). Mesostylids are usually lacking (26 specimens), but in three specimens the mesostylid is represented by a very small bump or swelling (incipient) on the lingual margin of the tooth between the metaconid and entoconid and in one specimen it is represented by a small, distinct cuspulid. Mesoconids are present on all specimens except one, where it is only a very slight swelling on the ectolophid. A mesolophid is present as a simple cristid that varies from very short to moderate in length and extends from the mesoconid into the central valley in 24 specimens. In two specimens, the mesolophid exhibits a slight bifurcation or spur at its lingual end and in four specimens, a mesolophid is lacking. Thirteen m3s are identified in the new sample. The occlusal outline is subrectangular with the anterior transverse width greater than the posterior transverse width (Figure 2K-L). The anterior cingulid is robust and extends labially from the anterior side of the metaconid to the anterolabial margin of the tooth. A well-developed adlophulid is present on all but one specimen, where the adlophulid is incomplete, not reaching the metalophid. The anteroconid is present as a small cuspulid along the center of the anterior cingulid in 11 specimens, but is represented by only a slight swelling on the anterior cingulid in two specimens. The metaconid is the largest and tallest primary cusp. The entoconid is a well-developed cusp positioned at the posterior lingual margin of the tooth and is taller than the hypoconid. The metalophid is a complete cristid connecting the metaconid and protoconid in all specimens. The hypolophid is a

KELLY—DUCHESNEAN RODENTS FROM CALIFORNIA

55

  complete crest that extends as a posteriorly directed arc from the hypoconid to the entoconid. In unworn teeth, the hypolophid has a slight notch along its lingual half between the entoconid and hypoconid. A mesolophid is present in all specimens, where it exhibits the following variation: 1) a short cristid, extending lingually from the mesoconid straight into the central valley (five specimens); 2) a short cristid, extending into the central valley, but with a slight posterior turn at its lingual end (four specimens); 3) a moderate to moderately long cristid, extending lingually straight into the central valley (three specimens); and 4) a moderately long cristid that extends into the central valley and then turns posteriorly at its lingual end (one specimen). On three specimens, one or two minute posteriorly directed spurs are present on the mesolophids. On all specimens, a mesoconid is present as a small cuspulid or slight swelling along the ectolophid. The ectolophid extends between the hypoconid and protoconid. It is well connected to the hypoconid and almost as tall as the hypoconid along its posterior edge. From the mesoconid, the ectolophid decreases in height anteriorly, where there is a small notch present in unworn teeth at the point where it meets the posterior side of the protoconid. With wear this notch disappears, giving the appearance that the ectolophid is well connected to the protoconid. A mesostylid is lacking in five specimens, incipient in six specimens (represented as a very slight swelling or irregularity in the enamel), and a distinct, small cuspulid in two specimens. On one specimen, the mesostylid has a short labially extending spur. A small posterior cingulid is present at the posterolabial corner of the tooth on two specimens, where it extends a short distance from the posterolingual wall to the posterolabial wall of the entoconid. Dental statistics for the entire sample of Metanoiamys korthi from the Simi Valley Landfill Local Fauna, including those specimens described by Kelly (1992) and Kelly and Whistler (1998), are provided in Tables 2-3. Measurements of the holotype m1 or m2 (LACM 132447) of M. korthi are ap = 1.16 mm, tra = 1.09 mm, and trp = 1.13 mm. Discussion―The newly recovered teeth of Metanoiamys korthi significantly increase the sample size for the species and provide additional information on individual variation. Also, the new teeth include the first records of dP4, M3, dp4, p4, and m3 of M. korthi. Genus Paradjidaumo Burke, 1934 Paradjidaumo reynoldsi Kelly, 1992 Figures 3,4 Referred Specimens―P4, LACM 153770, 153772, 153773,153774, 153775, 153776, 153777, 153782, 153786, 153787; M1 or M2, LACM 153793,

153794, 153795, 153796, 153798, 153799, 153800, 153801, 153802, 153803, 153804, 153805, 153806, 153807, 153808, 153810, 153817, 153818, 153819, 153820, 153821, 153822, 153827, 153828, 153829, 153830, 153831, 158333; M3, LACM 153835, 153837, 153839, 153840, 153841, 153842, 153846, 153847, 153848, 153849, 153851, 153852, 153853; Ldp4, LACM 153778; p4, LACM 153788, 153790, 153791; m1 or m2, LACM 153811, 153812, 153813, 153814, 153815, 153824, 153825, 153826, 153832, 153834, 153900; m3, LACM 153856, 153857, 153858, 153859, 153860, 153861, 153862, 153863, 153890, 153947. Description―Without intact dentitions of Paradjidaumo reynoldsi for comparison, differentiation of M1 from M2 and m1 from m2 cannot be made confidently from isolated teeth. First and second lower molars are differentiated from first and second upper molars by having anteroposterior lengths that are greater than their transverse widths, whereas in the upper molars, the transverse widths are greater than the anteroposterior widths. First and second lower molars further differ from first and second upper molars by having an elongated anterior cingulid that extends to the labial margin of the tooth connecting with the anterior base of the protoconid, whereas in first and second upper molars the lingual extension of the anterior cingulum connects with the protoloph and does not extend to the lingual margin of the tooth. The lingual/labial connections of the posterior cingulum/cingulid of the first and second upper and lower molars, respectively, differ similarly, but in the mirror image of the anterior cingulum/cingulid connections. Kelly (1992) tentatively referred three teeth (LACM 130841, 131028, 131050) to m3, but with the discovery of definitive m3s in the new sample it is clear that these teeth actually represent m1 or m2. Kelly (1992) provided detailed descriptions of P4, M13, p4, and m1-2, so these will not be repeated here. However, the new sample does provide additional information on the individual variation of these teeth, which is included below. The new sample of Paradjidaumo reynoldsi also includes the first known examples of dp4 and m3, so detailed descriptions of these teeth are also included below. Ten teeth are identified as P4 in the new sample. Their occlusal morphology agrees with those described by Kelly (1992). They all exhibit a short, anterior cingulum that extends from the posterolabial portion of the metaloph to the anterolabial corner of the paracone, a small, distinct mesostyle between the paracone and metacone, a small mesocone at about the center of the endoloph, and lack an anterocone (Figure 3A-C). All but one P4 (LACM 153774) have complete, high endolophs connecting the hypocone to the protocone. On LACM 153774, the endoloph is a high crest (about

56

PALUDICOLA, VOL. 8, NO. 1, 2010

 

FIGURE 3. Upper teeth of Paradjidaumo reynoldsi. A, LP4, LACM 153774; B, LP4, LACM 153782; C, RP4, LACM 153772; D, RM1 or M2, LACM 153808; E, LM1 or M2, LACM 153800; F, RM1 or M2, LACM 153804; G, RM1 or M2, LACM 153820; H, LM1 or M2, LACM 153794; I, LM1 or M2, LACM 153833; J, LM1 or M2, LACM 153801; K, LM1 or M2, LACM 153796; L, LM1 or M2, 153834; M, LM3, LACM 153835; N, RM3, LACM 153848; O, RM3, LACM 153849. All occlusal views, lingual at bottom, scale = 1 mm.

KELLY—DUCHESNEAN RODENTS FROM CALIFORNIA

57

 

FIGURE 4. Lower teeth of Paradjidaumo reynoldsi. A, Ldp4, LACM 153778; B, Lp4, LACM 153788; C, Rp4, LACM 153791; D, Lm1 or m2, LACM 153824; E, Lm1 or m2, LACM 153813; F, Lm1 or m2, LACM 153812; G, Lm1 or m2, LACM 153832; H, Lm3, LACM 153858; I, Lm3, LACM 153857. All occlusal views, lingual at top, scale = 1 mm. _____________________________________________________________________________________________________________

the height of the hypocone) that extends from the hypocone to the posterolabial corner of the protocone, where it is separated from the protocone by a transversely narrow and deep notch. LACM 153775 is unworn and it is apparent that the notch separating the endoloph from the protocone would disappear in late wear resulting in the appearance of a complete endoloph. A mesoloph is present in all specimens,

extending relatively straight labially from the mesocone into the central valley, but exhibits variation in its length. The mesoloph extends across the full length of the central valley to the mesostyle in four specimens, to about three-quarters the way across the central valley in three specimens, and to about one-half the way across the central valley in three specimens. The P4 of Paradjidaumo reynoldsi is similar in size to

58

PALUDICOLA, VOL. 8, NO. 1, 2010

  that of Metanoiamys korthi, but can easily be distinguished from those of M. korthi by lacking an adlophule, a distinct anterocone, a lingual extension of the anterior cingulum, and an indentation at the middle of the anterior cingulum, along with having a much better developed mesoloph (higher and longer). Except for the notable indentation of the anterior cingulum, these characters can also be used to distinguish the M12 of P. reynoldsi from those of M. korthi. ___________________________________________ TABLE 3. Dental statistics (in mm) for all upper teeth of Paradjidaumo reynoldsi from Simi Valley Landfill Local Fauna. Includes new sample provided herein plus specimens described by Kelly (1992). Tooth/ Dimension N Mean OR SD CV P4 ap 14 1.08 1.00-1.18 0.06 5.6 P4 tra 14 1.06 0.95-1.16 0.07 6.6 P4 trp 15 1.05 0.97-1.18 0.07 6.6 M1 or M2 ap 39 1.13 1.00-1.30 0.07 6.2 M1 or M2 tra 37 1.29 1.10-1.39 0.07 5.4 M1 or M2 trp 35 1.22 1.08-1.38 0.07 5.7 M3 ap 18 0.97 0.85-1.05 0.06 6.2 M3 tra 18 1.01 0.93-1.09 0.04 4.0 M3 trp 18 0.84 0.77-0.92 0.05 6.0 _______________________________________________________ TABLE 4. Dental statistics (in mm) for all lower teeth of Paradjidaumo reynoldsi from Simi Valley Landfill Local Fauna. Includes new sample provided herein plus specimens described by Kelly (1992). Tooth/ Dimension N Mean OR SD CV dp4 ap 1 1.19 ― ― ― dp4 tra 1 0.67 ― ― ― dp4 trp 1 0.75 0.70-0.80 ― ― p4 ap 8 1.15 1.05-1.29 0.08 7.0 p4 tra 7 0.83 0.75-0.90 0.05 6.0 p4 trp 7 1.11 1.00-1.23 0.09 8.1 m1 or m2 ap 28 1.21 1.07-1.36 0.08 6.6 m1 or m2 tra 24 1.14 1.03-1.26 0.09 7.9 m1or m2 trp 27 1.18 1.06-1.33 0.07 5.9 m3 ap 10 1.00 0.92-1.08 0.05 5.0 m3 tra 9 0.99 0.95-1.03 0.03 3.0 m3 trp 9 0.85 0.77-0.93 0.05 5.9 _______________________________________________________

The new sample contains 28 teeth identified as M1 or M2 (Figure 3D-L). The occlusal morphology of the first and second molars in the new sample agrees well with those described by Kelly (1992). The mesoloph extends labially across the central valley from the endoloph or incipient mesoconid, when present, to the labial margin of the tooth or mesostyle, when present, on 19 specimens. On six specimens, the mesoloph extends labially three-quarters or slightly more across the central valley and on three specimens, it extends a little over half way across the central valley. A small mesostyle is present between the paracone and metacone on 19 specimens, whereas on four specimens the mesostyle is represented by only a

slight bump or swelling and five specimens lack a mesostyle. A mesoconid is lacking on three specimens, whereas on the other 25 specimens, it is incipient, that is a minute cuspule (unworn teeth) or very slight swelling (worn teeth) on the endoloph. The new sample contains 13 teeth identified as M3 (Figure 3M-O), which exhibit more individual variation than the original sample of five teeth described by Kelly (1992). The morphology of the anterior cingulum agrees well with those that Kelly (1992) described in all but one specimen, where the anterior cingulum is lacking and the enamel along the anterior edge, where the cingulum would normally be, exhibits numerous small bumps. In this specimen, the posterior cingulum is also lacking, exhibiting also numerous small bumps along the posterior edge of the tooth where it would normally occur. This morphology is probably due to aberrant ontogeny during the development of this tooth. In eight specimens the mesoloph extends labially completely across the central valley to the labial edge of the tooth and in four specimens it extends only about three-quarters of the way across the central valley, giving all of these teeth a five crested occlusal pattern. On one specimen (LACM 153841), a mesoloph is lacking, yet the tooth is otherwise indistinguishable morphologically from the other 12 third molars. In the new sample, the protocone and hypocone are connected in seven specimens (Figure 3M) and separated by a distinct notch in six specimens (Figure 3N-O). A small mesostyle is present in two specimens, whereas in eight specimens it is represented by a small swelling on the enamel between the protocone and metacone and in three specimens it is lacking. The posterior cingulum also exhibits variation in the new sample; in eight specimens the posterior cingulum is well separated from the hypoloph by a relatively wide transverse valley and in five specimens the posterior cingulum and hypoloph are more closely positioned to each other, only separated by a narrow transverse valley. One tooth is identified as dp4 (Figure 4A). It is relatively narrower transversely and more elongated anteroposteriorly than the p4 of P. reynoldsi. The metaconid and protoconid are transversely compressed cusps of about equal height and are connected anteriorly by a very short, narrow cristid (? = metalophulid I). The entoconid and hypoconid are well-developed cusps with the entoconid taller than metaconid, protoconid, and hypoconid. A welldeveloped hypolophid connects the hypoconid and entoconid. A complete, high ectolophid connects the protoconid and hypoconid. The mesolophid is a tall cristid that extends lingually from the ectolophid to about half-way across the central valley. A cristid extends posteriorly from the metaconid towards the entoconid, although it is separated from the entoconid

KELLY—DUCHESNEAN RODENTS FROM CALIFORNIA

59

  by a wide notch. The posterior cingulid extends lingually from the posterolingual side of the hypoconid to the posterior side of the entoconid. An anteroconid, mesostylid, and mesoconid are lacking. The reference of LACM 153778 to dp4 of P. reynoldsi could be questioned because it differs from p4 by having a mesoloph that is not as complete and the posterior cingulid connects with the posterior lingual side of the hypoconid instead of near the middle of the hypolophid. However, it is similar to the p4 in having a complete, high ectolophid, high mesolophid, welldeveloped hypolophid, and lacking an anteroconid. These characters, along with its larger size, also distinguish LACM 153778 from the dp4 of Metanoiamys korthi. Furthermore, LACM 153778 appears to be too small to be referable to any of the larger, undetermined genera of Eomyidae described below. Moreover, other investigators have noted considerable variation in the mesoloph/mesolophid length and in the posterior cingulum/cingulid morphology of dp4, P4/p4, and M3/m3 of Paradjidaumo (e.g., Black, 1965; Korth, 1980; Storer, 1984). For these reasons, LACM 153778 is tentatively assigned to P. reynoldsi. Three teeth are identified as p4. The anterior transverse widths are narrower than the posterior transverse widths (Figure 4B-C). The occlusal morphology of the new sample of p4s agrees well with those described by Kelly (1992). They all exhibit the following characters: 1) an anteroconid is lacking; 2) a short, low anterior cingulid is present that extends from the anterolabial side of the protoconid to the anterolingual side of the metaconid; 3) a complete short metalophid that is slightly taller than the anterior cingulid and connects the posterolabial side of protoconid to the posterolingual side of the metaconid; 4) a complete, high ectolophid that connects the protoconid to the hypoconid; 5) a complete mesolophid that extends lingually across the central valley from the ectolophid to a small metastylid between the metaconid and entoconid; 6) a well-developed, high hypolophid that connects the hypoconid and entoconid; and 7) a well-developed, high posterior cingulid that extends from about the middle of the hypolophid to the posterolingual side of the entoconid. Two specimens are lacking a mesoconid and one specimen exhibits an incipient mesoconid as a very slight expansion of the enamel at the junction of the mesolophid with the ectolophid. Eleven teeth in the new sample are identified as m1 or m2 (Figure 4D-G). The occlusal morphology of the first and second lower molars in the new sample agrees well with those described by Kelly (1992). In the new sample, the mesolophid extends completely across the central valley from the ectolophid to the lingual margin of the tooth or the mesostylid, when

present, on seven specimens; it extends three-quarters or a little more across the central valley on four specimens. A small, distinct mesostylid is present between the metaconid and entoconid on two specimens and is lacking on nine specimens. An incipient mesoconid, which is expressed as a slight swelling or widening at the ectolophid-mesolophid junction, is present on six specimens. A mesoconid is lacking on the other five specimens. In the new sample, 10 teeth are confidently assigned to m3. The occlusal outlines are subrectangular with the anterior transverse widths notably wider than the posterior transverse widths (Figure 4H-I, Table 4). The anterior cingulid is welldeveloped and extends labially from the anterolingual side of the metaconid to the anterolabial side of the protoconid. The metaconid is the largest and tallest primary cusp. The protoconid and hypoconid are well developed and of about equal height. The entoconid is reduced to a transversely compressed cusp that is only slightly taller than the protoconid and hypoconid in unworn teeth and positioned just anterior to the posterolingual corner of the tooth. Following wear, the entoconid is difficult to distinguish on the hypolophid. The metalophid is a complete crest, connecting the metaconid to the protoconid. The mesolophid is a complete crest in all specimens, extending lingually from the ectolophid to the lingual margin of the tooth between the metaconid and entoconid (eight specimens) or to the labial side of the entoconid (two specimens). The ectolophid is a complete crest connecting the anterolabial edge of the hypoconid to the posterolabial edge of the protoconid. The hypolophid is a complete crest extending from the hypoconid in an uninterrupted arc to the entoconid. A posterior cingulid is lacking in all specimens. A mesostylid is lacking in seven specimens and incipiently represented as a slight swelling along the enamel edge between the metaconid and entoconid in three specimens. A mesoconid is lacking in eight specimens and incipiently represented by a slight widening on the ectolophid in two specimens. The m3 of Paradjidaumo reynoldsi is similar in size to those of Metanoiamys korthi, but it can be easily distinguished from those of M. korthi by lacking an adlophulid and having higher, better developed crests (anterior cingulid, metalophid, complete mesolophid, hypolophid) giving them a distinctive four crested occlusal pattern. Dental statistics for the entire sample of Paradjidaumo reynoldsi from the Simi Valley Landfill Local Fauna, including those specimens described by Kelly (1992), are provided in Tables 4-5. Measurements of the holotype Rp4 (LACM 131042) of Paradjidaumo reynoldsi are ap = 1.14 mm, tra = 0.90, and trp = 1.10 mm.

60

PALUDICOLA, VOL. 8, NO. 1, 2010

  Discussion―The newly discovered teeth of Paradjidaumo reynoldsi significantly increase the sample size for the species and provide additional information on individual variation. The new teeth include also the first records of dp4 and m3. Genus Simiacritomys Kelly, 1992 Simiacritomys whistleri Kelly, 1992 Figure 5 Referred Specimens―P4, LACM 153758, 153759; p4, LACM 153760, 153766; m2, LACM 153763, 153764; m3, LACM 153765. Description―Of the two teeth identified as P4, one is in early wear (Figure 5B) while the other is in late wear and missing a portion the labial margin of the tooth (Figure 5A). Because of the worn and broken condition of LACM 153759, the following description is based primarily on LACM 153758. The occlusal outline is subquadrate with a five-crested occlusal pattern. The paracone, protocone, metacone, and hypocone are well-developed conical cusps with the paracone the tallest and largest cusp, the metacone slightly lower in height than the paracone, and the protocone and hypocone of about equal height. The anterior cingulum is a well-developed crest that extends lingually from the anterior side of the paracone to about the midline of the tooth and then turns posteriorly to connect with protoloph, near its junction with the protocone. A very small anterocone is present on the anterior cingulum at this junction. The protoloph is almost complete, connecting the protocone to the paracone, but with a shallow notch or valley present on the protoloph at its junction with the anterior cingulum. A central transverse valley is present that extends from the posterolingual margin of the paracone to the lingual margin of the tooth, where it is open lingually. The metaloph is a complete crest connecting the hypocone to the metacone. From about the center of the metaloph, a slightly lower crest (= posterior portion of the endoloph) extends anterolabially into the central valley to an indistinct mesocone (incipient) and then continues anterolabially as a very small, low spur (= anterior portion of the endoloph) that terminates just short of the protoloph. A short, distinct mesoloph is present that extends labially from the endoloph to a ridge that extends from the posterior side of the paracone. The lingual portion of the mesoloph is narrowed anteroposteriorly, but then as it extends further labially it expands into a thick crest. The posterior cingulum is a well-developed crest that extends from the posterolabial side of the hypocone to the posterolingual edge of the metacone. A mesostyle is lacking. Even in its worn state, the occlusal morphology of LACM 153759 agrees well with that of LACM 153758, including the five crested pattern. The

primary differences seen in LACM 153759 are that the shallow valley or notch is lacking on the protoloph so that it becomes a complete crest and that the anterior portion of the endoloph is connected to the protoloph and interrupts the transverse central valley. These differences are regarded as resulting from the extremely worn condition of LACM 153758 and are not considered taxonomically significant. Welldeveloped appression facets are present on the posterior borders of LACM 153758 and 153759, but are completely lacking on their anterior borders, confirming that these teeth are P4s and indicating that a small P3 was probably not present in either specimen. The two upper premolars can be confidently assigned to Simiacritomys whistleri because they fit well within the observed range of variation noted by Kelly (1992) for M1-2 and exhibit the following shared characters with M1 or M2: 1) a five-crested occlusal pattern; 2) the paracone, protocone, metacone, and hypocone are well-developed conical cusps; 3) the anterior cingulum connects with the protoloph near its junction with the protocone; 4) the protoloph is interrupted by a shallow valley in early wear; 5) the metaloph is a complete, well developed crest; 6) the posterior cingulum extends from the metacone to the posterolabial edge of the hypocone; 7) the valleys between the crests are deep, including a central transverse valley that essentially separates the protoloph from the metaloph in early wear; 8) a welldeveloped posterior endoloph and weakly developed anterior endoloph; 9) a well-developed mesoloph; 10) an indistinct mesocone; 11) and absence of a mesostylid. They differ primarily from the M1 or M2 by being considerably smaller in size. Two teeth are identified as p4 (Figure 5C-D). One premolar is in a moderate stage of wear and has a portion of the enamel margin missing along the posterolingual margin of the entoconid and posterior cingulid. The other premolar is complete, but is heavily worn with the entoconid, hypoconid, hypolophid, and posterior cingulid worn down to a single, merged occlusal surface. Much of the following description of the occlusal morphology is derived from the less worn premolar with additional information provided by the heavily worn premolar. The anterior transverse width is considerably narrower than the posterior transverse width. The protoconid and metaconid are low, compressed cusps of about equal height that are connected by a short, slightly lower crest (? = anterior cingulid). The entoconid is the largest primary cusp and is connected to the hypoconid by a complete hypolophid. The posterior cingulid is well developed and slightly lower in height than the hypolophid, and extends from the posterolingual edge of the entoconid to the posterolabial corner of the hypoconid. In the less worn premolar, a transverse valley extends from the

KELLY—DUCHESNEAN RODENTS FROM CALIFORNIA

61

 

FIGURE 5. Teeth of Simiacritomys whistleri. A, partial RP4, LACM 153759; B, LP4, LACM 153758; C, Rp4, LACM 153760; D, Lp4, LACM 153766; E, Rm2, LACM 153763; F, partial Rm2, LACM 153764; G, Lm3, LACM 153765. All occlusal views, A-B with lingual at bottom, C-G with lingual at top, scale = 1 mm.

62

PALUDICOLA, VOL. 8, NO. 1, 2010

  lingual margin of the tooth to the labial margin, whereas in the worn premolar, only vestiges of this valley remain at the labial and lingual margins. In both premolars, a valley is present between the metaconid and protoconid that extends posterolingually to the hypolophid. In the less worn premolar, two small crests or spurs extend posteriorly from the metaconid, whereas in the worn premolar, one thicker crest extends posteriorly from the metaconid. In the less worn premolar, a moderately developed crest extends posterolingually from the posterior side of the protoconid, where it comes close to the anterior side of the hypoconid and then turns lingually to terminate a little over halfway across the central valley. A very small, low spur extends anteriorly from the labially directed portion of this crest to the base of the anterior wall of the hypolophid. This crest presumably represents the ectolophid and mesolophid. In the worn premolar, this crest extends posterolingually from the protoconid to connect with the hypolophid. In both premolars, a second small crest (lingual to the moderately developed crest) extends posterolingually from the protoconid to connect with the putative mesolophid in the unworn premolar or to the hypolophid in the worn premolar. On the worn premolar, this second crest exhibits a small labially directed spur and a slight connection with the moderately large crest at about midpoint. What is more significant than the minute details of the connections of these small crests in each premolar, which are probably attributable to individual variation and wear stage, is that in both premolars a distinct, relatively deep valley is present that separates the protoconid and metaconid and extends posterolingually towards the lingual margin of the tooth, similar to those seen in m1 or m2. In addition, the morphology of the entoconid and hypoconid, along with the connections of the hypolophid and posterior cingulid, are very similar to those of m1 or m2. Furthermore, a relatively deep transverse valley is also present, before late wear, between the hypolophid and the presumed mesolophid, similar also to those of m1 or m2. The premolars are considerably smaller than the m1 or m2, but this appears to correlate well with the size of the upper premolars, which are confidently assigned to the genus. For all of these reasons, the lower premolars are provisionally referred to Simiacritomys whistleri. Based on the presence of anterior and posterior appression facets, Kelly (1992) identified six lower molars as m2. However, without any intact lower dentitions of Simiacritomys for comparison and the fact that P4/p4 are now known to occur in the genus, these teeth should be regarded as m1 or m2. Two teeth are identified as m1 or m2 in the new sample (Figure 5EF). Their occlusal morphology agrees well with the

range of variation noted by Kelly (1992) in the original sample of m1 or m2s and does not provide any new descriptive information. One tooth identified as m3 (Figure 5G) is very worn and missing the enamel along the anterolabial margin of the protoconid. It can be confidently assigned to Simiacritomys whistleri because it exhibits the following characters: 1) the anterior cingulid extends labially beyond the small anteroconid; 2) the metalophid is incomplete, where it is interrupted by a transverse valley that extends from the lingual margin of tooth to the anterolabial margin of the tooth between the anterior cingulid and protoconid; 3) the mesolophid is a well developed, long crest extending from an indistinct metaconid (slight widening of the enamel) on the ectolophid to the labial margin of the tooth; 4) the mesolophid and the anterior portion of the ectolophid are separated from the hypolophid and hypoconid by a transverse valley; and 5) the hypolophid is a well developed, thick crest connecting a distinct hypoconid to a very weakly expressed (incipient) entoconid. It differs from the other m3s described by Kelly (1992) in lacking a distinct posterior cingulid. However, this is probably due to the extreme wear on the tooth because a wide, heavily worn shelf is present along the posterior part of the tooth where a short posterior cingulid would normally occur and the hypolophid exhibits a posteriorly directed protrusion along its posterior border where the posterior cingulid would normally connect with the hypolophid, suggesting that a posterior cingulid may have been present, but has been worn away. Measurements of the teeth in the new sample of Simiacritomys whistleri are provided in Table 5 and the dental statistics of the entire sample of S. whistleri, including those reported by Kelly (1992), are provided in Table 6. Measurements of the holotype Lm1 or m2 (LACM 131462) of Simiacritomys whistleri are ap = 1.76 mm, tra = 1.70 mm, and trp = 1.72 mm. Discussion―Based on 17 isolated teeth from the Simi Valley Landfill Local Fauna, Kelly (1992) named Simiacritomys. At the time, no upper or lower premolars were known for the genus. Kelly (1992) noted that the molar occlusal morphology of Simiacritomys exhibits similarities to those of both Eomyidae and certain members of Zapodidae (= Sicistinae and Zapodinae of Dipodidae sensu Flynn, 2008a). Kelly suggested that Simiacritomys may be a zapodid because its molar occlusal pattern appeared to be morphologically similar to those of the sicistine zapodid Plesiosminthus Viret, 1926, which was followed by Korth (1994). Flynn (2008b) placed Simiacritomys in Eomyidae incertae sedis, but noted that he was not convinced of this assignment.

KELLY—DUCHESNEAN RODENTS FROM CALIFORNIA

63

  P4 and p4 are present and molariform in Eomyidae, whereas p4 is absent and P4 is reduced to a small, single-rooted peg in Tertiary Sicistinae and ___________________________________________ TABLE 5. Measurements (in mm) of new sample of teeth of Simiacritomys whistleri from the Simi Valley Landfill Local Fauna. LACM # Tooth/position ap tra trp 153758 P4 1.27 1.37 1.26 153759 P4 1.27 ― ― 153760 p4 1.21 0.80 1.04 153766 p4 1.21 0.83 1.08 153763 m1 or m2 1.70 1.44 1.54 153764 m1 or m2 ― ― 1.49 153765. m3 1.64 1.41 1.17 _______________________________________________________ TABLE 6. Dental statistics for all teeth of Simiacritomys whistleri from Simi Valley Landfill Local Fauna. Includes sample reported on herein and specimens described by Kelly (1992). Tooth/ Dimension N Mean OR SD CV P4 ap 2 1.27 1.27 ― ― P4 tra 1 1.37 ― ― ― P4 trp 1 1.26 ― ― ― M1 or M2 ap 4 1.65 1.60-1.73 0.06 3.6 M1 or M2 tra 3 1.75 1.71-1.81 0.06 ― M1 or M2 trp 4 1.58 1.50-1.62 0.06 3.8 M3 ap 3 1.30 1.27-1.35 0.06 ― M3 tra 3 1.47 1.42-1.52 0.08 ― M3 trp 3 1.29 1.28-1.31 0.03 ― p4 ap 2 1.22 ― ― ― p4 tra 2 0.82 0.80-0.83 ― ― p4 trp 2 1.06 1.04-1.08 ― ― m1 or m2 ap 8 1.73 1.63-1.81 0.7 4.0 m1 or m2 tra 8 1.57 1.43-1.71 0.10 6.4 m1 or m2 trp 9 1.60 1.49-1.72 0.09 5.6 m3 ap 4 1.67 1.55-1.64 0.11 6.6 m3 tra 4 1.43 1.41-1.53 0.08 5.6 m3 trp 4 1.25 1.17-1.35 0.09 7.2 _______________________________________________________

Zapodinae (Korth, 1994; Flynn, 2008a-b). With the discovery that a molariform P4 and putative p4 are present in Simiacritomys, its referral to the Eomyidae appears to be confirmed. Without intact dentitions of Simiacritomys, the presence or absence of P3 cannot be unequivocally determined, but the lack of a small, anterior appression facet on the two known specimens of P4 suggests that P3 was lacking. The relationship of Simiacritomys to other Eomyidae is difficult to determine. It differs from early members of the Yoderimyinae Wood, 1955 (also see Black, 1965; Wood, 1974; Storer, 1987; Emry and Korth, 1993; Korth, 1994; Flynn, 2008b), by having the following: 1) P3 is apparently absent; 2) the P4/p4 are considerably smaller than M1-2/m1-2, respectively; 3) a distinct p4 anteroconid is lacking; 4) transverse valleys are present on the cheek teeth that usually interrupt the endolophs/ectolophids and protolophs/metalophids during early wear; and 5) the

molars are less lophodont and their anterior cingula/cingulids are less developed (narrower). Within the subfamily Eomyinae Winge, 1887, the Namatomyini Korth, 1992, are generally regarded as the most primitive (Storer, 1987; Korth, 1992, 1994). Simiacritomys differs from members of the Namatomyini by having the following: 1) relatively higher crowned cheek teeth; 2) P4-M2 and m1-3 exhibiting a five-crested occlusal pattern with much better developed mesolophs and mesolophids, respectively; 3) cheek teeth with more rounded, inflated primary cusps; and 4) a p4 anteroconid is lacking. Of the above characters that distinguish Simiacritomys from members of the Namatomyini, numbers one, three, and four are generally regarded as shared derived characters for the subfamily Eomyini (Storer, 1987; Korth, 1992, 1994). Simiacritomys differs from members of the Namatomyini and the Eomyini in having transverse valleys present that usually interrupt the protoloph and endoloph on the upper cheek teeth and metalophid and ectolophid on the lower cheek teeth, especially during early wear stages. In addition, if the p4s described herein are correctly assigned to Simiacritomys, it differs from all other eomyid genera by its unique p4 anterior occlusal morphology. For these reasons, Simiacritomys is herein regarded as an eomyid, but not assigned to any recognized subfamily. Eomyidae, genera and species undetermined Figure 6 Referred Specimens―M1 or M2, LACM 153712, 153767, 153768, 153797, 153809, 153816, 153823; M3, LACM 153713, 153845, 153850; m1 or m2, LACM 153761, 153762. Description―Twelve cheek teeth appear to represent the occurrence of additional eomyid species in the Simi Valley Landfill Local Fauna (Figure 6). In his preliminary identifications provided for the proprietary mitigation report for Waste Management of California, Inc., Walsh (2008) allocated some of these specimens as follows: 1) LACM 153712, 153713, 153714, Griphomys sp.; 2) LACM 153767 and 153768, Eomyidae; 3) LACM 153797, 153809, 153816, and 153823, Paradjidaumo reynoldsi; 4) LACM 153761, 153762, Simiacritomys whistleri. Except for LACM 153767 and 153768, his identifications are not supported by the morphological evidence presented here. In the following descriptions, the teeth are described and, in those that appear to have been originally identified incorrectly, the reasons are given for their new allocation. The three teeth that Walsh (2008) referred to Griphomys sp. are a RM1 or M2 (LACM 153712, Figure 6A) and two M3s (LACM 153713 and 153714).

64

PALUDICOLA, VOL. 8, NO. 1, 2010

 

FIGURE 6. Teeth of undetermined species of Eomyidae from Simi Valley Landfill Local Fauna. A, RM1 or M2, LACM 153712; B, partial RM1 or M2, LACM 153768; C, RM1 or M2, LACM 153767; D, LM1 or M2, LACM 153816; E, LM1 or M2, LACM 153797; F, partial RM1 or M2, LACM 153823; G, RM1 or M2, LACM 153809; H, RM3, LACM 153850; I, RM3, LACM 153845; J, LM3, LACM 153713; K, Rm1 or m2, LACM 153761; L, Lm1 or m2, LACM 153762. All occlusal views, A-J with lingual at bottom, K-L with lingual at top, scale = 1 mm.

KELLY—DUCHESNEAN RODENTS FROM CALIFORNIA

65

 

The M3s do not exhibit the bilophodont occlusal patterns characteristic of Griphomys (Wilson, 1940a; Lillegraven, 1977; Kelly and Whistler, 1994), where one actually represents an undetermined eomyid (Figure 6J) and the other Simimys landeri (Figure 7D). LACM 153712 is moderately well-worn and exhibits the following occlusal morphology: 1) an anterocone is lacking; 2) the primary cusps are distinct with the labial cusps (paracone and metacone) taller than the lingual cusps (protocone and hypocone); 3) the paracone has a short crest present that extends posteriorly to a point close to the labial margin of the middle of the central transverse valley; 4) a welldeveloped, complete protoloph and metaloph are present; 5) the anterior cingulum is a distinct crest that extends lingually from the anterior side of the paracone to the anterolabial side of the protocone, near its connection with the protoloph, thus forming an enclosed, relatively deep transverse valley between the anterior cingulum and protoloph; 6) the posterior cingulum is a distinct, relatively high crest that extends lingually from a point relatively high on the posterior side of the metacone to the posterolabial corner of the hypocone, thus forming an enclosed, relatively deep transverse valley between the posterior cingulum and the metaloph; and 7) a distinct endoloph with a small bulge along its center is present that extends anteriorly from the metaloph to about the middle of the protoloph. In the upper molars of Griphomys, the anterior cingulum extends nearly the entire width of the tooth connecting the posterior side of the paracone to the anterior side of the protocone and a protoloph spur (very small, thin crest) is sometimes present that extends posterolabially, usually for a very short distance, from the protocone or the lower portion of the protoloph near its connection with the protocone into the central transverse valley (see Lillegraven, 1977, fig. 21; Kelly and Whistler, 1994, fig. 15B-G). Similar to the upper molars, the lower molar anterior cingulid of Griphomys extends nearly the entire width of the tooth connecting the metaconid to the anterior part of the protoconid (see Lillegraven, 1977, figs. 22-24; Kelly and Whistler, 1994, fig. 15K-N). LACM 153712 differs from the upper molars of Griphomys in having an endoloph present that extends from the metaloph to the protoloph, and, more importantly, a much shorter anterior cingulum, where the connection of the anterior cingulum forms a Y-shaped occlusal pattern, like those of the Eomyidae. Walsh (2008) identified two M1 or M2s (LACM 154767 and 153768) as an undetermined eomyid. LACM 153767 is in early wear and exhibits the following occlusal morphology (Figure 6C): 1) the primary cusps are distinct, but somewhat lophate, with

the paracone slightly compressed transversely and the protocone compressed anteroposteriorly; 2) a thick, lingually curved ridge or crest extends posteriorly from the posterior side of the paracone to about the middle of the labial part of the central transverse valley; 3) the protoloph and metaloph are well-developed, high, complete crests connecting the protocone to the paracone and the hypocone to the metacone, respectively; 4) the anterior cingulum is a distinct, high crest that extends lingually from the anterior side of the paracone to the anterolabial side of the protocone, near its connection with the protoloph, thus forming an enclosed, relatively deep transverse valley between the anterior cingulum and protoloph; 5) the posterior cingulum is a well developed, high crest that extends from the posterolingual side of the metacone to the posterolabial side of the hypocone, forming an enclosed, relatively deep transverse valley between the posterior cingulum and metaloph; 6) the endoloph is a complete crest that extends anterolabially from the anterolabial corner of the hypocone to about the middle of the metaloph; and 7) an additional cuspule is present in the transverse central valley that is positioned at the anterior margin of the metaloph near its junction with the metacone. LACM 153768 is well worn with the labial portions of the paracone and metacone abraded away (Figure 6B). The occlusal morphology of LACM 153768 is very similar to that of LACM 153767, but differs in having a slightly thicker endoloph with a weak anteriorly directed bulge extending from its center (incipient mesoloph or mesocone) and lacking an additional cuspule along the anterolabial margin of the metaloph. Walsh (2008) assigned four M1 or M2s (LACM 153797, 153809, 153816, 153823) to Paradjidaumo reynoldsi (Figure 6D-G). Except for LACM 153823, the measurements of these teeth are at the high end of the observed range or slightly larger than those of P. reynoldsi (Table 3). LACM 154797, 153809, and 153816 exhibit a similar occlusal morphology, including the following: 1) the anterior cingulum, protoloph, metaloph, and posterior cingulum are welldeveloped crests that give the tooth a four-crested occlusal pattern; 2) the endoloph is a complete, high crest that extends from the metaloph to about the middle of the protoloph; 3) a mesostyle is present near the base of the posterolingual wall of the paracone; and 4) a mesoloph is lacking. These three specimens appear to represent the same species and differ primarily from those of P. reynoldsi by lacking a well-developed, long mesoloph and having the endoloph extending from the metaloph to about the middle of the protoloph. In P. reynoldsi, the endoloph extends from the anterolabial side of the hypocone to the posterolabial side of the

66

PALUDICOLA, VOL. 8, NO. 1, 2010

  protocone. LACM 153823 is missing a portion of the anterolabial corner of the tooth, including the paracone, and exhibits the following occlusal morphology (Figure 6F): 1) the anterior cingulum, protoloph, metaloph, and posterior cingulum are well-developed crests; 2) the endoloph extends from the anterolabial side of the hypocone to the middle of the protoloph with a distinct, labially directed spur or cuspule (incipient mesoloph or metacone) present at about the middle of the endoloph; and 3) a distinct cuspule (?mesostyle) is present, abutting the anterior base of the metacone. LACM 153823 is similar in size and occlusal morphology to LACM 153767, but differs by having a cuspule (?mesostyle) present near the anterior side of the metacone and an incipient mesoloph or mesocone present and lacking the small cuspule along the anterior margin of the metaloph. It is possible that LACM 153823 represents the same species as LACM 153767 and 153768. Walsh (2008) assigned three M3s as follows; LACM 153845 and 153850 to Paradjidaumo reynoldsi and LACM 153713 to Griphomys sp. However, these assignments are not supported by the occlusal morphology. LACM 153845 and 153850 (Figure 6H-I) exhibit the following characters: 1) a subtriangular occlusal outline; 2) the anterior cingulum extends lingually from the anterolingual side of the paracone to metaloph, connecting near the junction of the protocone and metaloph; 3) the protoloph and metaloph are well developed, connecting the protocone with the paracone and the metacone with the hypocone, respectively; 4) a distinct crest (endoloph) is present that extends anteriorly from the metaloph across the central valley to connect with the protoloph, and a small lingually directed spur or bulge (incipient mesoloph or mesocone) is present near the middle this crest; 5) the paracone and metacone are connected by a lingual crest (ectoloph) that closes off the central valley lingually; 6) the protocone is centrally positioned along the lingual margin of the tooth; 7) a small hypocone is present that is separated from the protocone by a valley; and 8) the posterior cingulum extends lingually from the posterolingual side of the metacone to the anterolabial side of the hypocone. LACM 153713 (Figure 6J) exhibits an occlusal morphology that is somewhat similar to LACM 153845 and 153850, but differs by having a shallow valley present between the labial end of the protoloph and the lingual side of the paracone and lacking a distinct valley between the protocone and hypocone, resulting in a distinct Yshaped occlusal pattern for the protocone, posterior cingulum, and metaloph. Walsh (2008) assigned two lower molars (LACM 153761 and 153762) to Simiacritomys whistleri. The occlusal morphology of these two molars includes the following (Figure 6K-L): 1) the anterior cingulid,

metalophid, hypolophid, and posterior cingulid are well developed, uninterrupted crests; 2) the anterior cingulid extends almost the entire width of the tooth from the anterior side of the metaconid to the anterior side of the protoconid; 3) the ectolophid extends anteriorly from the posterolingual corner of the protoconid to the anterior side of the hypoconid, where it is either weakly connected (LACM 153761) or separated from the hypoconid by a shallow notch (LACM 153762); 4) a well-developed, moderately long mesolophid is present that ends lingually from one-half to threequarters the way across the central transverse valley; and 5) the posterior cingulid extends from the posterior wall of the entoconid to the posterolabial side of the hypoconid, near the junction of the hypoconid and hypolophid. LACM 153761 and 153762 differ significantly from the m1 or m2 of S. whistleri by having the following characters: 1) an anteroconid is lacking; 2) the anterior cingulid is a continuous crest between the metaconid and protoconid, whereas in S. whistleri there is a labial extension of the anterior cingulid that is separated from the protoconid by a valley; 3) a complete metalophid and hypolophid are present, whereas valleys or distinct notches interrupt these crests in S. whistleri (compare Figure 5F with Figure 6L); and 4) smaller size. The overall occlusal patterns of LACM 153761 and 153762 are similar to those of m1 or m2 of P. reynoldsi, except that the mesolophids are slightly shorter and their size is in the uppermost observed range of P. reynoldsi or slightly larger. Measurements of the Sespe eomyid teeth are presented Table 7. ____________________________________________ TABLE 7. Measurements (in mm) of teeth of Eomyidae from Simi Valley Landfill Local Fauna. LACM # Tooth/position ap tra trp 153712 M1 or M2 1.21 1.32 1.28 153767 M1 or M2 1.13 1.23 1.08 153768 M1 or M2 1.23 ― ― 153797 M1 or M2 1.16 1.41 1.36 153809 M1 or M2 1.28 1.36 1.34 153816 M1 or M2 1.21 1.39 1.36 153823 M1 or M2 ― ― 1.08 153713 M3 1.16 1.22 1.03 153845 M3 1.12 1.30 1.00 153850 M3 1.14 1.31 1.08 153761 m1 or m2 1.31 ― 1.20 153762 m1 or m2 1.39 1.31 1.35 _______________________________________________________

Discussion―All of the Sespe M1 or M2s described above exhibit an eomyid pattern with four well-developed transverse crests (anterior cingulum, protoloph, metaloph, hypoloph). They differ from those of Paradjidaumo in lacking distinct mesolophs and having an endoloph that extends anterolabially

KELLY—DUCHESNEAN RODENTS FROM CALIFORNIA

67

  from the anterolabial side of the hypocone or metaloph to about the middle of the protoloph, whereas on M1 or M2 of Paradjidaumo, the mesoloph is usually a long, well-developed crest and the endoloph typically extends from the hypocone to the protocone (Burke, 1934; Wood, 1937; Black, 1965; Setoguchi, 1978; Kelly, 1992). They can easily be distinguished from those of Metanoiamys in having the following: 1) much larger size; 2) complete, better developed crests (more lophodont); and 3) lacking an adlophule and labial extension of the anterior cingulum. They can also be easily distinguished from those of Simiacritomys by their much smaller size and by having a more complete endoloph and protoloph (not interrupted by a valley), and lacking a distinct mesoloph. Two of the specimens (LACM 153767, 153823) are slightly smaller than the other five M1 or M2s and four of the M1 or M2s (LACM 153712, 153767, 153768, 153823) appear to differ from the others by exhibiting slightly more lophodont crests, especially LACM 153767, where the anterior and posterior cingula are relatively taller and connect higher on the paracone and metacone, respectively, resulting in enclosed, relatively deep valleys between the anterior and posterior cingula and the protoloph and metaloph, respectively. These differences suggest that LACM 153712, 153767, 153768, and 153823 may represent a different eomyid species from the other three M1 or M2s. LACM 153797, 153809, and 153816 are similar in size and have very similar occlusal patterns, indicating they probably represent the same species. These teeth exhibit some similarity in their occlusal morphology to those of P. reynoldsi, but the differences cited above indicate that they do not represent this species. They also exhibit some similarity in their occlusal morphology to those of Protadjidaumo Burke, 1934, and Cristadjidaumo Korth and Eaton, 2004. They differ from both of these genera by having the endoloph connecting anteriorly with the protoloph instead of the protocone. They further differ from those of Cristadjidaumo in lacking a high, moderatelydeveloped mesoloph. In most Eocene genera of Eomyidae where the M1 and M2 are known, the endoloph is positioned lingual of the center of the tooth and extends from the hypocone to connect with the protocone (e.g., Wood, 1936, 1937, 1974; Jacobs, 1977; Setoguchi, 1978; Storer, 1984, 1987; Korth, 1989, 1992; Wilson and Runkel, 1991; Korth and Bailey, 1992; Korth and Eaton, 2004). However, in members of the Yoderimyinae, the endoloph is positioned less lingually, often near the center of the tooth, and extends from the hypocone or the lingual portion of the metaloph just above its connection with the hypocone to the protoloph (Wood, 1955, 1974; Emry and Korth, 1993). Interestingly, all of the Sespe M1 or M2s

described above also exhibit more centrally positioned endolophs that extend from the hypocones or metalophs to the protolophs. In addition, on LACM 153767, which is in very early wear, the crests and cusps form a moderately lophodont occlusal pattern, somewhat similar to the early yoderimyid Litoyoderimys Emry and Korth, 1993. However, the Sespe eomyid M1 or M2s lack a long anterior cingulum with a lingual extension, a well developed mesoloph, and an ectoloph, which are typical characters observed in the upper molars of the Yoderimyinae (Emry and Korth, 1993). Although the overall occlusal patterns of all of the M1 or M2s are eomyid-like, they do not agree specifically with any recognized eomyid genus and, therefore, are referred to Eomyidae, genera and species undetermined. Of the three new M3s, LACM 153845 and 153850 (Figure 6H-I) differ from those of P. reynoldsi by being much larger in size and lacking welldeveloped mesolophids. LACM 153713 (Figure 6J) differs from those of Griphomys by having the following: 1) an anterior cingulum that extends lingually from the paracone to the metaloph, connecting near its junction with the protocone and resulting in a Y-shaped occlusal pattern for the protocone, anterior cingulum, and protoloph; 2) a welldeveloped crest in the central valley that extends from the protoloph to the metaloph; 3) an ectoloph is present; and 4) a more centrally positioned protocone (compare Figure 6J with fig. 15G of Kelly and Whistler, 1994). Furthermore, the M3 of Griphomys exhibits a prominent bilophodont occlusal pattern and, when present, the protoconal spur or mesocone is represented as a small, low spur or cuspule that is connected to the posterolabial side of the protocone and only extends a short distance into the central valley (Kelly and Whistler, 1994). The differences in occlusal morphology between LACM 154713 and LACM 153845 and 153850 noted above indicate that it represents a different species. The three new M3s do not agree in occlusal morphology with any of the recognized rodent species from the Simi Valley Landfill Local Fauna, but do exhibit a four-crested occlusal pattern, like those of the Eomyidae. They are smaller than those of Simiacritomys and Simimys landeri, but comparable in size to the other undetermined eomyid M1 or M2s described above. Without intact dentitions for comparison, confident generic assignment is not possible, so these specimens are referred to Eomyidae, genus and species undetermined. The two lower molars (LACM 153761, 153762) exhibit occlusal patterns that are very similar to those of Paradjidaumo reynoldsi, but they are either larger than (LACM 153762) or in the upper limit of the observed range of P. reynoldsi (LACM 153761) and,

68

PALUDICOLA, VOL. 8, NO. 1, 2010

  on one of them (LACM 153761), the mesolophid is less developed (relatively shorter). However, Korth (1980) noted that in large samples of Paradjidaumo from Nebraska, although the mesolophid is usually a long crest, its development can be variable. LACM 153761 and 153762 may represent very large m1 or m2s of P. reynoldsi, but they could instead represent the lower molars of the undetermined eomyid species described above that has upper molars (LACM 153797, 153809, 153816) that are similar in size to LACM 153761 and 153762. In addition, the lower molars exhibit an occlusal morphology that is similar to those of Cristadjidaumo, but differ in the connections of the anterior cingulid and by having a less V-shaped ectolophid. In the Sespe lower molars, the anterior cingulid extends from the anterior side of the protoconid to the anterolingual base of the metaconid, whereas in Cristadjidaumo the anterior cingulid connects with the anterior arm of the protoconid and both the labial and lingual ends are free, not connecting with any cusp (Korth and Eaton, 2004). Because the lower molars cannot be confidently assigned to any recognized eomyid genus, they are referred to Eomyidae, genus and species undetermined. Family Simimyidae Wood, 1980 Genus Simimys Wilson, 1935b Simimys landeri Kelly, 1992 Figure 7 Referred Specimens―M1, LACM 153956, 153957; M2, LACM 153967, 153968, 153969, 153970; M3, LACM 153714, 153976, 153977, 153978, 153979, 153980, 153981; m1, LACM 153958, 153959, 153960, 153961, 153962, 153963, 153964, 153965, 153966; m2, LACM 153971, 153972, 153973, 153974, 153975; m3, LACM 153982, 153983, 153984. Discussion―The occlusal morphology of the cheek teeth in the new sample of Simimys landeri fits well within the variation described by Kelly (1992) and provides no new descriptive information (Figure 7AH). It should be noted that Kelly (1992) incorrectly stated in the description of S. landeri, which was also repeated in the diagnosis, that “the m3 lingual metalophid is bifurcated at its lingual end, with one spur of the bifurcation connecting to mesostylid and other spur to the base of the entoconid.” This was a typographical error and should have read m3 lingual “mesolophid.” The diagnostic characters that Kelly (1992) used to distinguish S. landeri from S. simplex (Wilson, 1935a) are supported by the new sample, including much larger cheek teeth with less development of the small accessory crests and stylids, a more prominent M3 hypocone, and a lingually bifurcated m3 mesolophid with one end connecting to

the mesostylid and the other end connecting to the base of the entoconid. Dental statistics for the entire sample of Simimys landeri from the Simi Valley Landfill Local Fauna, including those specimens described by Kelly (1992), are provided in Table 8. Measurements of the holotype (RM1, LACM 131062) are ap = 1.71 mm, tra = 1.45 mm, trp = 1.52 mm. ____________________________________________ TABLE 8. Dental statistics (in mm) for all teeth of Simimys landeri from Simi Valley Landfill Local Fauna. Includes new sample provided herein plus specimens reported by Kelly (1992). Tooth/ Dimension M1 ap M1 tra M1 trp M2 ap M2 tra M2 trp M3 ap M3 tra M3 trp m1 ap m1 tra m1 trp m2 ap m2 tra m2 trp m3 ap m3 tra m3 trp

N 7 6 6 8 8 8 11 11 10 10 10 11 9 9 9 7 7 7

Mean 1.76 1.45 1.53 1.72 1.54 1.47 1.31 1.22 1.11 1.78 1.03 1.22 1.89 1.35 1.42 1.46 1.16 0.95

OR 1.64-1.89 1.28-1.55 1.41-1.66 1.51-1.80 1.33-1.64 1.28-1.58 1.23-1.36 1.15-1.33 1.03-1.21 1.67-1.87 0.89-1.10 1.08-1.32 1.75-2.05 1.21-1.50 1.28-1.56 1.34-1.56 1.08-1.20 0.83-1.08

SD 0.10 0.09 0.07 0.10 0.11 0.09 0.03 0.06 0.08 0.07 0.08 0.08 0.10 0.09 0.09 0.07 0.05 0.10

CV 5.7 6.2 4.6 5.8 7.1 6.1 2.3 4.9 5.4 3.9 7.8 6.6 5.2 6.7 6.3 4.8 4.3 10.5

____________________________________________ Family Cylindrodontidae Miller and Gidley, 1918 Genus Pareumys Peterson, 1919 Pareumys sp. Figure 8 Referred Specimens―partial M1 or M2, LACM 153955. Description―The partial upper molar is missing the portion anterior to the protoloph, the anterolabial portion of the paracone, the posterolabial side of the metacone, and the lingual margin of the tooth (Figure 8). Even with the missing portions, the overall occlusal outline appears round or oval in shape. The paracone, metacone, and metaconule are moderately developed cusps with distinct wear facets. The protoloph is a tall, complete loph extending between the paracone and the protocone. A moderately deep valley is present between the protoloph and metaloph that also extends posteriorly and then turns labially, separating the metaloph from the protocone and posterior cingulum. The metaloph extends, in an almost straight line, from the anterolingual corner of the metacone to the metaconule. The partial protocone is represented by a bulge in the enamel at the anterolingual side of the broken tooth. A similar, but smaller, bulge in the

KELLY—DUCHESNEAN RODENTS FROM CALIFORNIA

69

 

FIGURE 7. Teeth of Simimys landeri. A, LM1, LACM 153957; B, LM2, LACM 153969; C, RM3, LACM 153980; D, RM3, LACM 153714; E, Lm1, LACM 153960; F, Rm1, LACM 153965; G, Rm2, LACM 153973; Rm3, LACM 153984. All occlusal views, A-D with lingual at bottom, E-H with lingual at top, scale = 1 mm. ________________________________________________________________________________________________________________________

enamel at the posterolingual side of the broken tooth indicates that a hypocone was present. The morphology of the anterior cingulum cannot be determined because

it is broken off from the tooth. The posterior cingulum is a well-developed loph that extends from the posterolingual corner of the metacone to the

70

PALUDICOLA, VOL. 8, NO. 1, 2010

  posterolabial corner of the hypocone. An endoloph is present between the protocone and hypocone. The endoloph and the posterior cingulum together form a continuous curved arch from the protocone to the metacone. All of the lophs are nearly equal in height, giving the tooth a flat-topped appearance. Measurements of the partial tooth are ap = 2.10 mm and tr = 2.13 mm. ____________________________________________

FIGURE 8. Pareumys sp. from Simi Valley Landfill Local Fauna, partial upper molar, LACM 153955, occlusal view, lingual at bottom, scale = 1 mm. ___________________________________________________

Discussion―LACM 153955 can be confidently assigned to Pareumys because it exhibits a flat-topped, lophate occlusal morphology, an incomplete metaloph that is separated from the protocone, an overall circular occlusal outline, and is lacking a mesoloph (Figure 8). Pareumys is a relatively rare taxon in the Eocene of California (Wilson, 1940b; Lindsay, 1968; Lillegraven, 1977; Kelly, 1990; Walsh, 1991, 1996). Kelly (1990) provided the most recent summary of the dental characters and known occurrences of Pareumys from southern California. LACM 153955 is similar to Pareumys sp. near P. grangeri Burke, 1935, from the Uintan Friars and Mission Valley formations of the San Diego area (Wilson, 1940b; Lillegraven, 1977) in having a straight metaloph that is well separated from the posterior cingulum, but appears derived relative to this taxon by having much better developed (higher) lophs (protoloph, metaloph, and posterior cingulum) resulting in a flatter topped occlusal surface. It differs from Pareumys sp., aff. P. milleri Peterson, 1919, from the Uintan Brea Canyon and Duchesnean Pearson Ranch local faunas (Wilson, 1940b; Kelly, 1990) in having a straight metaloph, rather than a posteriorly curved metaloph that joins or nearly joins the posterior cingulum. Although LACM 153955 may represent a new species of Pareumys, it is assigned to an

indeterminate species due to its fragmentary nature. This is the first record of the genus in the Simi Valley Landfill Local Fauna. CONCLUSIONS This report documents the discovery of a number of additional isolated rodent teeth from the Duchesnean Simi Valley Landfill Local Fauna that were recovered during a paleontologic mitigation program at the Simi Valley Landfill and Recycling Center. Included in these teeth are larger samples of Metanoiamys korthi, Paradjidaumo reynoldsi, Simiacritomys whistleri, and Simimys landeri. These larger sample sizes allow a detailed assessment of the individual variation of the occlusal morphology of the cheek teeth in these species, which provides criteria for determining the validity of their specific diagnostic characters. Prior to this report, the dP4, M3, dp4, p4, and m3 of M. korthi, the dp4 and m3 of P. reynoldsi, and the P4 and putative p4 of S. whistleri were unknown. The familial status of S. whistleri has been previously questioned (Kelly, 1992; Flynn, 2008b). However, with the discovery that a molariform P4 and putative p4 are present in S. whistleri, this species is now confidently referred to the Eomyidae. Additional new records for the Simi Valley Landfill Local Fauna include at least one or more undetermined eomyid species and Pareumys sp. A revised faunal list for the Simi Valley Landfill Local Fauna is provided in Table 9 based on the new records provided herein and those of Kelly (1992, 2009, 2010), Kelly and Whistler (1994, 1998), and Kelly et al. (1991). ___________________________________________ TABLE 9. Revised faunal list for the Simi Valley Landfill Local Fauna (Kelly, 1992, 2009, 2010, this paper; Kelly and Whistler, 1994, 1998; Kelly et al., 1991). Reptilia Squamata Glyptosaurini, gen. and sp. indet. Melanosaurini, gen. and sp. indet. Mammalia Didelphimorpha Herpetotheriidae Herpetotherium sp. Peradectidae Peradectes californicus Stock, 1936 Erinaceomorpha Sespedectidae Sespedectes singularis Stock, 1935 Soricomorpha Geolabididae Batodonoides walshi Kelly, 2010 Oligoryctidae cf. Oligoryctes sp. Primates Microsyopidae cf. Uintasorex sp. Rodentia

KELLY—DUCHESNEAN RODENTS FROM CALIFORNIA

71

  Ischyromyidae Leptotomus sp. Eomyidae Metanoiamys korthi Kelly and Whistler, 1998 Paradjidaumo reynoldsi Kelly, 1992 Simiacritomys whistleri Kelly, 1992 Eomyidae, gen. and spp. undetermined Heliscomyidae Heliscomys walshi Kelly, 2009 Simimyidae Simimys simplex (Wilson, 1935a) Simimys landeri Kelly, 1992 Artiodactyla Camelidae Camelidae, gen. and sp. undetermined Hypertragulidae Simimeryx sp., aff. S. hudsoni Stock, 1934 _______________________________________________________

ACKNOWLEDGEMENTS I am indebted to Samuel A. McLeod of the LACM for his support and providing access to the new sample of small mammal teeth from Sespe Formation. I am grateful to Rachel Dolbier of the W. M. Keck Earth Science and Mineral Engineering Museum, University of Nevada, Reno, for help in securing an interinstitutional loan of the specimens reported on here. Special thanks are given to E. Bruce Lander of Paleo Environmental Associates, Inc. (PEA) and the LACM, and Mark A. Roeder of the San Diego Natural History Museum (SDMNH) and PEA for their support and advice during the preparation of this paper. I am thankful to William W. Korth of the Rochester Institute of Vertebrate Paleontology, David P. Whistler of the LACM, and Mary R. Dawson of the Carnegie Museum of Natural History for their constructive comments and advice on the original draft of this report. My appreciation and admiration goes out to the late Steven L. Walsh of the SDMNH for his many insightful comments and conversations regarding Eocene small mammals that he provided me over a 20 year period. Waste Management of California, Inc., provided funding for the Paleontologic Resource Impact Mitigation Program at the Simi Valley Landfill and Recycling Center. LITERATURE CITED Black, C. C. 1965. Fossil mammals from Montana. Annals of Carnegie Museum 38:1-48. Burke, J. J. 1934. New Duchesne River rodents and a preliminary survey of the Adjidaumidae. Annals of Carnegie Museum 23:391-398. Chiment, J. J. and W. W. Korth. 1996. A new genus of eomyid rodent (Mammalia) from the Eocene (Uintan-Duchesnean) of southern California. Journal of Vertebrate Paleontology 16:116124.

Emry, R. J. and W. W. Korth. 1993. Evolution in Yoderimyinae (Eomyidae: Rodentia), with new material from the White River Formation (Chadronian) at Flagstaff Rim, Wyoming. Journal of Paleontology 67:1047-1057. Flynn, L. J. 2008a. Dipodidae, in Janis, C. M., G. F. Gunnell, and M. D. Uhen, eds., Evolution of Tertiary mammals of North America Volume 2: small mammals, xenarthans, and marine mammals. Cambridge, Cambridge University Press, p. 406-414. Flynn, L. J. 2008b. Eomyidae, in Janis, C. M., G. F. Gunnell, and M. D. Uhen, eds., Evolution of Tertiary mammals of North America Volume 2: small mammals, xenarthrans, and marine mammals. Cambridge, Cambridge University Press, p. 415-427. Golz, D. J. 1976. Eocene Artiodactyla of southern California. Natural History Museum of Los Angeles County, Science Bulletin 26:1-85. Golz, D. J. and J. A. Lillegraven, 1977. Summary of known occurrences of terrestrial vertebrates from Eocene strata of southern California. University of Wyoming Contributions in Geology 15:43-65. Jacobs, L. L. 1977. Rodents of the Hemphillian age Redington Local Fauna, San Pedro Valley, Arizona. Journal of Paleontology 51:505-519. Kelly, T. S. 1990. Biostratigraphy of Uintan and Duchesnean land mammal assemblages from the middle member of the Sespe Formation, Simi Valley, California. Natural History Museum of Los Angeles County, Contributions in Science 419:1-42. Kelly, T. S. 1992. New Uintan and Duchesnean (middle and late Eocene) rodents from the Sespe Formation, Simi Valley, California. Southern California Academy of Sciences Bulletin 91:97-120. Kelly, T. S. 2009. A new species of Heliscomys (Rodentia, Heliscomyidae) from the Duchesnean (middle Eocene) Simi Valley Landfill Local Fauna, Sespe Formation, California. Paludicola 7:67-77. Kelly, T. S. 2010. New records of Marsupialia, Lipotyphla, and Primates from the Duchesnean (middle Eocene) Simi Valley Landfill Local Fauna, Sespe Formation, California. Paludicola 7:158-169. Kelly, T. S. and D. P. Whistler. 1994. Additional Uintan and Duchesnean (middle and late Eocene) mammals from the Sespe Formation, Simi Valley, California. Natural History Museum of Los Angeles County, Contributions in Science 439:1-29.

72

PALUDICOLA, VOL. 8, NO. 1, 2010

  Kelly, T. S. and D. P. Whistler. 1998. A new eomyid rodent from the Sespe Formation of southern California. Journal of Vertebrate Paleontology 18:440-443. Kelly, T. S., E. B. Lander, D. P. Whistler, M. A. Roeder, and R. E. Reynolds. 1991. Preliminary report on a paleontologic investigation of the lower and middle members, Sespe Formation, Simi Valley Landfill, Ventura County, California. PaleoBios 13:1-13. Korth, W. W. 1980. Paradjidaumo (Eomyidae, Rodentia) from the Brule Formation, Nebraska. Journal of Paleontology 54:933941. Korth, W. W. 1989. Geomyoid rodents (Mammalia) from the Orellan (middle Oligocene) of Nebraska, in Black, C. C. and M. R. Dawson, eds., Papers on Fossil Rodents in Honor of Albert Elmer Wood, Natural History Museum of Los Angeles County, Science Series 33:3146. Korth, W. W. 1992. Cylindrodonts (Cylindrodontidae, Rodentia) and a new genus of eomyid, Paranamatomys, (Eomyidae, Rodentia) from the Chadronian of Sioux County, Nebraska. Transactions of the Nebraska Academy of Sciences 19:75-82. Korth, W. W. 1994. The Tertiary record of rodents in North America. Topics in Geobiology, Vol. 12. New York, Plenum Press, 319 p. Korth, W. W. and B. E. Bailey. 1992. Additional specimens of Leptodontomys douglassi (Eomyidae, Rodentia) from the Arikareean (late Oligocene) of Nebraska. Journal of Mammalogy 73:651-662. Korth, W. W. and J. G. Eaton. 2004. Rodents and a marsupial (Mammalia) from the Duchesnean (Eocene) Turtle Basin Local Fauna, Sevier Plateau, Utah. Bulletin of Carnegie Museum of Natural History 36:109-119. Lander, E. B. 2008. Simi Valley Landfill and Recycling Center landfill expansion, Ventura County, California, paleontologic resource impact mitigation program, fourteenth progress report for period January 1, 2003 to December 31, 2007, project 2006-8. Prepared for Waste Management of California, Inc., Simi Valley Landfill and Recycling Center, 35 p. Lillegraven, J. A. 1977. Small rodents (Mammalia) from Eocene deposits of San Diego County, California. Bulletin of the American Museum of Natural History 158:221-261. Lillegraven, J. A. and R. W. Wilson. 1975. Analysis of Simimys simplex, an Eocene rodent

(?Zapodidae). Journal of Paleontology 49:856-874. Lindsay, E. 1968. Rodents of the Hartman Ranch Local Fauna, California. PaleoBios 6:1-22. Luterbacher, H. P., J. R. Ali, H. Brinkhuis, F. M. Gradstein, J. J. Hooker, S. Monechi, J. G. Ogg, J. Powell, U. Röhl, A. Sanfilippo, and B. Schmitz. 2004. The Paleogene period, in Gradstein, F. M., J. G. Ogg, and A. G. Smith, eds., A geologic time scale 2004, Cambridge, Cambridge University Press, p. 384-408. Mason, M. A. 1988. Mammalian paleontology and stratigraphy of the early to middle Tertiary Sespe and Titus Canyon Formations, southern California. Ph.D. dissertation, University of California, Berkeley, 257 p. Miller G. S., Jr. and J. W. Gidley. 1918. Synopsis of the supergeneric groups of rodents. Journal of the Washington Academy of Sciences 8:431448. Peterson, O. A. 1919. Report upon the material discovered in the upper Eocene of the Uinta Basin by Earl Douglass in the years 19081909 and O. A. Peterson in 1912. Annals of the Carnegie Museum 12:40-168. Prothero, D. R., J. L. Howard, and T. H. Huxley Dozier. 1996. Stratigraphy and paleomagnetism of the upper middle Eocene to lower Miocene (Uintan to Arikareean) Sespe Formation, Ventura County, California, in Prothero, D. R. and R. J. Emry, eds., The Terrestrial Eocene-Oligocene Transition in North America, Cambridge University Press, New York, p. 171-188. Setoguchi, T. 1978. Paleontology and geology of the Badwater Creek area, central Wyoming, Part 16. The Cedar Ridge Local Fauna (late Oligocene). Bulletin of Carnegie Museum of Natural History 9:1-61. Stock, C. 1934. A hypertragulid from the Sespe uppermost Eocene, California. Proceedings of the National Academy of Sciences 20:625629. Stock, C. 1935. Insectivora from the Sespe uppermost Eocene, California. Proceedings of the National Academy of Sciences 21:214-219. Stock, C. 1936. Sespe Eocene didelphids. Proceedings of the National Academy of Sciences 22:122124. Storer, J. E. 1984. Mammals of the Swift Current Creek Local Fauna (Eocene: Uintan, Saskatchewan). Natural History Museum Contributions 7:1-158. Storer, J. E. 1987. Dental evolution and radiation of Eocene and early Oligocene Eomyidae (Mammalia, Rodentia) of North America,

KELLY—DUCHESNEAN RODENTS FROM CALIFORNIA

73

  with new material from the Duchesnean of Saskatchewan. Dakoterra 3:108-117. Viret, J. 1926. Nouvelles observations relatives á la faune de Rongeurs de Saint-Gérand-le-Puy. Comptes-rendus hebdomadires des Séances 183:71-72. Walsh, S. L. 1991. Eocene mammal faunas of San Diego County, in Abbott, P. L. and J. A. May, eds., Eocene Geologic History San Diego Region. Pacific Section, Society of Economic Paleontologists and Mineralogists Book 68, p. 161-177. Walsh, S. L. 1996. Middle Eocene mammalian faunas of San Diego County, California, in Prothero, D. R. and R. J. Emry, eds., The Terrestrial Eocene-Oligocene Transition in North America, New York, Cambridge University Press, p. 75-119. Walsh, S. L. 1997. New specimens of Metanoiamys, Pauromys, and Simimys (Rodentia: Myomorpha) from the Uintan (middle Eocene) of San Diego County, California, and comments on the relationships of selected Paleogene Myomorpha. Proceedings of the San Diego Society of Natural History 32:1-20. Walsh, S. L. 2008. Appendix C, in Lander, E. B., Simi Valley Landfill and Recycling Center landfill expansion, Ventura County, California, paleontologic resource impact mitigation program, fourteenth progress report for period January 1, 2003 to December 31, 2007. Paleo Environmental Associates, Inc., project 20068. Prepared for Waste Management of California, Inc., Simi Valley Landfill and Recycling Center, p.32-35. Wilson, J. A. and A. C. Runkel. 1991. Prolapsus, a large sciuravid rodent and new eomyids from the late Eocene of Trans-Pecos, Texas. Pearce-Sellards Series, Texas Memorial Museum 48:1-30.

Wilson, R. W. 1935a. Cricetine-like rodents from the Sespe Eocene of California. Proceedings of the National Academy of Sciences 21:26-32. Wilson, R. W. 1935b. Simimys, a new name to replace Eumysops Wilson, preoccupied―A correction. Proceedings of the National Academy of Sciences 21:179-180. Wilson, R. W. 1940a. Two new Eocene rodents from California. Contributions to Paleontology, Carnegie Institution of Washington Publication 514:85-95. Wilson, R. W. 1940b. Pareumys remains from the later Eocene of California. Contributions to Paleontology, Carnegie Institution of Washington Publication 514:97-108. Winge, H. 1887. Jordfundne og nulevende Gnavere (Rodentia) fra Lagoa Santa, Minas Geraes, Brasilien. E. Mueso Lundii, University of Copenhagen 1:1-178. Wood, A. E. 1936. A new rodent from the Pliocene of Kansas. Journal of Paleontology 10:392-394. Wood, A. E. 1937. The mammalian fauna of the White River Oligocene, Part II, Rodentia. Transactions of the American Philosophical Society 28:155-269. Wood, A. E. 1955. Rodents from the lower Oligocene Yoder Formation of Wyoming. Journal of Paleontology 29:519-524. Wood, A. E. 1974. Early Tertiary vertebrate faunas Vieja Group Trans-Pecos Texas: Rodentia. Texas Memorial Museum Bulletin 21:1-112. Wood, A. E. 1980. The Oligocene rodents of North America. Transactions of the American Philosophical Society70:1-68. Wood, A. E. and R. W. Wilson. 1936. A suggested nomenclature for the cusps of the cheek teeth of rodents. Journal of Paleontology 10:388391.

ROCHESTER INSTITUTE OF VERTEBRATE PALEONTOLOGY The Rochester Institute of Vertebrate Paleontology (RIVP) was founded in December of 1992 as a nonprofit corporation for the purpose of furthering scientific research in Vertebrate Paleontology. Officers President: William W. Korth Provost: Judy A. Massare

Research Associates Jon A. Baskin Department of Biological and Health Sciences, Texas A&M University-Kingsville. Kingsville, TX 78363 Jeffrey Chiarenzelli Department of Geology St. Lawrence University 23 Romoda Drive Canton, NY 13617 Robert L. Evander Department of Vertebrate Paleontology American Museum of Natural History Central Park West at 79th St. New York, NY 10024 Margaret Diamond 5215 Ledge Lane Williamsville, NY 14221

Board of Directors William W. Korth, Rochester, NY Judy A. Massare, Rochester, NY George McIntosh, Rochester, NY

Mark C. Lawler 33 Wigwam Road Pennellville, NY 13132 Robert A. Martin Department of Biological Sciences Murray State University Murray, KY 42071 M. Amy Sheldon 345 Stevens St. Raton, NM 87740 William R. Wahl Wyoming Dinosaur Center 110 Carter Ranch Rd. Thermopolis, WY 82443

CONTENTS First report of fossil Amphiuma (Amphibia: Caudata: Amphiumidae) from South Carolina, USA JAMES L. KNIGHT AND DAVID J. CICIMURRI

1-7

Mammals from the Blue Ash local fauna (late Oligocene), South Dakota. Rodentia, Part 6: Family Castoridae and additional Eomyidae with a summary of the complete rodent fauna WILLIAM W. KORTH

8-13

North American Promimomys immigration event

ROBERT A. MARTIN

14-21

An Ichthyosaurus (Reptilia, Ichthyosauria) with gastric contents from Charmouth, England: first report of the genus from the Pliensbachian DEAN R. LOMAX

22-36

Fossil chimaeroid remains (Chondrichthyes: Holocephali) from Williamsburg County, South Carolina, USA DAVID J. CICIMURRI

37-48

New records of Rodentia from the Duchesnean (middle Eocene) Simi Valley landfill local fauna, Sespe Formation, California THOMAS S. KELLY

49-73