BUREAU OF ECONOMIC GEOLOGY The University of Austin 12, Texas

Peter T. Flawn, Director

Report of Investigations—No. 48

The Hill-Shuler Local Faunas of the Upper Trinity River, and Denton Counties, Texas

By BOB H. SLAUGHTER, WILSON W. CROOK, JR.,R. K. HARRIS, D. C. ALLEN, AND MARTIN SEIFERT

December 1962

BUREAU OF ECONOMIC GEOLOGY The University of Texas Austin 12, Texas

Peter T. Flawn, Director

Report of Investigations—No. 48

The Hill-Shuler Local Faunas of the Upper Trinity River, Dallas and Denton Counties, Texas

By BOB H. SLAUGHTER, WILSON W. CROOK, JR., R. K. HARRIS, D. C. ALLEN, AND MARTIN SEIFERT

December 1962

Contents

Abstract —.. 1

Introduction - 1 Abbreviations and terms 4

Geology of the terrace deposits - - 5 Comparisons and interpretations 8

Hill-Shuler local faunas , 10 Location and discussion of sites 10 Lagow pit 10 Gifford-Hill pit 10 Pemberton Hill 10 Moore pit 10 Wood pit 11 Coppel locality 11 Hickory Creek locality 11 Lewisville site 11 Seale pit 11 Haymarket pit 11 Vertebrate from the T-2 deposits 12 Systematic paleontology 13 Phylum Vertebrata Class Mammalia Order Marsupialia Family Didelphidae Didelphis cf. D. marsupialis Linnaeus 13 Order Insectivora Family Talpidae Scalopus aquaticus (Linnaeus) 13 Order Edentata Family Glyptodontidae ?Glyptodon sp. 13 Family Mylodontidae ?Paramylodon sp. 14 Family Megalonychidae Megalonyx cf. M. brachycephalus McAnulty 14 Family Dasypodidae Holmesina septentrionalis (Leidy) 14 Dasypus bellus (Simpson) . 14 Order Rodentia Family Sciuridae Cynomys ludovicianus (Ord) 14 Sciurus cf. S. carolinensis Gmelin 15 Sciurus niger Ord 15 Spermophilus cf. S. tridecemlineatus (Mitchill).... 15 Family Cricetidate Peromyscus gossipinus (Le Conte) 15 Peromyscus leucopus (Rafinesque) or Peromyscus maiculatus (Wagner) 15 IV Report of Investigations—No. 48

Neotoma sp 15 Sigmodon hispidus Say and Ord 15 Family Microtinae Synaptomys sp. 16 Microtus cf. M. (Pitymys) sp. 16 Ondatra cf. 0. zibethicus (Linnaeus) 16 Family Geomyidae Geomys cf. G. bursarius (Shaw) 16 Family Castoridae Castor cf. canadensis Kuhl 16 Order Carnivora Family Canidae Canis sp. (coyote) 16 Canis sp. (wolf) 17 Aenocyon sp. 17 Family Ursidae Arctodus cf. A. simus (Cope) 17 Ursus americanus Pallas 18 Family Mustelidae Procyon lotor (Linnaeus) 18 Mustela cf. M. vison (Bangs) 19 Mephitis mephitis (Schreber) 19 Family Felidae Smilodon trinitiensis Slaughter 19 Lynx cf. L. rufus (Schreber) 20 Order Proboscidea Family Mastodontidae Mammut americanus (Kerr) 20 Family Elephantidae Elephas columbi Falconer 20 Order Lagomorpha Family Leporidae

Sylvilagus cf. S. floridanus (Allen) ... 21 Lepus sp. 21 Order Artiodactyla Family Tayassuidae Platygonus sp. 21 Dama virginianus aplodon Slaughter, n. subsp. 21 Family Antilocapridae Breameryx sp. 24 Tetrameryx shuleri Lull 24 Family Camelidae Camelops huerfanensis dallasi Lull 24 ?Camelops sp. 25 Tanupolama sp. 26 Family Bovidae Bison alleni Marsh . 27 Order Perrissodactyla Family 29 Contents V

Equus midlandensis Quinn 30 Equus cf. E. fraternus (Leidy) 31 - 32 Equus sp. , conversidens (Owen) 32 Equus? quinni Slaughter, n. sp 33 Equus cf. E, caballus Linnaeus 35 Family Tapiridae Tapirus sp. 36 Class Reptilia Order Testudinata Family Emydidae Graptemys geographica (Le Sueur) . 37 Terrapene canaliculata Hay 38 Terrapene cf. T. Carolina (Linnaeus) 38 Family Testudinidae Geochelone sp. (= Testudo) 38 Order Crocodilia Family Crocodylidae Alligator mississippiensis (Daudin) 38 Order Squamata Family Colubridae ?Coluber constrictor Linnaeus 39 Class Amphibia Order Anura 39 Class Aves Order Falconiformes Family Accipitridae Aquila chrysaetos (Linnaeus) 39 Order Family Tetraonidae Tympanuchus cf. T. pallidicinctus (Ridgeway).... 39 Order Passeriformes Family Corvidae Corvus brachyrhynchos Brehm 39 Order Strigiformes Family Tytonidae Tyto alba (Scopoli) 39 Class Pisces Order Lepisosteiformes Family Lepisosteidae Lepisosteus cf. L. spatula (Lacepede) 39 Invertebrate fossils from the T-2 deposits 1 40 Phylum Arthropoda

Class Insecta , 40 Plant fossils from the T-2 deposits 40 Molluscan fauna from the Lower Shuler member, T-2 deposits 42 Wood pit pond deposit 42 VI Report of Investigations—TSo. 48

Phylum Class Family Planorbidae

Gyraulus parvus (Say) —. 42 Gyraulus circumstriatus (Tryon) .. 43 Helisoma anceps (Menke) 43 Helisoma trivolvis (Say) 43 Planorbula armigera (Say) 43 Family Lymnaeidae

Lymnaea caperata Say ... 44 Lymnaea dalli Baker ._ 44 Lymnaea humilis modicella (Say) 44 Lymnaea obrussa Say 44 Family Physidae Aplexa hypnorum (Linnaeus) 44 Physa gyrina Say 44 Family Succineidae Succinea cf. S. avara Say 45 Family Bulimulidae Bulimulus dealbatus (Say).. 45 Family Carychiidae Carychium exile H. C. Lea 45 Family Zonitidae Hawaiia minuscula (Binney) 46 Retinella cf. R. identata (Say) 46 Zonitoidesarboreus (Say) 46 Family Pupillidae Gastrocopta armifera (Say) 46 Gastrocopta contracta (Say) 46 Gastrocopta pentodon (Say) 46 Gastrocopta procera (Gould) 47 Pupoides albilabris (C. B. Adams).... 47 Family Strobilopsidae Strobilops cf. S. texasiana (Pilsbry and Ferris) 47 Family Endodontidae Helicodiscus parallelus (Say) 47 Family Helicinidae Helicina orbiculata tropica Pfeiffer 47 Family Polygyridae Mesodon thyroidus (Say) 48 Polygyra texasiana (Moricand) 48 Stenotrema leai (Binney) 48 Class Pelecypoda Family Sphaeriidae Sphaerium, striatinum (Lamarck) 48 Wood pit channel deposit 48 Family Pupillidae Pupisoma cf. P. dioscoricola (C. B. Adams) 49 Discussion of Mollusca from T-2 deposits 49 Sub-Recent comparative molluscan fauna 49 Contents VII

Fossil man in alluvial terraces of the upperTrinity River 51 T-0 deposits 51

T-l deposits - 51 Archaic Stage 51 Neo-American Stage 51

Historic Stage : 52 52 T-2 deposits ~ Lewisville site 52 Hickory Creek 52 Lagowpit 53 Seagoville site 53 Pemberton Hill site 53 Boatwright pit 54 T-3 and T-4 deposits 54 T-5 deposits : 54 uplands . 54 Comparison with other localities and local faunas 55 Trinidad, Texas, locality 55 Iron Bridge Dam locality 56 Ingleside and Berclair localities 57 Friesenhahn Cavern locality 58 Jinglebob locality 59 Rancho La Brea locality 59 Conclusions 61 Bibliography 64 Index 71 Illustrations FIGURES— 1. Index and rainfall map 2 2. Location map for Hill-Shuler localities, region of Dallas, Texas 3 3. Schematic section showing Trinity River alluvial terraces, Dallas, Texas 5 4. Typical section, Pemberton Hill—Lewisville (T-2) terrace deposits 6 5. Comparison of left ulna of Aenocyon sp. and of Canis niger 17 6. Upper jaw fragment of Arctodus cf. A. simus (Cope) 18 7. Jaw and teeth of Dama virginianus aplodon Slaughter, n. subsp., and tooth of

of Dama virginianus texanus ... 22 8. Metatarsal of ?Camelops sp. 26 9. Left mandible fragment of Tanupolama sp. 26 10. Metapodials of Bison, alleni Marsh 28 11. Teeth of Equus midlandensis Quinn 30 12. Equine metapodials 31 13. Teeth of Equus cf. E. fraternus (Leidy) 32

14. Teeth of Equus sp...... 32 15. Teeth of A sinus co nversidens (Owen) 32 16. Teeth of Equus? quinni Slaughter, n. sp. 33 17. Teeth of Equus cf. E. caballus Linnaeus 36 PLATE— I. Vertebrate fossils from T-2 terrace deposits 68 VIII Report of Investigations—No. 48 Tables

1. Measurements of sloth teeth 14 2. Comparative measurements of Aenocyon ulnas 17 3. Comparative measurements of of Arctodus 19 4. Measurements of Platygonus 21 5. Measurements of Dama virginianus 23 6. Measurements of teeth referred to Breameryx sp. 24 7. Skull and dentition measurements of C. huerfanensis dallasi. Dallas, and

C. hesternus, Rancho La Brea 25 8. Measurements of camel metapodials from T-2 deposits 26 9. Measurements of Tanupolama dentition 27 10. Measurements of B. alleni horn cores 29 11. Measurements of Dallas bison metapodials 29 12. Measurements of Hill and Lower Shuler equine metapodials 33 13. Measurements of teeth of small Equidae 36 14. Measurements of Dallas Tapirus material and observed size ranges of T. terrestris 37 The Hill-Shuler Local Faunas of the Upper Trinity River, Dallas and Denton Counties, Texas

BOB H. SLAUGHTER, WILSON W. CROOK, JR., R. K. HARRIS, D. C. ALLEN, AND MARTIN SEIFERT1

ABSTRACT The Pemberton Hill—Lewisville (T-2) probably reflect change of climate from terrace deposits of the Trinity River in more humid to more arid than is present Dallas and Denton counties are divided in the area today. The terrace deposits are into four consecutive members. Sixty- tentatively referred to the Sangamon three vertebrates, thirty-two mollusks, interglacial on the basis of two radiocar- and several insects and plants have bon dates in excess of 37,000 years and on been collected from three members and the association of the giant bison, Bison al- are discussed. Fluorine-uranium-nitrogen leni, with a typical interglacial fauna. (F-U-N) tests on fossil material from each Other Trinity River terraces and local member indicate that the entire unit is a archeology are described, and strati- single depositional sequence with a short graphic and faunal relationships with time span. During this period there were other well-known fossil localities in Texas, local and new arrivals which Oklahoma, and are suggested.

INTRODUCTION Shuler (1918, pp. 25-26, 28, 47) first Recent study shows the Pemberton Hill— alluded to the fossil-rich alluvial terrace Lewisville (T-2) terrace deposits to be late deposits along the Trinity River near Dal- Pleistocene. Because recently determined las, Texas (fig. 2) ; Lull (1921) published radiocarbon dates of the surface of the the first study of these deposits. Stovall and Beaumont clay off the Texas coast indicate McAnulty (1941) described a similar oc- the Wisconsin to have begun about 25,000 currence at Trinidad, Henderson County, years BP 2 and because two radiocarbon Texas, about 80 miles downstream from dates of fossil material from the Pember- Dallas (fig. 1). ton Hill-Lewisville (T-2) terrace deposits Most of the fossils have been discovered at Dallas indicate their age to be in excess during large-scale economic exploitation of years BP, the Pemberton Hill- for sand and gravel of the 50-foot thick 37,000 Lewisville terrace are al- Pemberton Hill-Lewisville (T-2) terrace (T-2) deposits located to the last interglacial Sanga- deposits of the Trinity River. Because so or much new material has been added to the mon. older Pemberton Hill-Lewisville fossil col- Although this paper is jointly authored, lections and because there has been a gen- the authors assume individual responsi- eral increase in knowledge over the years bibility for certain sections. These sections of ancient , plants, and climates, a and their authors are: Geology of the Ter- more accurate interpretation of the fossils race Deposits, Wilson W. Crook, Jr.; is now possible. Vertebrate Fossils from the T-2 Deposits,

¹ Dallas, Texas. Bob H. Slaughter; Molluscan Fauna from ² Abbreviation used throughout paper to indicate "years before present." the Lower Shuler Member, T-2 Deposits, 2 Report of Investigations—No. 48

FIG. 1. Index and rainfall map: (1) Area of the Hill-Shuler local faunas: (2) Trinity terrace (Boatwright pit, etc.), Trinidad, Texas; (3) Iron Bridge local fauna; (4) Ingleside local fauna; (5) Berclair local fauna; (6) Friesenhahn Cavern local fauna. Dallas occupies a position between forested areas to the east and prairie-plains to the west; up- stream and downstream migration of fauna and flora, therefore, may occur with relatively small increase or decrease in rainfall. Numbers on the rainfall lines indicate inches of precipitation annually. D. C. Allen and Martin Seifert; and Fossil ing workers for their help in making and Man in the Alluvial Terraces of the Upper preparing the collections: J. L. Connolly, Trinity River, R. K. Harris. Virgil Hudson, Theodore Saldivar, and The writers are indebted to the follow- L. P. Starrett, Jr. Pit operators Roy Moore Hill—Shuler Local Faunas 3

FIG. 2. Location map for Hill-Shuler fossil localities, region of Dallas, Texas. (1) Hickory Creek locality. (2) Lewisville site. (3) 'Coppel locality. (4) Seale pit. (5) Lagow pit. (6) Pemberton Hill. (7) Moore pit. (8) Wood pit. (9) Gifford-Hillpit. and Charles' Scott gave permission to exca- T. Lonsdale, former Director of theBureau vate certain fossiliferous zones, and this of Economic Geology; and to Dr. E. L. help is gratefully acknowledged. Lundelius, Assistant Professor of Geology, Appreciation is also expressed to Dr. A. The University of Texas, for offering sug- Byron Leonard, University of Kansas, for gestions and for permission to examine reading and criticizing the molluscan por- fossil and recent material in the collections tion of the manuscript; to the late Dr. John at Austin. 4 Report of Investigations—No. 48

The authors are most grateful to Mr. SMUMP .. Southern Methodist University Mu- Harry W. Elliott for the drawings of the seum of Paleontology, Dallas, T exas. specimens, exclusive of the equine tooth SMUP Southern Methodist University Bi- patterns. ology Department (Pleistocene mollusks). ABBREVIATIONS AND TERMS SMUR Southern Methodist University Bi- ology Department (Recent mol- The following abbreviations are used in lusks) . the text and tables of measurements: LSM Los Angeles Countv Museum, Los Angeles, California. American Museum Natural AMNH of His- TMM Texas Memorial New New York. Museum, Austin, tory, York, Texas. BEG . Bureau of Economic Geology, The University of Texas, Austin, UC .. .University of California, Los An- Texas. geles, California. Technology Wagner CITVPC .. California Institute of WFIS Free Institute of Science. Vertebrate Paleontology Collec- Philadelphia, Pennsylvania. Pasadena, tion, California. BP Years before present DPSC , .Dallas Prehistorical Society Collec- tions. Deposited jointly at South- ern Methodist University, Dallas, Sand and gravel size terms are accord- Texas (bulk of study collection) ing to the grain size classification of and at Dallas Museum of Natural History (display specimens). Wentworth. GEOLOGY OF THE TERRACE DEPOSITS

Alluvial terrace deposits of the Trinity name indicates the Trinity River terrace River are well defined, easily recognized, deposits at Dallas; the second indicates the and remarkably consistent in content and comparable terrace deposits upstream on sequence from 40 or more miles upstream Elm Fork. from Dallas to at least 80 miles down- The following Trinity River terrace de- stream. The writers have investigated and posits, in order from highest to lowest, studied these terrace deposits from Tar- occur in the Dallas area (fig. 3) : rant and Denton counties, west and north- west of Dallas, to Henderson County, southeast of Dallas. Shuler (1935) first described the Trin- Cretaceous uplands 195-425 terrace at T-5, Buckner Home—Hackberry Creek 165 ity deposits Dallas; Pattillo T-4, Love Field—Bethel 120 (1940) described those of the Elm Fork T-3, Travis School—Farmers Branch.... 90 of the Trinity River near Carrollton, traced T-2, Pemberton Hill—Lewisville 70 T-l, Union Terminal—Carrollton 50 this sequence to its junction with the Trin- T-0. Flood nlain (annrox.l 20 ity River sequence described by Shuler, and established a lithologic correlation The Hill-Shuler local faunas described between the two. The present writers have in this paper are from Pemberton Hill- confirmed this correlation and have fur- Lewisville (T-2) terrace deposits (fig. 3)' ther refined the sequence and definition of in Dallas, Denton, and Tarrant counties the terrace deposits, including the recog- (fig. 2); these terraces lie along the Trin- nition of two additional terraces. The cur- ity River and its tributaries, Elm Fork, rent binomial nomenclature, used locally, West Fork, and Hickory Creek. Uni- emphasizes the relation of the Trinity and formity of age of the terrace deposits and Elm Fork type localities for the several the fossil bone within them was proven by terrace deposits (Crook and Harris, 1957, fluorine-uranium-nitrogen (F-U-N) tests p. 37). In the binomial system, the first on bone samples submitted to the Univer-

FIG. 3. Schematic section showing Trinity River alluvial terraces, Dallas, Texas. Numbers indicate elevation above present stream grade of top of present terrace surface. 6 Report of Investigations—No. 48 sity of London (Oakley and Howells, deposits and produced the cut into which 1961). the T-2 fill was deposited. There were only minor differences in fluorine, uranium, and nitrogen content of bone from four different localities as much as 40 miles apart; four species of animals were represented and bone was analyzed from each of the three lowest members of the Pemberton Hill-Lewisville (T-2) ter- race fill. Uniformity in age of the three lowest members of the terrace deposits both geographically and stratigraphically is thus demonstrated. Because the differ- ence in age between the three lowest mem- bers of the Pemberton Hill-Lewisville (T-2) fill is so small, they must have been deposited within a brief period despite the pronounced erosional break that occurs in places between members and despite the change from gravel to sand, to clay, and so on. Individual members of the Pemberton Hill-Lewisville (T-2) terrace fill vary in thickness from exposure to exposure, de- pending upon the amount of erosion at a given locality during any of the erosion in- tervals separating the different members. This variation in thickness is typical of the meandering cut-and-fill deposits of low- gradient coastal plain streams. Despite its distance of nearly 250 miles from the Guli Coast, the gradient of the Trinity River at Dallas is only approximately 0.7 foot per mile. Hereinafter, the terms "T-2 deposits" and "T-2 fill" designate the Pemberton Hill-Lewisville (T-2) terrace deposits. Similar terms are used to designate other terrace deposits of the Trinity River system. A typical section of T-2 deposits in the Dallas area is as follows, beginning at the

base (fig. 4) : (1) At Dallas bedrock is Cretaceous, either Eagle Ford shale or Austin chalk. (2) A marked angular unconformity separates the Cretaceous and Quaternary deposits. Its last was the valley refinement FIG. 4. Typical section, Pemberton Hill— trenching which established relief on T-3 Lewisville (T-2) terrace deposits. Hill—Shuler Local Faunas 7

(3) A deposit of gravel with much in- formity, the formation of which, in places cluded sand overlies the bedrock and is probably close to the stream channel, re- called locally the Hill member. The gravels sulted in the removal of all of the Lower range in size from very large pebbles at Shuler sands. the base to small pebbles at the top. Thick- (7) Yellow sandy clay, often massive ness varies from section to section, from a and indurated, comprises the next higher minimum of about 3 feet to an observed member, the Upper Shuler member, which maximum of 20 feet, and averages about is 3 to 20 feet thick and averages about 12 10 to 12 feet. In many exposures, the top- feet. Although this fill is relatively homo- most foot or so is cemented by iron oxide geneous, it does contain five to seven depo- to form a brownish, rust-colored con- sitional units which vary in thickness. The glomerate. Fossil bone is heavily mineral- upper layer of each of these units was a ized, is usually stream abraded, and has surface long enough for minor erosional small pebbles and sand grains firmly ce- features to develop thereon and for a sub- mented to it. surface zone of caliche to form; it is the (4) A minor disconformity separates caliche that delineates each of these minor the Hill member from the next higher erosional surfaces. The caliche is com- member. posed of popcorn-like nodules and concre- (5) Clean, finely laminated yellow tions which are progressively larger and sands comprise the Lower Shuler member. more abundant upward. A number of cal- The laminae are approximately 2 mm careous root fillings, caliche-filled croto- thick. The yellow color is limonite stain, vinas (presumably prairie-dog holes and in places there are narrow bands of judging from the size of the holes manganese stain. Many calcareous root and the presence of fossilized remains of molds occur throughout at various levels these animals), and smaller, filled cray- which must have been surfaces at the time fish holes identical to filled holes in pres- the roots were emplaced. Layers of cal- ent creek beds occur at each of the succes- careously cemented sand as much as 6 sive unit surfaces; there are also vague, inches thick also occur at certain localities. whitish, limy zones. Fossil bones are min- Occasional lenses of small pebbles inter- eralized, primarily by calcium carbonate, spersed with layers of laminated sand oc- and are usually heavily encrusted by ca- cur near the base. liche. They show no evidence of stream Average thickness of the member is ap- abrasion and are commonly cracked and are proximately 5 feet, although it may be as fragmentary, although the fragments much as 10 feet. At many localities suc- recemented, usually by calcium carbonate. ceeding erosion has completely removed (8) Another minor disconformity oc- the yellow laminated sand; here, higher curs above the yellow sandy clays of the members rest directly upon the eroded Upper Shuler member. In places, 3- to 8- Hill gravels below. Included fossil bones foot deep stream channels have been cut are well mineralized, unabraded, and com- in the Upper Shuler, then filled by the monly complete; a few small logs also overlying member. occur. Some bones and logs are encrusted (9) The Richards, locally derived gray with iron oxide. In places near the top of alluvium, is the uppermost member of the 5 the member a thin (5-mm thick) layer of T-2 depositional sequence. It averages ironstone has formed beneath the buried feet thick, excluding the deeper channel erosion surface which separates the Lower fills in the underlying Upper Shuler mem- Shuler from the Upper Shuler. ber; in places the Richards has been al- (6) The yellow sands of the Lower Shu- most totally removed by subsequent ero- ler member are separated from the over- sional periods. The Richards has remained lying Upper Shuler by a minor discon- the exposed surface of the T-2 fill essen- 8 Report of 4B tially ever since it was laid down as the trenches to glacial maxima of low sea level. uppermost member of the T-2 deposits. Thus, the T-2 fill is related, depending up- The member contains numerous small cal- on recent interpretation and revisions, careous nodules and concretions through- either to the last interglacial or to a major out, but no fossils have yet been found. It interstadial within the last glacial. is possible that leaching, during the long Two radiocarbon dates from the lower period that the Richards has comprised part of the Upper Shuler member at Lewis- the T-2 surface, may have removed fossil ville, near the junction of Hickory Creek bone and caused the small size of the and Elm Fork in Denton County, are "more caliche nodules. than 37,000 years BP" (Brannon et al., (10) A major episode of valley cutting 1957, p. 149). created the T-2 relief. Numerous radiocarbon assays from gla- (11) Another depositional cycle par- cial tills in the Midwest have established tially filled the valleys cut in the T-2 de- that the standard Wisconsin, or Fourth posits; this fill is the T-l deposit. Thick- Glaciation as it has been previously con- ness of the T-l fill does not equal that of sidered, encompasses the period roughly the T-2 deposits (50 feet above present from 25,000 to 10,000 BP. These dates are stream grade as compared to 70 feet). The substantiated by Magnolia Petroleum Com- T-l deposits, red Albritton clay and over- pany (a division of Socony Mobil) which lying gray Pattillo sand, do form a veneer has obtained radiocarbon dates of samples in places on the eroded slopes of the T-2 from submerged offshore deposits in the fill. Especially near the present valley, Gulf of Mexico that can be correlated with typical red sandy clay and gray sand mem- sea-level fluctuations. One sample from the bers of the T-l fill commonly rest directly surface of the Beaumont clay, a marine de- on truncated sections of T-2 fill. posit presumably of the last interglacial, (12) Another major valley cutting es- was from the subaerially weathered zone tablished the present relief on the T-l de- exposed by the last major regression of posits. This was followed by deposition of the sea—assumed to have occurred during the present flood plain (T-0) deposits to a the last glacial or Wisconsin. Certainly the feet above height of some 20 present sea had to be lowered considerably for this stream grade (fig. 3). surface to be exposed. The oldest sample (13) The channels incised present were measured some 28,000 BP (Bray and Nel- to bedrock. son, 1956, p. 173). On the other hand, the radiocarbon dates of offshore borings by COMPARISONS AND INTERPRETATIONS Magnolia, from the base of Recent marine The sequence of alluvial terrace deposits deposits laid down during the transgression of the Trinity River has been tentatively related to modern high sea stand, range correlated by Crook and Harris (1957, p. from 9,000 to 12,000 BP. 53; 1958, p. 237) with those of the neigh- In Germany, recent radiocarbon meas- boring reaches of the Red, Sulphur, Sa- urements have dated the Masurian Inter- and Leon Rivers. bine, Brazos, stadial deposits at Rixdorf, near Berlin, as Crook and Harris' conclusions agree es- 28,000 to 42.000 BP (DeVries, 1958, p. sentially those of Fisk (1938, pp. with 15), presumably that interval between the 169-172; 1940, p. 58; and revised by early and late phases of the Wurm or Doering, 1956, pp. 1832-1837) concern- Fourth Glacial. In more southerly climatic ing Fisk's investigations of the Red River such as at this non-glacial in northern Louisiana. areas, Dallas, Briefly, these investigators allocated the period may have begun somewhat earlier perhaps terrace fills to non-glacial periods of high and lasted somewhat later, cover- sea stand and allocated the cuts and ing the period 25,000 to 45,000 BP. Hill-Shuler Local Faunas 9

There is also some recent evidence from Exposures near Iron Bridge Dam (fig. the Midwest that there an early Wisconsin 1,no. 3) on the upper Sabine River in Yan predates standard Wisconsin, but this is not Zandt and Hunt counties east of Dallas yet widely accepted. exhibit a sequence of units almost identi- Based on radiocarbon dates the T-2 fill cal to those at Dallas and Trinidad (Peter- must pertain to a non-glacial period, either son and Slaughter (MS). T-0 deposits are major instadial in the last glacial or the composed of locally derived alluvium, with last interglacial. Climatic implications one historic Indian village site upon its based upon the fossil fauna and flora defi- surface—the Tawakoni Village of 1759- nitely allow this interpretation; it is also 1767. The T-l fill is the same red clay—- in agreement with conclusions of Fisk gray sand sequence as at Dallas, contain- (1938, pp. 169-172; 1940, p. 58) and ing in-place Archaic archeological arti- Doering (1956, pp. 1832-1837) concern- facts also directly related to those of the ing correlation of river terraces and coastal Dallas area. The T-2 deposits of the two deposits with glacial and non-glacial areas exhibit the same general lithologic periods. sequence, including the most recent caliche Further confirmation is supplied by the zone, and extinct faunal species are also next-younger (or T-l) fill, which should similar ( Tetrameryx shuleri is known only in theory be related to the Recent post- from these two localities). Higher terraces, glacial period characterized by high sea not yet intensively studied, are capped, as stand. One unpublished radiocarbon date at Dallas, by a remnant of the typical from near the mid-point of T-l deposits Dallas area T-5 cobble fields of quartzite just southeast of Dallas is approximately and petrified wood. Similarities of the 6,000 BP. terrace deposits and faunas of the Iron Archeological materials of the Archaic Bridge and Dallas areas are striking. Stage, in place and undisturbed, are com- Other Texas late Pleistocene areas pale- mon throughout the T-l deposits. The ontologically, geologically, and archeologi- Archaic Stage in the United States, based cally described in the literature—the Ingle- on radiocarbon dates, ranges from 4,000 side pond fill, theBerclair terrace deposits, to 10,000 BP. As yet only a modern verte- and the Friesenhahn Cavern deposits of brate fauna has been recovered from the south Texas (fig. 1, nos. 4, 5, 6) are related T-l deposits on the Trinity River system, in varying degrees. Berclair terrace depos- and the snail assemblage of the de- T-l its (Sellards, 1940), Bee County, which posits appears to be modern in contrast to have been directly correlated with the that of the T-2 deposits. The present flood plain (T-0) deposits Beaumont clay shore line and its non- have yielded, appropriately, inclusions of contradictory fossils, seem to be a direct late pottery-and-agriculture Indians, scant correlative of the T-2 terrace deposits at evidences of possible Spanish explorations, Dallas. Ingleside pond deposits in San a surficial capping of Anglo-American Patricio County (studied by the writers) materials, plus a totally modern fauna. It have a fauna similar to the fauna of the is obviously but yesterday in origin. T-2 deposits at Dallas, but the stratigraphic The writers have visited and studied the position of the Ingleside atop the Beau- downstream terrace evidence near Trini- mont clay (implying considerable regres- dad and Malakoff, Henderson County (fig. sion of sea level) suggests the Ingleside 1, no. 2). The Trinity terrace deposits during last inter- described by Stovall and McAnulty (1941) fill was deposited late, and early last glacial are directly correlated with those of the glacial interstadial, than T-2 at Dallas on the basis of the elevation time. This is apparently slightly later above stream grade, lithology, and the in- the T-2 deposits at Dallas. Friesenhahn cluded fauna. Cavern local fauna (Bexar County), for 10 Report of Investigations—No. 48 various reasons, may be somewhat later, the Texas and Pacific Railroad. The site perhaps early-to-late last glacial. and general area have been visited by the A last glacial fauna of pluvial deposits, writers, and their observations, added to such as the late deposits of west Texas, has Shuler's (1923) study, definitely correlate not been found in the Dallas area, perhaps these deposits with other T-2 deposits along because of the excessive local downcutting the Trinity River. One side of the original occurring at that time. pit displayed an almost complete section of In summary, the fauna of the T-2 depos- the terrace deposit while the opposite side its along the Trinity River and its tribu- sloped to the river. The Richards and part taries near Dallas probably pertains to the of the Upper Shuler were removed and last interglacial or a major interstadial of red Albritton clay of the T-l terrace fill was the last glacial. These deposits and their deposited on the cut surface. faunas, far south of the actively glaciated Gifford-Hill pit—ln 1961, this pit (fig. regions, are directly related to the Gulf 2, no. 9), exposed only clay, sand, and Coastal Plain alluvial deposits which can gravel of the lowest terrace section. In be correlated over wide areas with fluctuat- 1955, however, there was an island rem- ing sea levels of the late Quaternary. There nant of T-2 deposits near the middle of the is little doubt, however, that these obvious Gifford-Hill properties. Hill-Shuler fossils fluctuations of sea level can be directly from the Gifford-Hill pit were collected correlated with major cycles of glaciation from this island remnant, which had the and deglaciation, and because of this the same elevation and lithology as better pre- abundant fossils in the terrace deposits are served exposures of the T-2 deposits. This most useful and instructive. pit is just east of Beckett Road, 0.6 mile south of Belt Line Railroad and 3 miles HILL-SHULER LOCAL FAUNAS northeast of Seagoville, Dallas County. Pemberton Hill—Excavations for sand Fossils comprising the Hill-Shuler local and gravel have almost completely de- faunas were collected from commercial ex- stroyed Pemberton Hill, which stood on cavations and one natural of the exposure the northeast corner of Pemberton Hill T-2 fill along Trinity River from southern Road and Loop 12 in the Dallas City limits Denton County to southern Dallas County, (fig. 2, no. 6). Early excavations into this Texas. These specimens, almost without portion of the T-2 fill exposed a complete exception, were in situ and therefore can section, with all members present in maxi- be assigned to one of three members of the mum thickness. The Pemberton Hill loca- T-2 fill. Specimens found without an over- tion will be covered with a housing project lying soil profile have been omitted to within a year. This site is an excellent ex- avoid the possibility of mixed faunas. With ample of the amount of erosion that has the exception of a small collection made in taken place between the deposition of the the early twenties by the late Dr. Ellis W. T-2 and T-l deposits; road cuts just to the Shuler, all material was collected by the south and west and an excavation immedi- writers and their colleagues. ately to the east display eroded sections in which the Richards and Upper Shuler LOCATION AND DISCUSSION OF SITES members are missing. —Moore pit (fig. no. 7) Lagow pit—ln 1920, Shuler collected Moore pit 2, side of Loop 12 just 400 fossils from the Lagow sand and gravel pit is on the south southeast of Pemberton Hill. The pit (fig. 2, no, 5) in the T-2 fill. The pit, lo- yards all members cated justeast of Fitzhugh Street and south walls opposite theriver display of East Grand Avenue in Dallas, Texas, of the T-2 deposits except the Richards. On has since been refilled, and the site is now the river side, however, both the Richards occupied by the locomotive roundhouse of member and the Upper Shuler member Hill—Shuler Local Faunas 11

have been removed along with some of the Road 2181. This terrace deposit of Hickory Lower Shuler member, and there is a cover Creek has been traced 5 miles downstream of T-l deposits. There are several ancient where it merges with T-2 terrace deposits channel fills within the Lower Shuler of of Trinity River. The Hickory terrace se- this pit, and these have produced most of quence has a slightly different composition the fossils. because of a slightly more westerly source; Wood pit—Wood pit (fig. 2, no. 8), the Hill gravel equivalent, Upper Shuler, 700 yards downstream from Moore pit, is and Richards are well preserved. Lower at the south end of Deepwood Street. Rapid Shuler is either missing or represented by expansion of excavation has joined Wood a 4-foot clay bed above the gravel; this and Milton pits making it practical to com- clay bed is more sandy than the Upper bine Milton and Wood pit collections. As Shuler clay and lacks its massive caliche. in Moore pit, Upper Shuler is present only Lewisville site—A barrow pit (fig. 2, in the wall opposite the river. no. 2), from which material was excavated Coppel locality—This site (fig. 2, n0.3) to build the Garza-Little Elm Dam, is im- is on Grapevine Creek where it has cut into mediately northeast of the west end of the a terrace of Denton Creek near where dam. Fossils from this pit were discovered Denton Creek joins the Trinity River. The by Theodore White and later described by site is on the Good farm, 100 yards up- Crook and Harris (1957, p. 54). All fossils stream from Coppel Road bridge over recovered from this site are from the Up- Grapevine Creek, half a mile southeast of per Shuler member, although a complete the town of Coppel in Dallas County. Al- section is present. Lewisville site was des- though this deposit was too far upstream ignated as the Elm Fork type locality of from the main river to be affected by back- the T-2 deposits; this locality has since wash from the river and although the al- been inundated by waters of the reser- luvial material is all locally derived, the voir. sequence is the same as that of the Trinity Seale pit —This pit, an excavation for River—gravel capped by sand which is, in sand, is on property of Mr. W. Q. Seale and turn, capped by clay charged with caliche is located at the south end of County Line nodules. The fossils and the 70-foot eleva- Road and the west end of Oakdale Road, tion of the location above Denton Creek approximately 4 miles north of Grand indicate direct correlation a with T-2 de- Prairie, Texas (fig. 2, no. 4). Unlike the posits of the Trinity River. other localities, the Seale pit is on the T-2 Hickory Creek locality—Fossils were terrace of the West Fork of Trinity River, collected from this site (fig. 2, no. 1) its junction with the main river. subsequent to erosion of a large excavation near —Because of the close made to obtain material to build the dump Haymarket pit of the Santa Fe Railroad, which crosses proximity of this pit to the Wood-Milton Hickory Hill Road 150 yards west of the pit complex and also because only a single site. The locality, the Ernest D. Calvert specimen was recovered from it, the Hay- farm, is on Hickory Hill Road, 1.4 miles market pit is considered part of the Wood west of the junction with Farm-to-Market pit local fauna. VERTEBRATE FOSSILS FROM THE T-2 DEPOSITS The specimens collected from the T-2 deposits and described in the following pages are listed below so that their member and site location may be readily at hand. Chart showing distribution of vertebrate fossils from T-2 deposits. Locality designations: C, Coppel locality; G, Gifford-Hill pit; Ha, Haymarket pit; H, Hickory Creek locality; La, Lagow pit; Le, Lewisville site; M, Moore pit; S, Seale pit; P, Pemberton Hill; W, Wood pit. Member Hill Lower Shuler Upper Shuler Mammalia Marsupialia M Insectivora Le Edentata ?Glyptodon sp. Le ?Paramylodon sp La Megalonyx cf. M. brachycephalus McAnulty G M, W Holmesina septentrionalis (Leidy) M, C Dasypus bellus (Simpson) M H Rodentia Cynomys ludovicianus (Ord) Le, H Sciurus cf. S. carolinensis Gmelin .... M, C Sciurus niger Ord M Le Spermophilus cf. S. tridecemlineatus (Mitchell) C Peromyscus gossipinus (Le Conte) Le Peromyscus leucopus (Rafinesque) or P. maniculatus (Wagner) M, C Le, H Neotoma sp. M, C Le, H Sigmodon hispidus Say and Ord M, C H

Synaptomys sp. ... C H Microtus cf. M. (Pitymys) sp. M, C Le, H H Geomys cf. G. bursarius (Shaw) M, C Le, H Castor cf. C. canadensis Kuhl M H Carnivora Canis sp: (coyote) ... M Le Le P, M Arctodus cf. A. simus (Cope) S JJrsus americanus Pallas Le Procyon lotor (Linnaeus) . M Le, H M Mephitis mephitis (Schreber) M Le M,La M, W,P M Proboscidea Mammut americanus (Kerr) ...W, G M H Elephas columbi Falconer .... W, G, S, P W, M, C, P, Le, H Langomorpha P, H, S Sylvilagus cf. S. floridanus (Allen) M, C Le, H Lepus sp. W Le Artiodactyla Platygonus sp. Le *Dama virginianus aplodon, n. subsp. M, W *Dama sp. ...La, W M, W, C, P Le, H P, c La, M La, P ?Camelops sp. W, G, M M, W, C, H Le M, La, W La, G, M, S Bison sp — — W, G C, M, W, P Le, H

* Since Lhis paper was written and in type the International Congress of Zoological Nomenclature recommended the use of Odocoileus in place of Dama. Hill-Shuler Local Faunas 13

Member Perrisod'actyla Hill Lower Shuler Upper Shuler Equus midlandensis Quinn W M, W Equus cf. E. fraternus (Leidy) — W M, W Equus sp. W, M, P C, M, P, W

Asinus conversidens (Owen) .... ' W, M M, W, C Equus? quinni Slaughter, n. sp. W,M W,M Equus cf. E. caballus Linnaeus Le, P Tapirus sp. W, M M, W Reptilia Testudinata Graptemys geographica (Le Sueur) M Terrapene canaliculata Hay Le, H Terrupene cf. T. Carolina (Linnaeus) W, M, C Geochelone sp W,G W, M, C, P Le Alligator mississippiensis (Daudin) C M Squamata ?Coluber constrictor Linnaeus Le Amphibia Anura M, C Le, H Aves Falconiformes Aquila chrysaetos (Linnaeus) W Galliformes Tympanuchus cf. T. pallidicinctus (Ridgeway).... c Passeriformes Corvus brachyrhynchos Brehm . ... M Stringiformes Tyto alba (Scopoli) ...... c Pisces Lepisosteiformes Lepisosteus cf. L. spatula (Lacepede) M SYSTEMATIC PALEONTOLOGY Most of the specimens upon which this cept for its slightly smaller size, which is study is based were originally in the col- certainly within individual size range. lections of the Dallas Prehistorical Society. Stovall and McAnulty (1941, p. 242) These have now been transferred to South- state that their record of the opossum in ern Methodist University Museum of Pale- the Trinity terrace at Trinidad, just 80 ontology, and most of the specimen num- miles downstream from Dallas, is the sec- bers reflect that change. Specimens on loan ond Pleistocene report of the form. If this to other universities, however, still carry is so, the Dallas occurrence is apparently the number of the Dallas Prehistorical the third. Society; both numbers are given for speci- Order INSECTIVORA mens that have been previously reported Family TALPIDAE and described the literature. in SCALOPUS AQUATICUS (Linnaeus) Phylum VERTEBRATA Referred specimen.—Humerus (DPSC Upper Lewisville Class MAMMALIA 1060); Shuler, site. Discussion. —This specimen was identi- Order MARSUPIALIA fied in 1956 by Dr. Claude Hibbard when Family DIDELPHIDAE he examined material from the Lewisville DIDELPHIS cf. D. MARSUPIALIS Linnaeus site. Referred specimen.—Right M 3 (SMU- Order Edentata Family MP 60152) ; Lower Shuler, Moore pit. GLYPTODONTIDAE Discussion. —This is a single tooth; ref- ?GLYPTODON sp. erence identification is provisional. It is Theodore E. White, a salvage paleon- identical to several Recent specimens ex- tologist for the River Basin Survey Party 14 Report of Investigations—No. 48 which conducted operations in the Garza- was collected, and they are convinced that Little Elm reservoir, reported (personal the deposits of this type locality are a Hen- communication to W. W. Crook, Jr., 1955) derson County equivalent of the T-2 de- "Glyptodon scutes'', from the Upper Shuler posits at Dallas. at the Lewisville site. It is possible to iden- Family DASYPODIDAE tify scutes of various members of Edentata (Leidy) even specifically, but apparently these par- HOLMESINA SEPTENTRIONALIS ticular specimens, now in the U. S. National Referred specimens.—Cheek tooth (SM- Museum, have not been studied; Glyptodon UMP 60068) ; Lower Shuler, Moore pit. scutes scutes do not occur in Dallas site collections Buckler (SMUMP 60092) ; Lower available to the writers. Shuler, Moore pit. Casque scute (SMU- MP 60267) ; Lower Shuler, Coppel locality. Family MYLODONTIDAE Discussion.—The cheek tooth (SMUMP ?PARAMYLODON sp. 60068) was compared with the dentition Shuler (1918, p. 26) stated that dermal of a mandible of H. septentrionalis, in the scutes of ground sloths were collected at collection at the Bureau of Economic Ge- the Lagow sand pit. These specimens were ology, The University of Texas, which not used in this study, but because Paramy- came from a terrace of the Brazos River. lodon is the only sloth that had such The same specimen was sent to Mr. Walter scutes, the genus is added provisionally to Auffenburg, University of , for the Hill-Shuler faunal list. comparison with material from that State. The Dallas specimen is somewhat smaller Family MEGALONYCHIDAE than both. Auffenburg felt, however, that MEGALONYX cf. M. BRACHYCEPHALUS McAnulty size is not an important specific character- Referred specimens.—Second upper istic of this . cheek tooth (SMUMP 60046) ; Lower The buckler scutes are similar to scutes Shuler, Moore pit. Third upper cheek tooth found at other localities in Texas and Flor-

(SMUMP 60058) ; Lower Shuler, Moore ida. They are 8 to 10 mm in thickness and pit. Fused first and second phalange of the 30 to 40 mm in diameter. third pes digit (SMUMP 60023); Hill The casque scute (SMUMP 60267) is member, Gifford-Hill pit. very thin and kite-shaped; dimensions are Discussion.—lsolated sloth teeth are not 5 X 37 mm. identified but these easily specifically, spec- DASYPUS BELLUS (Simpson) imens are almost identical in size and form one hundred to M. brachycephalus, which is smaller than Referred specimen.—Over buckler, movable ring, and leg scutes plus other species of the genus. It is on this three thoracic vertebrae belonging to a basis that the Dallas specimens are referred single (SMUMP DP- provisionally (table 1). individual 60297; SC 705). Upper Shuler, Hickory Creek TABLE 1. Measurements of sloth teeth (in mm). locality. Upper M. brachy- cheek teeth cephalus Discussion.—This individual has been (SMUMP (SMUMP (BEG described by Slaughter (1959) and is the 60046) 60058) 30907-60) first record of this species in Texas. Anteroposterior diam- eter of 2nd tooth 15 16 Order RODENTIA Transverse diameter of 2nd tooth 2'2.5 23 Family SCIURIDAE Antero-posterior diam- CYNOMYS LUDOVICIANUS (Ord) eter of 3rd tooth 16.5 15.3 Transverse diameter of Referred specimens.—Six lower jaws 3rd tooth 23 24 (SMUMP 60668) ; Upper Shuler, Lewis-

The writers visited the pit near Trinidad, ville site. Lower jaw (SMUMP 60296) ; Texas, where the holotype of this species Upper Shuler, Hickory Creek locality. Hill—Shuler Local Faunas 15

Discussion.—In addition to the six jaws Family CRICETIDAE (SMUMP 60668), post-cranial elements PEROMYSCUS GOSSIPINUS (Le Conte) have been collected from the Lewisville Referred specimens.-—Lower jaws (DP- site and Hickory Creek and Coppel locali- SC 1013, 1069); Upper Shuler, Lewisville abundant ties. Although fairly prairie-dog site. occurs the not material in Upper Shuler, Discussion.—The presence of this spe- a single specimen has been recovered from cies in the Upper Shuler member is a para- the Lower Shuler members. This Hill or is dox. Because it is a species whose normal an indication of more arid conditions dur- habitat is somewhat farther east, it does Upper Shuler than ing deposition during not conform with the previously suggested chat of the Hill and Lower Shuler members. climatefor this member. There no reliable records of the are oc- The climate of earliest Upper Shuler currence of dogs within a hundred prairie times may have been much like that of miles of Dallas during Recent The time. Lower Shuler time, and perhaps it was not Upper Shuler specimens do not to appear until well into Upper Shuler time that arid be intrusive because they were collected conditions began to affect the fauna. from a depth of over 20 feet, and several encrusted with PEROMYSCUS LEUCOPUS (Rafinesque) were caliche. or PEROMYSCUS MANICULATUS (Wagner) SCIURUS cf. S. CAROLINENSIS Gmelin specimens.—Lower jaw (DP- Referred specimens.—Lower jaw (SM- Referred SC 1070) ; Upper Shuler, Lewisville site. UMP 60117); Lower Shuler, Moore pit. Tooth (provisional reference) (SMUMP Ml (SMUMP 60283) ; Lower Shuler, Cop- 60287) ; Lower Shuler, Coppel locality. pel locality. Discussion.—Both species are present in Discussion.—These specimens were com- the area today. pared with S. carolinensis and S. Niger and the size of both matches that of S. NEOTOMA sp. carolinensis more closely. Gray squirrels Referred specimens.—lsolated teeth are not present in the area today but are (SMUMP 60122) of this genus are abun- found east of Dallas in the pine-hardwood dant in the Lower Shuler at the Moore pit forests. This is an indication of the exist- and Coppel locality and in the Upper ence in the Dallas area of a humid climate, Shuler at the Hickory Creek locality and with accompanying fauna and flora, during Lewisville site. Lower Shuler deposition. Discussion.—N. floridanus occurs in the Dallas area today. SCIURUS NIGER Ord Referred specimen.—Lower jaw (DPSC SIGMODON HISPIDUS Say and Ord —Lower jaw frag- 1061) ; Upper Shuler, Lewisville site. Referred specimens. Discussion. —This species is abundant ment containing M 3 (SMUMP 60150) ; today both east and west of Dallas. Lower Shuler, Moore pit. Maxilla fragment containing (SMUMP 60158) ; Lower SPERMOPHILUS cf. S. TRIDECEMLINEATUS M 3 (Mitchill) Shuler, Moore pit. M 3 (SMUMP 60307); Referred specimens.-—Partial jaw (DP- Upper Shuler, Hickory Creek locality. SC 1034); Upper Shuler, Lewisville site. Discussion.—Although S. hispidus rang- Single tooth (SMUMP 60286); Lower es throughout most of southern United Shuler, Coppel locality. States today, Sigmodon s center of distri- Discussion. —Dallas is near the eastern bution is definitely south of the United edge of the range of this species. Because States border. Although currently abun- the species ranges west and slightly east, dant within its range, this species has not it cannot be used to infer climate. been previously collected from any of the 16 Report of Investigations—-No. 48 well known Sangamon localities. Its occur- because its present range includes most of rence in the Upper Shuler at Dallas may the arid and semi-arid areas of the South- indicate that at least this member of the west. A steady supply of ground water is T-2 terrace sequence is younger than the all that is necessary for its existence. Kansas and Oklahoma local faunas refer- Family GEOMYIDAE red to the Sangamon. GEOMYS cf. G. BURSARIUS (Shaw) Family MICROTINAE Referred specimens.—lsolated teeth SYNAPTOMYS sp. (SMUMP 60307) ; Upper Shuler, Hickory Referred specimen.—Ml (SMUMP Creek locality. 60306); Upper Shuler, Hickory Creek lo- Discussion. —These cheek teeth were cality. compared with Recent specimens of this Discussion. —Fossil Synaptomys from species and no important differences were Pleistocene deposits of approximately the seen. Specific identification is provisional, same age as those of the T-2 are usually re- however, because of the fragmentary na- ferred to S. australis Simpson, whose ture of the specimens. larger size distinguishes it from the living Family CASTORIDAE form 5. cooperi. The small size (2.6 mm) CASTOR cf. C. CANADENSIS Kulil of the Dallas specimen makes it impossible to refer it with confidence to the fossil Referred specimens.—lncisor (SMU- species, although the Dallas form could MP 60567); Lower Shuler, Moore pit. represent a young individual. On the other Molar (SMUMP 60294) ; Upper Shuler, hand, if it belongs to the Recent species, Hickory Creek locality. there is the subtle suggestion that Upper Discussion.—The incisor is slightly Shuler time was getting cooler as well as larger thanRecent specimens with which it drier. has been compared; it is, no doubt, within the size range of the modern species. MICROTUS cf. M. (PITYMYS) sp. The molar was collected from the same Referred specimens.-—lsolated teeth zone in the basal Upper Shuler at Hickory (SMUMP 60277); Lower Shuler at the Creek that produced the muskrat and other Coppel locality and Moore pit; Upper evidence of marsh conditions. Shuler atHickory Creek locality and Lewis- ville site. Order CARNIVORA Discussion.—Single teeth are difficult to Family CANIDAE identify, even subgenerically, but M. CANIS sp. (coyote) (Pitymys) pinetorum (Le Conte) is pres- Referred specimen.—Left lower jaw ent today in east Texas. There is one re- with complete dentition except incisors port of M. (Pitymys) sp. near Kerrville, (SMUMP 60315) ; Upper Shuler, Lewis- Texas, but this subspecies is considered ville site. primarily a deep east Texas type. Discussion. —Although this specimen is ONDATRA cf. O. ZIBETHICUS (Linnaeus) coyote size, comparison with numerous Re- cent specimens of C. latrans and C. Niger specimen.—Ml (SMUMP Referred indicate that it is distinct from these forms. 60556); Upper Shuler (?), Hickory Creek wolf-like locality. Because of certain characteristics, such as the presence of an additional pos- Discussion.—This specimen was col- lected in the lowest portion of the Upper terior cusp on P2, further study and addi- Shuler where that member is very sandy tional material may prove that this speci- and almost free of caliche nodules. Al- men (SMUMP 60315) from the Lewisville though indicative of a marsh condition, site is a new type. It is also possible that O. zibethicus is not a good climate marker the specimen will prove to belong to C. Hill—Shuler Local Faunas 17 caneloensis Skinner (1942),, the type of which is a skull and therefore not directly comparable. C. caneloensis also is very wolf-like (Skinner, 1942, p. 164).

CANIS sp. (wolf) Referred specimen. Claw (DPSC 1001) ; Upper Shuler, Lewisville site. Discussion.—This claw seems to be slightly larger than claws of C. latrans or C. Niger but is hardly large enough to be considered Aenocyon. It is possible that it belonged to a small individual of C. lupus.

AENOCYON sp. Referred specimens.—Left ulna (SMU- MP 60037) ; Lower Shuler, Pemberton Hill. Phalanx (SMUMP 60103); Lower Shuler, Moore pit. Discussion. —Although these specimens are not adequate for specific identification, they were compared with Recent specimens of Canis latrans and C. Niger (fig. 5) and found to be almost one-third longer and twice as robust. The referred specimens are, however, comparable in size with FIG. 5. Comparison of left ulna of (a) Aeno- cyon sp. (SMUMP 60037) and of (bj Canis three ulnas of Aenocyon , one from Califor- niger. x% nia and two from Texas. The ulna from California, with which the Dallas specimen P4 (BEG 40442—1); Lower Shuler, Seale was compared, is referred to A. dirus. A. pit. ayersi is more commonly known from This specimen, collected several years Texas, but only the tooth row is diagnostic. ago by W. Q. Seale and submitted to Dr. Table 2 compares measurements of several J. H. Quinn, formerly with the Bureau of Aenocyon ulnas. Economic Geology, was made available to the authors by Quinn. Family URSIDAE Discussion.—The maxilla and pre-max- ARCTODUS cf. A. SIMUS (Cope) illa are fused. II and 12 are equal size and Fig. 6 flattened transversely, as indicated by the Referred specimen.—Fragmentary max- alveoli. 13 is larger, with only the inner illa and complete pre-maxilla containing side flattened. There is no diastema be- the canine and P4, plus the alveoli of the tween 13 and the canine. The canine is left incisors and one premolar anterior to large and the crown is relatively short,

TABLE 2. Comparative measurements of Aenocyon ulnas (in mm).

Aenocyon sp. Aenocyon sp. Aenocyon dirus Friesenhahn Ingleside Ranch o Aenocyon sp. Cavern Terrace La Brea (SMUMP 60037) (TMM 933-1847) (BEG 30967-172) (TMM 31021) Over-all length '267 260 Greatest width of olecranon process 40 41.6 43.7 38.6 Greatest diameter at distal end 13 15.8 18 Report of Investigations —No. 48

a more or less triangular shape with the protocone less prominent than the paracone or the metacone. This tooth is in an ad- vanced stage of wear in the Dallas speci- men, but a rather sharp ridge on the an- terior inner edge indicates a tendency for the paracone to form a shearing blade instead of the conical cusp present in spe- cies of Ursus. It is unfortunate that this specimen is not comparable with an Arc- todus cranium from the Jinglebob occur- rence described by Rinker (1949) . The Kansas skull is 18 mm wider at the orbits than either A. simus or A. californicum. The Dallas specimen is almost 18mm wider at the incisors than are these species. It is possible that the extra width at the muz- zle extended the length of the skull, and that the materialfrom Texas is more closely related to the Kansas specimen than to the two well-known species. The Jinglebob local fauna is certainly more closely re- lated to the Hill-Shuler faunas, both geo- graphically and geologically, than to Rancho La Brea. The wider and shorter muzzle, plus the reduction of the number of premolars to two, accompanied by the crowding of the anterior premolar out of line, strongly sug- gests that this material represents an un- FIG. 6. Upper jaw fragment of Arctodus cf. A. described form. The proposal of another simus (Cope) (BEG 40442-1). (a) Lateral view, "short-faced" bear must, however, await (b) Occlusal view. x% the recovery of additional material. comprising less thanone-thirdof the tooth's Comparison of three specimens of Arc- total length. The anterior premolar is di- todus is givenin table 3 (p. 19). rectly behind the canine but inside a line URSUS AMERICANUS Pallas drawn from the canine to P4. Although a Referred specimen.—Single superior ca- portion of bone is broken away at this nine (DPSC 1023) ; Upper Shuler, Lewis- it is not a point, enough to remove possible ville site. alveoli of an additional premolar. There- Discussion.—This tooth is small, remi- the number premolars fore, of appears to niscent of the small black bears found in have been reduced to two. The number of extreme west Texas and in contrast to the premolars in Recent bears and described huge Arctodus collected from the Lower of species Arctodus varies, but the space Shuler member. between the canine and P4 is not reduced accordingly. Family MUSTELIDAE P4 has three roots. The protocone is op- PROCYON LOTOR (Linnaeus) posite the notch between the paracone and Referred specimens.—Maxilla (DPSC metacone, as in other described material 1021); Upper Shuler, Lewisville site. Frag- referred to Arctodus. This gives the crown ment of lower jaw (DPSC 1026) ; Upper Hill-Shuler Local Faunas 19

TABLE 3. Comparative measurements of species of Arctodus (in mm).

A. cf. simus A. simus A. californicum (BEG 40442-1) (UC 3001) (LAM Z-l) Greatest transverse diameter of the incisor series 74* 57 J2 64,2 Length from posterior edge of canine to anterior edge of P4 20 27 30 Antero-posterior diameter of canine 34 27.9 29.3 Antero-posterior diameter of anterior premolar 7t 8.5 7.9 Antero-posterior diameter of P4 22 20.5 20.5 Transverse diameter of P4 17.5 15 15.8

* Measurement made from inner border of 11 alveoli to outer edge of 13 and doubled, f Measurement taken from alveoli. Shuler, Lewisville site. Right lower jaw was collected by the late E. W. Shuler and

(SMUMP 60202) ; Lower Shuler, Moore described by Lull (1921, p. 160) as S. pit. cf. S. fatalis. A re-evaluation was under- Discussion. —All three specimens are taken when a right mandible (SMUMP justwithin the extreme upper size range of 60030; DPSC 304) was recovered from this species. While it is possible that only the Lower Shuler member at Pemberton large males have been collected, this series Hill; a new species was proposed, with of specimens suggest that this species may the mandible as the holotype and the cra- have averaged somewhat larger in the area nium as the paratype (Slaughter, 1960). during the Pleistocene. Large anteroposterior diameter of the

MUSTELA cf. M. VISON (Bangs) cheek teeth, relative to the mandibular measurements, plus an extremely short Referred, specimen.—Fragment of left post-canine diastema are the specific char- lower jaw containing Ml and the alveoli of acters. Although the mandibular measure- P3, P4, and (SMUMP 60047); Lower M 2 ments of the Dallas specimen compare with Shuler, Moore pit. the smallest 20 percent of the mandibular Discussion. —This specimen is referred measurements of twenty-five described provisionally because of its fragmentary specimens of S. calijornicus from Rancho nature. Because Dallas today is on the ex- five of the California jaws treme western range line of this species at La Brea, only this latitude, a Lower Shuler climate at contain Ml with antero-posterior diameter least as moist as present is indicated. as large as that found in the Dallas speci- and only one of the California jaws MEPHITIS MEPHITIS (Schreber) men contains a P4 of equal size. Furthermore, specimens.—Fragment of low- Referred have teeth the size of the teeth er jaw (DPSC 1032); Upper Shuler, approaching Lewisville site. Fragment of lower jaw of the Dallas specimen are much larger and (SMUMP 60139); Lower Shuler, Moore longer. Still further comparison with the pit. specimens from California shows that the Discussion.—Little can be learnedabout P4 of the Dallas specimen is thinner trans- Shuler climate by the presence of this ani- versely, relative to the antero-posterior di- mal in the fauna. M. mephitis is present ameter, and the inferior canine is smaller, almost all of Texas—moist today in parts relative to P4, than that of the California and arid sections, woodlands and plains. specimens. Family FELIDAE The larger size of the cheek teeth and SMILODON TRINITIENSIS Slaughter the lack of P3 in S. trinitiensis are sug- A cranium (SMUMP 60006; DPSC gested differences between it and S. trog- 152) from the Hill gravel of the Lagow pit lodytes and S. gracilis. 20 Report of Investigations—No. 48

The cranium of 5. trinitiensis (SMUMP The ratio of mastodon to mammoth in the 60006) is comparable with that of S. flori- T-2 fill is less than one to fifty, while the danus, but the Hill gravel cranium of Dal- ratio in a seemingly equivalent terrace las was so closely associated geographically east of Dallas on the Sabine River is one to and stratigraphically with the mandible of two or three. No doubt the Sabine drain- S. trinitiensis that the cranium was re- age offered a more wooded environment, ferred to the new species and used to dis- as it does today, and the difference in these tinguish S. trinitiensis from S. floridanus. proboscidean faunas is probably eco- The occiput of the Dallas cranium is much logical. narrower and triangular than the oc- more Family ciput of S. floridanus, and the mastoid-post- ELEPHANTIDAE ELEPHAS COLUMBI glenoid processes are more forwardly di- Falconer rected in the Dallas specimen. Referred specimens.—Skull, complete The recent recovery, from lower in the except for cranium and lower jaws T-2 deposits, of two additional lower jaws (SMUMP 60817); Lower Shuler, Pem- that may be more closely related to S. trog- berton Hill. Two lower jaws (SMUMP lodytes than to S. trinitiensis creates 60027); Lower Shuler, Pemberton Hill. doubt concerning the reference of the Three lower jaws (SMUMP 60063); cranium from the Hill gravel to S. trini- Lower Shuler, Moore pit. Three lower tiensis. Although S. trinitiensis now is con- jaws (SMUMP 60016); Hill, Gifford-Hill sidered distinct from S. floridanus, this pit. Ninety-four isolated teeth from Hill, sufficient refer hardly seems reason to the Lower Shuler, and Upper Shuler at all Hill gravel cranium to S. trinitiensis. This localities under discussion. discussed more a problem is fully in paper Discussion. —There has been some dis- by Slaughter (in preparation). cussion as to whether E. imperator and E. LYNX of. L. RUFUS (Schreber) columbi are truly distinct. All studied lo- Referred specimen. Metacarpal calities have produced elephant teeth, and

(SMUMP 60145) ; Lower Shuler, Moore there is a rather wide variation displayed pit. in the spacing of the enamel plates. If low- Discussion. —This metacarpal is identi- ers and uppers are grouped together, they cal to that of L. rufus, which species is oc- range from 3% to 8% plates in a 100-mm casionally reported in the area today. The line. Several of the specimens with widest specimen may also be within the size range spacing are from deep in the Hill gravel of L. canadensis but is slightly smaller and could possibly have been redeposited than the average individualof that species. from older formations. The vast majority, Order PROBOSCIDEA however, most certainly belong to the T-2 Family MASTODONTIDAE deposits. These could be divided into E. columbi and E. imperator according to the MAMMUT AMERICANUS (Kerr) tooth measurements for these species as specimens.—Maxilla fragment Referred given by Osburn (1942, pp. 1001, 1002, with two teeth (SMUMP 60256); Hill, 1075) . There, if these species are distinct, Wood pit. Isolated tooth (SMUMP it appears that characters other than 60018) ; Lower Shuler, Gifford-Hill pit. enamel should be considered Discussion.—The American mastodon plate spacing for the is present in all three members of the T-2 separation. terrace, although very rare. Three speci- Semi-articulated skeletons occur in some mens from the Wood and Moore pits and a quantity; often two or more are inter- single tooth fragment from Hickory Creek mixed. One tusk at the Moore pit was are all that are available to the writers. measured in excess of 13 feet. Hill-Shuler Local Faunas 21

Order LAGOMORPHA than the observed range of the specimens Simpson Family LEPORIDAE of P. compressus described by (1949, 29), but all other measurements SYLVILAGUS cf. S. FLORIDANUS (Allen) p. fall within the ranges of his measure- Referred specimens.—Maxilla (DPSC ments. The small difference in P4 diame- 1010) ; Upper Shuler, Lewisville site. ters is not important, because Simpson has Tooth (SMUMP 60264) ; Lower Shuler, shown that this species is quite variable. Coppel locality. The Dallas material may possibly rep- Discussion.—This species has a wide resent a population of slightly smaller habit-range and is not a useful climatic individuals Texas during the last indicator. in interglacial. LEPUS sp. TABLE 4. Measurements of Platygonus (in mm). Referred specimens.-—Single tooth Platygonus compressus (SMUMP 60231) ; Lower Shuler, Wood Platygonus sp. Observed range after (DPSC 1025) Simpson (1949, p. 30) pit. Lower jaw (DPSC 1038) ; Upper Shuler, Lewisville sites. Lower jaw (DPSC Anteroposterior diameter of P4 8.5 9.3-12.7 1042) ; Upper Shuler, Lewisville site. examined Transverse diameter Discussion. —C. W. Hibbard of P4 14.5 12.3-14.8 the two mandibles from the Lewisville site Antero-posterior and remarked (personal communication diameter ofMl 14 13.3-15.4 to Crook, 1956) that if more complete ma- Transverse diameter terial could be found, the Dallas jack rab- of Ml 12.4 11.9-16.0 bit might prove to be a new species. Although the jack rabbit inhabits plains DAMA VIRGINIANUS APLODON Slaughter, n. subsp. and open fields in the arid and semi-arid Fig. 7 also inhabits more west, it areas moist Holotype.—Right mandible containing than Dallas today. P2 through M 3 (SMUMP 60077). Order ARTIODACTYLA Referred specimen.—Right mandible Family TAYASSUIDAE fragment containing M2-M3 (SMUMP

PLATYGONUS sp. 60060); Lower Shuler, Moore pit. Type locality.—Lower Shuler, Moore Referred specimen.—P4 and Ml of a pit, one-fourth mile northeast of Pember- single individual (DPSC 1025) ; Upper ton Hill Road and Loop 12, Dallas, Texas. Shuler, Lewisville site. Diagnosis subspecific characters.— Discussion. —P4 in Mylohyus sp. and of Third or posterior lobe of MS is repre- Tayassu sp. contains two pairs of cusps; sented by a simple loop without suggestion the Dallas specimen has only one pair. Ml of the internal enamel fold or fossette pres- in these genera also has two pairs of cusps, cent in all living subspecies of D. virgini- each pair of which retains its identity anus; shorter diastema relative to length throughout life; each pair in the Dallas of the cheek tooth series; more nearly ver- forms a because of Ml continuous ridge tical ascending ramus, the worn state of the tooth. Because of Discussion.—The third or posterior lobe these differences, the Dallas teeth are re- of M 3 is a simple loop without internal ferred to Platygonus. enamel fold representing the fossette pres- Teeth are not the diagnostic features ent in the anterior lobes of the fossils and which separate valid species; therefore, in all three lobes of other described species the Dallas material is not identified spe- and subspecies of Dama. In the Dallas cifically. As shown in table 4, the antero- specimens this lobe has the same antero- posterior diameter of P4 is slightly less posterior and transverse diameters for the 22 Report of Investigations—-No. 48

jaw, lower Right a) x2. JM3, of 60077). view (SMUMP occlusal suhsp. texanus, n. Slaughter, virginianus Dama aplodon (c) x2. virginianus view, Duma occlusal 7. M3, FIG. (b) xl. Hill—Shuler Local Faunas 23 entire height, whereas in other forms it creating the illusion of a short diastem. tapers from bottom to top. This feature is The diastema index was calculated as identical in the only two Dallas specimens follows:

Length of entire cheek tooth series

— — = = diastema index Length from anterior edge of P2! to posterior edge of symphysis Dama virginianus aplodon 74 mm ■ 29 mm which contain this tooth. Seventeen lower jaws of Recent Florida There are also two specimens in the white-tail deer and twenty-seven lower Ingleside assemblage (Sellards, 1940, p. jaws of Recent Texas white-tail deer were 1650) that match these characters and measured and compared. The diastema in- seemingly should be referred to this race. dex of these specimens ranged from 1.54to These specimens are the only ones from 2.46 with all but one less than 2.28 and all that fauna that are comparable to D. vir- but seven less than 2.23. This feature is ginianus aplodon. reminiscent of the shorter diastema of the Another specimen collected from the more primitive Stockoceras rather than Seymour (late Kansan) beds of west that of the Recent form Antilocapra. Texas was examined through the courtesy The ascending ramus is more nearly of Dr. Walter Dalquest, Midwestern Uni- vertical in the black-tail and mule deer versity. Its diastema index and general than in all white-tail deer other than Dama size seem to exclude this specimen from virginianus couesi. None of these, how- the new race, but it shares with the ma- ever, is as nearly vertical as the ramus in terial from Dallas and Ingleside the sim- the Dallas specimen. This near-vertical ple loop feature of the third lobe of M3. ramus together with the shorter diastema Because the diastema is not complete in would certainly shorten the profile of the the Dallas specimens, measurements dem- living animal. There can be no doubt that the Dallas onstrating this feature were taken from the form is subspecifically distinct from other edge of anterior edge of P2 to the posterior described forms, although antler frag- the fossil Recent symphysis in and speci- ments from the same stratum affiliate the mens alike. The position of the mental for- Dallas deer with white-tail deer. amen relative to both points gives assur- D. virginianus aplodon and D. virgini- ance that a possible long symphysis is not anus texanus are compared in table 5.

TABLE 5. Measurements of Dama virginianus (in mm) .

Dama virginianus aplodon Dama virginianus texanus (SMUMP 60077) (SMUMP 60060) Average Range observed Length of cheek tooth series: P2-M3 75 77 65—89' Length from anterior edgeof P'2 to posterior edge of symphysis 31 38.5 32-46 Diastema index 2.55 1.98 1.54-2.46

Antero-posterior diameter of— P2 10 P3 11 P4 11 Ml 13 M2 15 14.5 M3 17.5 19 24 Report of Investigations—No. 48

It is probable that the numerous antler Referred specimens.—Skull, complete fragments and post-cranial elements col- except for premaxilla (SMUMP 60029; lected at all and localities levels belong to DPSC 303) ; Lower Shuler, Pemberton the same species and race, but there is no Hill. positive assurance that another species is Discussion.—Hay (1914b) described present, so most are referred to Dama sp. the occiput of C. huerfanensis as follows: Family ANTILOCAPRIDAE "The occipital surface bears a median de- scending ridge, rough and rounded, sepa- BUEAMERYX sp. rating two deep excavations, on each side Referred specimens.—M 2 (SMUMP of which is another deep excavation at the 60034) ; Lower Shuler, Pemberton Hill. bottom of which is placed the lateral fora- P4 (SMUMP 60262); Lower Shuler, men." Lull (1921, p. 169) commented on Coppel locality. this: "In the Dallas specimen the median Discussion.—There are no important ridge is of less vertical extent, so that the differences between the Dallas specimens deep excavations on either side are con- and California material referred to B. fluent below, while in C. huerfanensis they minimus Meade. Comparative measure- are entirely separate." The occipital char- ments of the Dallas teeth are in table 6. acters of the new skull (SMUMP 60029) are identical to those of the holotype de- TABLE 6. Measurements of teeth referred to scribed by Lull. This further substantiates Breameryx sp. (in mm). the difference between C. huerfanensis and C. dallasi as stated by Lull. (SMUMP (SMUMP huerfanensis 60034) 60262) Merriam (1913, p. 311), speaking of the Anteroposterior diameter 9.5 7.5 median ridge, stated: "In the Rancho La Transverse diameter 4.5 3 Brea specimens [ C. hesternus] there is a Vertical diameter 28 13 short low median crest at the upper end of the occiput in no. 20028 and no. 20040; TETRAMERYX SHULERI Lull in 20049 it is scarcely visible." Holotype.—Skull cap with horn-cores This ridge, therefore, is longer if not and left maxilla with complete cheek tooth stronger in C. huerfanensis than in the series (SMUMP 60006; DPSC 152). Low- Dallas subspecies, whereas the same ele- er Shuler, Lagow pit. ment is decidedly weaker in the California Referred specimen.—Left M 2 (SMUMP specimens referred to C. hesternus. C. hu- 60134). Lower Shuler, Moore pit. erfanensis dallasi is thus intermediate be- Discussion. Specimen (SMUMP tween C. huerfanensis and C. hesternus on 60006) was collected from the Lagow pit the basis of the median ridge. by the late E. W. Shuler and described by In his discussion of the diastema be- Lull (1921, p. 163) as the holotype of the tween 13 and the superior canine in C. genotypic species. The horizon from exact (1914b, 269) said: , Hay p. which the specimen was collected is not huerfanensis "The canine must have emerged immedi- known, but the preservation indicates it behind the does may have been Lower Shuler. ately incisor, just as it in the Bactrian camel." This diastema is 45 Family CAMELIDAE mm in C. kansanus and this, according to CAMELOPS HUERFANENSIS DALLASI Lull Merriam (1913, p. 318), agrees with the P1 I, figs. I, 2 California specimens referred to C. hester- Holotype.—Partial skull and superior nus. This same measurement taken from dentition (SMUMP 60005; DPSC 151); the type of C. huerfanensis dallasi is 17 Lower Shuler, Lagow pit; collected by the mm. The length of the diastema is another late E. W. Shuler. character in which the Dallas subspecies is Hill—ShulerLocal Faunas 25 intermediate between C. huerfanensis and Perhaps the fact that C. huerfanensis C. hesternus. dallasi has features of both C. huerfanen- A large deep fossa near the upper mar- sis and C. hesternus suggests that all three gin of the maxillary above P4 was men- are so closely related that they are the tioned by Merriam (1913, p. 308) as be- same species and are no more than racial ing present in C. hesternus. This fossa is variants. identical in the skull (SMUMP 60029) In table 7, the holotype and newly re- from Dallas and may be another correla- ferred specimen of C. huerfanensis dallasi tion between the Dallas material and C. are compared to C. hesternus. hesternus. Because the fossa is not known for the other species in the genus, however, ?CAMELOPS sp. that it is possible the large deep fossa is P1. I, fig. 3; text fig. 8 not indicative. The posterior palatine foramen occupies Referred specimens.—Cranium poste- rior to the orbits (SMUMP 60237) ; a position opposite P3 or P4 in C. hester- Lower Shuler, Haymarket pit. Metatarsal nus from California, while in both C. huer- (SMUMP 60570) and C. dallasi, the ; Hill gravel, Moore pit. fanensis huerfanensis —The third camelid skull former is opposite Ml. The foramen lies Discussion. collected from T-2 deposits at Dallas differs at the anterior edge of Ml in the new skull in several respects from other skulls re- (SMUMP 60029) referred to C. huerfan- ferred to C. dallasi. The sagit- ensis dallasi. This is posterior to the posi- huerfanensis tal crest is very weak and the brain case tion occupied in C. huerfanensis dallasi or C. huerfanensis. The first two cheek teeth itself much less vaulted than any of the in this specimen are deciduous, however, species referred to the genus Camelops with and may allow the foramen to appear which it has been compared. The lamboi- farther forward than it would be in adult- dal crest overhangs the occiput more than C. or the hood. in C. hesternus, huerfanensis,

TABLE 7. Skull and dentition measurements of C. huerfanensis dallasi, Dallas, and C. hesternus, Rancho La Brea (in mm).

C. huerfanensis dallasi c. hestemus (SMUMP 60005; (SMUMP (UC (UC DPSC 151) 60029) 20028) 20040) Greatest skull width posterior to orbits 195 245 251 Greatest height of orbits 50 63.3 61 Least width of brain case immediately behind orbits 83 80* 77 83 Length from anterior edge of P4 to posterior edge of M3 123 127 141.9 Antero-posterior diameter of P3 8t 23 18.8 Transverse diameter of P3 6t 12 11 Antero-posterior diameter of P4 35f 26 23.5 28 Transverse diameter of P4 22f 23 '25 22.5 Antero-posterior diameter of Ml 40 35 24.4 42 Transverse diameter of Ml 21 29.5 31 33.6 Antero-posterior diameter of M2 49 44 42.1 52 Transverse diameter of M2 20 29 31.6 32.8 Antero-posterior diameter of M3 50t 51.5 49.5 45.8 Transverse diameter of M3 26 31.4 27.2

* Estimated, t Deciduous, t Unerupted. 26 Report of Investigations—No. 48

ments of post-cranial skeletons have been collected from both Upper and Lower Shu- ler which match very closely the material associated with the holotype skull. There are, however, some elements of the skeH ton fragments from the Shuler that are a little larger than the same elements known to belong to C. huerfanensis dallasi. This is not attributed to age because the type specimen represents a very old individual. There is much variation in the limb bones of Recent camels, and this may account for these fossil differences. On the basis of the strange cranium (SMUMP 60237) and the large metatarsal (SMUMP 60570), there may be two large camels in the fauna. Metapodial measurements of ?Camelops sp. compared with those of C. huerfanensi dallasi (table 8) show the distinctly larger size of ?Camelops sp.

TABLE 8. Measurements of camel metapodials from T-2 deposits (in mm).

C. huerfanensis dallasi (SMUMP ?Camelops 60014) (SMUMP sp. Belongs 60011) (SMUMP to type Referred 60570) Width at proximal end 60 66 75 Width at mid-shaft 36 39 47 Wi dth at distal en d 97 Over-all length 381

TANUPOLAMA sp. Fig. 9 FIG. 8. Metatarsal of ?Camelops sp. (SMUMP 60570). x% Referred specimens.—Fragment of left lower jaw containing P4, Ml, and M C. dallasi skulls. The occiput 2 huerfanensis (SMUMP 60015); Lower Shuler, Lagow is almost completely smooth without a left trace of the excavations present in the occi- pit. Upper M 2 (SMUMP 60195); Hill Milton Metatarsal (SMUMP puts of the skulls of C. huerfanensis dallasi gravel, pit. ; Wood and possesses only a hint of a descending 60239) Lower Shuler, pit. median ridge. Even the lateral foramen are abrupt holes without the slightest in- dentation around them. The brain case is considerably wider relative to its length than the types mentioned above, including the Dallas subspecies. The first cervical vertebra, the ulna, and a metatarsal of C. huerfanensis dallasi were directly with the skull associated FIG. 9. Left mandible fragment of Tanupolama designated as the holotype, and a few frag- sp. (SMUMP 60015). x% Hill—Shuler Local Faunas 27

Discussion. —The lower jaw (SMUMP horn core (SMUMP 60067) ; Lower Shu- t)0015) referred to this genus, like the Dal- ler, Moore pit. Metacarpal (SMUMP

las Camelops material, contains features of 60249) ; Lower Shuler, Wood pit. Meta- several described species and is exactly carpal (SMUMP 60248); Lower Shuler, ' none. The Dallas jaw lacks P3, which Moore pit. Metatarsal (SMUMP 60266) ; is said to be variable in T. stevensi, absent Upper Shuler, Coppel locality. Metatarsal Lower Moore in T. Americana, and consistently present (SMUMP 60186) ; Shuler, in T. mirifica (Simpson, 1929, p. 595). pit. The lower molars are similar in size to Discussion.—Post-cranial elements un- both T. stevensi and T. mirifica, except for identifiable to species occur at every lo- their shorter transverse diameter, which cality and in all three members. These are may be attributed to adolescence. Ml in referred to Bison sp. but no doubt belong T. stevensi contains a median external to the same species as represented by the style not present in the Dallas specimen or identifiable material. If size alone were in T. mirifica. considered, the only complete horn core The isolated M 2 in the Dallas collection (SMUMP 60067) would fall within the of B. antiquus as outlinedby Skinner also resembles more closely the same tooth range and Kaisen (1947, p. 178). Although the in T. than in T. stevensi, in that mirifica of this the horn the anterior lobe is considerably broader burr specimen suggests that of a the relatively smooth transversely than the posterior lobe. core is male, surface indicates that be immature. The Dallas metatarsal agrees almost per- it may (SMUMP 60003) and fectly with two metatarsals of T. stevensi Horn cores (SMUMP 60317) are the proper size and with which it was compared. proportion to be B. alleni. Post-cranial ele- Dentition measurements of several ments from the terrace deposits vary in specimens of Tanupolama are given in size from small to with meta- table 9. very large, carpals and metatarsals in the upper 20 Family BOVIDAE percent of the size range of the thou-

BISON ALLENI Marsh sands of elements measured by Skinner Kaisen (1947, 135, 136). Their P1. I, figs. 5-8; text fig. 10 and pp. measurements of several species even in- Referred specimens. Horn core cluded the gigantic B. latifrons. Horn cores (SMUMP 60317) Lower ; probably Shu- of B. alleni from theLower Shuler member ler, Seale pit. Horn core (SMUMP of the T-2 deposits are compared in table 60003) ; Hill or Lower Shuler, Lagow pit. 10 with the range of measurements of Pair of horn cores (SMUMP 60019); B. alleni as given by Skinner and Kaisen Lower Shuler, GifFord-Hill pit. Female (1947, p. 184).

TABLE 9. Measurements of Tanupolama dentition (in mm).

T. stevensi CITVPC Tanupolama sp. T. mirifica (SMUMP 60015) (AMNH 23487) 4 5 7 Antero-posterior diameter of P4 T2 18 15,2 14 17.8 Transverse diameter of P4 6 8 8.7 7.7 8.8 Antero-posterior diameter ofMl 22 23 21.2 21 Transverse diameter of Ml ( 11 15 14.4 14.9 Antero-posterior diameter ofM2 27.5 27 25.3 24 29.4 Transverse diameter of M2 11 15 14.1 16.6 16.8 Antero-posterior diameter of M'2 25 24 29.7 28.7 23.9 Transverse diameter of M2 22. 23 17 19 25,3 28 Report of Investigations—No. 48 Meta- (c) 60281). (SMUMP (female) Metacarpal

x% All (SMUMP 60186). (male) (SMUMP Metacarpal (female) (a) Marsh, Metatarsal alleni (d) Bison 60266). of Metapodials(SMUMP (male) (b) 60249).10 FIG tarsal Hill—Shuler Local Faunas 29

TABLE 10. Measurements of B. alleni horn cores (in mm).

B. cdleni range Skinner and Kaisen SMUMP (1947, p. 184) 60317 60003 60019 60067 SMUMP SMUMP SMUMP (Female) Minimum Average Maximum Core length on upper curve, tip to burr 540*-750f 400$ 305 275 400 561 720 Core length on lower curve, tip to burr 710*-900f 4504 357 330 480 647 890 Transverse diameter at base no* 128 '112 75 133 151 169 Vertical diameter at base 110* 100 97 66 107 1241 145 Circumference at base 340* 360 325 240 378 434 495 Tip to burr, straight line 440* 350$ 280 250 355 471 575 Index of curvature 128$ 123 132 121 135 155 Index of compression 78 86 77 72 82 91

* Exact measurement of fragment. f Estimated minimum according to solidness of fragment. J Measurement taken from reconstruction; must be considered approximate.

Skinner and Kaisen's (1947, pp. 142-143) method for determining indices of horn cores is as follows: Horn-core length on lower curve —— = index of curvature Distance, core tip to upper base of burr Vertical diameter of core at base

—— —— : index of compression 1 ransverse diameter of core at base

Table 11 is included for comparison of isolated teeth and the enamel patterns sug- measurements of bison metapodials col- gest a seemingly excessive number of spe- lected from the Hill-Shuler local fauna. cies, a thorough study of other collections It is concluded that the smaller elements was undertaken to prove or disprove pre- represent females and the larger, males; liminary identifications. After reviewing therefore, it is unlikely that one can rule the literature and examining numerous out the possibility of the larger bison forms Texas collections, the writers were unable being present in an assemblage where only to reduce the number of species in the post-cranial elements are known. Hill-Shuler local faunas. J. H. Quinn (oral Order PERRISSODACTYLA communication at site, 1958) suggested that the North American equine faunas of Family EQUIDAE the Pleistocene may have been almost as Vertical ranges after Quinn (1957) diversified as the present-day antelope of Discussion. Specific identifications Africa, with size which ranged from quite based on single teeth are unreliable be- large to very small, and with equally varied cause of the unknown range of variation habits. The ranges of eastern types, such possible within a single fossil species. Be- as Equus jraternus, may have overlapped cause most of the Dallas specimens are with southwestern types, such as A sinus

TABLE 11. Measurements of Dallas bison metapodials (in mm).

Metacarpals Metatarsals

Male Female Male Female (SMUMP (SMUMP (SMUMP (SMUMP 60249) 60281) 60266) 60186) Over-all length 229 224 298.5 264.9 Transverse diameter at proximal end 85.8 74.8 66.9 65,5 Transverse diameter at mid-shaft 65,4 42.4 47.5 40 Transverse diameter at distal end 91.3 74.6 80.8 72.6 30 Report of Investigations—No. 48

conversidens, only where optimum condi- bard and Taylor. If one adheres strictly to tions existed. Such conditions in the form the diagnosis of as outlined by of ample rainfall and an equable climate Quinn, each group of material from are suggested for Hill and Lower Shuler Dallas offers something of a paradox. The time at Dallas. A similar situation may not teeth referred provisionally to E. fraternus have existed to the east or west; conse- do not have elongated protocones nor is quently, the horse faunas may have been the similar-size metacarpal particularly more restricted in number of species. An long or slender. The metapodials of the equable climate is demonstrated at Dallas new species are quite long and slender, in the lower members of the T-2 fill by the but the similar-size teeth do not have espe- association of such animals as Synaptomys cially long protocones and the enamel pat- and Sigmodon plus certain snails. terns are not complicated. The writer Quinn (1957, p. 10) proposed the use of agrees withQuinn that will be much the generic name Onager for several Pleis- more helpful when they can be separated tocene forms from and into groups, which no doubt exist, but the Asia; among the diagnostic characters, he Dallas material is not referable to such listed complex enamel patterns of upper groups with confidence. cheek elongate protocones, and un- teeth, EQUUS MIDLANDENSIS Quinn usually long, slender metapodials. Hibbard (Wisconsin?) and Taylor (1960, p. 192) did not follow Quinn's use of Onager but instead pre- Figs. 11, 12(a) ferred Hemionus and quoted Ellerman and Referred specimens.—Right metacarpal Morris-Scott's statement that Brisson used (SMUMP 60079); Hill-Lower Shuler Onager for Equus asinus. Quinn (personal contact, Moore pit. Isolated teeth, Hill and correspondence with the writers, citing Lower Shuler members, Wood and Moore several references) has demonstrated, rath- pits. er convincingly, that the terms onager, Discussion.—The only difference be-

Asinus Equus , and Onager onager, onager tween teeth referred to E. midlandensis have rather been used consistently in con- and those referred to E. scotti, which the nection with the Persian ass. If, then, all writer has examined, is a slight tendency Asiatic asses are grouped together, it seems for the mesostyle in teeth of E. midlandensis proper to include them under the name be perpendicular to the Onager, whether it proves to be of generic to more nearly tooth. This or simply subgeneric rank, and Hemionus antero-posterior midline of the could be used only in a lesser capacity to slight difference might vary if numerous include Mongolian and Tibetan asses. specimens were available. The Dallas teeth The writer's reasons for not using Onager are referred to E. midlandensis provision- as generic rank in this paper, therefore, ally because this species is generally con- are somewhat different from those of Hib- sidered a late Pleistocene type, whereas

FIG. 11. Equus midlandensis Quinn. (a) Left P2 (SMUMP 60233). (b) Right P3 (SMUMP 60225). xl Hill—Shuler Local Faunas 31

E. scotti is found in middle Pleistocene Shuler members, Pemberton Hill, Wood, deposits. and Moore pits. Although the metacarpal (SMUMP Discussion. —Medium-size teeth with 60079) has a mid-shaft transverse diam- strongly folded enamel patterns collected eter slightly greater, relative to the prox- from the Hill and Lower Shuler members imal and distal transverse diameters, than have been referred provisionally to E. the Recent horse material at hand, the fraternus. metacarpal probably could be duplicated Four sizes of metapodials from the T-2 in some specimens of E. caballus. This is deposits are in the Dallas collection. These said to be true in both E. midlandensis and were placed with teeth of corresponding E. scotti (Gidley, 1901, p. 116; Quinn, size: the 1 argest teeth withthe largest meta- 1957, p. 26). podial, the smallest teeth with the smallest

EQUUS cf. E. FRATERNUS (Leidy) metapodial, and so on. The two smallest may grouped with (Post-Illinoian to post-Wisconsin) sizes and the largest be some confidence, but there are two types of Figs. 12(b), 13 teeth that would fit metacarpal (SMUMP Referred specimens.—Left metacarpal 60238) very nicely—those referred to (SMUMP 60238) ; Lower Shuler, Wood E. fraternus and those referred to Equus pit. Isolated teeth from Hill and Lower sp. Because this metacarpal is very horse-

FIG. 12. Equine metapodials. (a) Metacarpal of Equus midlandensis Quinn (SMUMP 60079). (b) Metacarpal of Equus cf. E. fraternus (Leidy) (SMUMP 60238). (c) Metacarpal of Asinus conversidens (Owen) (SMUMP 60013). (d) Metatarsal of Equus? quinni Slaughter, n.sp. (SMUMP 60228). All xYs. 32 Report of Investigations—No. 48

FIG. 13. Equus cf. E. fraternus (Leidy). (a) Right P2-P3 (SMUMP 60135). (b) Right M2 (SMUMP 60234). xl

like ( Equus s.s.) and because the proto- mens to that species because the enamel cones of the teeth referred to Equus sp. folds of the Hill-Lower Shuler teeth are are extremely long (a character attributed much stronger than in the holotype of E. to Onager), the metacarpal is referred lambei. According to Hay's figures and provisionally to E. fraternus along with the text (1917, p. 435), simplicity of the Dallas teeth which do not have exception- enamel pattern is one of the characters of tlly long protocones. the E. lambei. If this reference is proven by the future ASINUS CONVERSIDENS (Owen) recovery of a like metapodial in association (Sangamon? to late Wisconsin) with E. fraternus teeth, Quinn's reference of this species to the Onager group may be Figs. 12(c), 15 questioned. Referred specimens.—Right metacarpal EQUUS sp (SMUMP 60013); Lower Shuler, Lagow Left Fig. 14 pit. metacarpal (SMUMP 60194) ; Lower Shuler, Wood pit. Isolated teeth. Referred specimens.—Isolated teeth from the Hill and Lower Shuler members.

FIG. 15. Asinus conversidens (Owen), (a) Right P4 (SMUMP 60272). (b) Right M 2 FIG. 14. Equus sp. (a) Left Ml (SMUMP (SMUMP 60196). xl 60197). (b) Left Ml (SMUMP 60223). xl Discussion.—Some of the teeth from the Discussion.—There are a number of T-2 deposits referred to A. conversidens teeth in the Dallas collections that are sim- are similar to those referred to E.? quinni ilar to those of E. cf. E. fraternus but which except that they are larger than those of have protocones ranging up to three-fifths E.? quinni. The Lower Shuler teeth aver- of the total antero-posterior length. This age slightly larger than the same teeth of character is attributed to E. lambei, but the the holotype as given by Hibbarcl (1955a, writer is hesitant to refer the Dallas speci- p. 59). Hill—Shuler Local Faunas 33

Although the two Lower Shuler meta- 60208); Lower Shuler, Wood pit. Proxi- carpals are larger than the metacarpal from mal end of metatarsal, (SMUMP 60165) ; Papago Springs referred to this species Lower Shuler, Moore pit. by Skinner (1942, p. 171), their ass-like Type locality. —Lower Shuler, Wood character was used in their reference to pit. A. conversidens. Thus Skinner agreed with Diagnosis of specific characters.—Meta- Quinn that A. conversidens is an ass, podials are extremely long and slender, though not of the Onager type. The larger with diameters smaller than any specimens size of the Lower Shuler metacarpals and previously described from the late Pleisto- the referred teeth from the T-2 deposits cene (table 12). Referred teeth smaller suggest that the species averaged some- than other late Pleistocene species. what larger at Dallas than farther to the Discussion and comparisons.—Hibbard southwest. Both metacarpals can be dupli- (1955a, pp. 56-62; Hibbard and Taylor, cated in the A. hydruntius. European ass, 1960, pp. 189-193) concluded that the

EQUUS? QUINNI SLAUGHTER, n.sp late Pleistocene types—Asinus francesi, Figs. 12(d), 16 Equus littoralis, and Equus tau—are syno- nyms of A. conversidens. Based on differ- Specific dedication of the new form is ences between the small metatarsals from made, with pleasure and appreciation, to the Hill and Lower Shuler members at Dr. James H. Quinn. Dallas and the same element of A. Holotype.—Right metatarsal (SMUMP fran- evident that there least 60578); Lower Shuler, Moore pit. cesi, it is now are at Paratype.—Left metacarpal (SMUMP two small equine types in the late Pleisto- 60578; Lower Shuler, Moore pit. cene of North America. Although the Referred specimens.—Left metatarsal writer is not suggesting that Hibbard's sup- of the synonymity of A. (SMUMP 60174) ; Hill, Moore pit. M2 position francesi, tau incorrect, E. littoralis , and E. is he is (SMUMP 60101) ; Lower Shuler, Moore pit. Ml (SMUMP 60146); Lower Shuler, comparing E.? quinni with these species separately to show that the new material Moore pit. M 2 (SMUMP 60467) ; Lower Shuler, Moore pit. Distal end of metapodial does not fall within a range of variation (SMUMP 60045) ; Lower Shuler, Moore represented by material referred to all of pit. Distal end of metapodial (SMUMP these species.

FIG.16. Equus? quinni Slaughter, n.sp. (a) M 2 (SMUMP 60101). (b) MJ_ (SMUMP 60146) (c) M 2 (SMUMP 60467). xl TABLE 12. Measurements of Hill and Lower Shuler equine metapodials (in mm).

E. E. A. Eqilus? quinni, n.sp. midlandensis fraternus convursidens Metacarpal Metatarsals Metacarpals (Paratype) (Holotype) (SMUMP (SMUMP (SMUMP (SMUMP (SMUMP (SMUMP 60079) 60238) 60013) 60578) 60228) 60174) length Over-all .... 245 219 221 235 287 277

Transverse diameter at proximal end ...... 53 48.8 42 31.4 35 32.2

Transverse diameter at mid-shaft ..... 40.5 36 30 23.4 22 23.5 Antero-posterior diameter at mid-shaft . . 27.5 27 24 21.4 25 25.5

Transverse diameter at distal end ..... 51 43 39 31.6 30.9 29 34 Report of Investigations—No. 48

Quinn (1957, p. 13) stated that Onager Although smaller, the simple enamel littoralis (Hay) is the "smallest of the folds of the small Dallas teeth, both upper Equini, teeth slightly larger than those of and lower, are very nearly like those of A. Nannippus phlegon ." The measurements francesi and E. tau. Quinn (personal com- of E. littoralis as by Hay (1913, munication, 1959) as out, how- , given p. pointed 575), are almost identical to those of A. ever, that enamel patterns tend to become conversidens. In addition, the very simple simpler and the characters less certain with enamel patterns of tiny teeth from Dallas excessive wear. Therefore, if the teeth re- referred to E.? quinni are in direct con- ferred to E.? quinni are not of maximum trast with Hay's (1915, p. 575) statement: size for this species and the teeth of the A. "Enamel surrounding the lakes [in E. lit- francesi holotype are not of minimum toralis] rather strongly folded." It has size, it would be difficult, if not impossible, been pointed out by Quinn (personal com- to refer isolated teeth of intermediate size munication, 1959) that the length of upper to either. Excessive wear also tends to re- teeth of horses plays an important role in duce the diameters of check teeth, and this determining tooth diameters and the com- may account for the size differences be- plexity of enamel patterns. This point is tween the horses compared. The length of well taken; it can be demonstrated that the Dallas teeth (Ml, 75 mm; M2, 67.4 the antero-posterior and transverse tooth mm) is greater than that of any of the diameters are reduced as the crown is specimens here compared; therefore, shortened through wear. Mi's in the holo- larger, instead of smaller, diameters should types of the other species under discussion, be expected. Although the diameters of the however, are as short as or even shorter E. ? quinni metatarsals are but 2 to 3 mm metatar- than the Dallas Ml. The diameters of the less than those of A. francesi, the teeth of E.? quinni should thus be as large sal lengths of 277 and 287 mm are striking or larger, but they are not. when compared with the over-all length of The size of E. tau has been somewhat 225 mm for the metatarsal of A. francesi. confused in the literature. When Owen Comparison between the metacarpal of (1869, p. 565) proposed the species, he E.? quinni and the same elements of A. stated that it was much smaller than A. conversidens and A. francesi is equally conversidens. Cope (1884, p. 12) consid- striking. The diameters of E.? quinni ered the two synonymous but later (1893, range from 2.6 to 10.6 mm less than those p. 80), in discussing E. semiplicatus, said: of either A. conversidens and A. francesi, "Teeth indicate a species of about the same and the over-all length of the metacarpal of dimensions of the E. tau." He thus sug- E.? quinni is 14 and 15 mm greater re- gested a larger size for E. tau. Hibbard spectively than the metacarpals of the (1955a, p. 60) stated: other two species. In fact, the E.? quinni (M3) "In Equus tau this molar has a greater metacarpal is longer than the metatarsal of relative size [than the same tooth in Asinus con- versidens], especially in antero-posterior length. A. francesi. Because the figure of is dis- (1957, p. 12) referred a meta- Equus [The Quinn torted, that of tau probably is also. pit in holotype of E. tau is lost.] Measurement of the tarsal from the Holloman gravel illustration published by Owen of Equus tau es- Oklahoma (late Kansan) to Onager semi- tablishes a antero-posterior length of P3- greater plicatus. The Holloman metatarsal, al- M 3 series than for the same series in the illustra- than those of E.? tion of Equus conversidens. If, however, the illus- though much larger tration by Owen of Equus tau, is correct, the longrelative to its di- quinni, is also quite length for the P3-M3 series is the same as that in ameters. The same is true of E. calobatus the type specimen of Equus conversidens (meas- ured from the type). According to the measure- from the late Kansan of Kansas. E.? ments Equus tau was not a smaller horse than quinni is most certainly closely related to Equus conversidens. . . . The name Equus tau is probably a synonym of Equus conversidens." these earlier equine types. Hill—Shuler Local Faunas 35

Although the Dallas teeth referred to cesi, with an over-all length of 78.5 mm E.? quinni were not collected in direct as- compared to 69 mm for A. francesi, and is sociation with the five metapodials, these 30 percent narrower with a proximal teeth are referred with some confidence as transverse diameter of 25 mm compared to both the teeth and the metapodials have 36 mm for A. francesi. The comparison of smaller diameters thanany described from this longer, more slender phalanx the late Pleistocene. To say that the teeth with the phalanx of A. francesi is remi- and the metapodials may not belong to the niscent of the comparison between the same species would necessitate suggesting metatarsals of E.? quinni and A. francesi. that there are two equine types in the It is probable, therefore, that E,? quinni is Lower Shuler member that are smaller associated with A. conversidens in the Pa- than any described heretoforefrom the late pago Springs Cave, as it is in the Lower Pleistocene, Shuler at Dallas. Reference of the new species to Equus Considering the possible association of is rather doubtful. The splint bone scars on E.? quinni with (1) Nannippus in the un- the complete metatarsals are very long and known northwest Texas locality, (2) Ste- are quite like metapodials of the three-toed gomastodon at Broadwater, Nebraska, and genus Nannippus, which is known to have (3) a typical Wisconsin fauna at Papago survived into the early Pleistocene. The Springs, Arizona, E.? quinni may have protocones of Nannippus are isolated, been a rather long-ranging species. Why, the small Dallas teeth however; in they then, should it be so rare elsewhere and yet are not. one of the most abundant forms in the In a small collection at the Bureau of Lower Shuler at Dallas? Perhaps, being a Economic Geology, The University of southern form, it moved north from Mex- Texas, there are teeth of two distinct horses ico only during interglacials. Yarmouth de- from an unknown locality in northwest posits are practically unknown in Texas, Texas. One is N. phlegon; the other is and reported Sangamon deposits are very slightly larger based on cannot and, teeth, rare south of Kansas. Both size and form be from E.? quinni. Associ- distinguished of the 3 figured by Hibbard (1955a, fig. ated with these specimens is a complete M 4-d, p. 63) from Mexico compare well with metatarsal of the proper size for Nannippus Dallas teeth referred to E.? quinni thus teeth and a broken metatarsal exactly like the supposition that E.? the holotype of the new species from strengthening the Dallas. quinni found refuge in Mexico during Barbour and Schultz (1937, p. 4) list an glacial periods. "Equid (very light-limbed form; known Table 13 (p. 36) compares teeth meas- only from a single metapodial)" from urements of E.? quinni with other small Broadwater, Nebraska, in association with Pleistocene Equidae. Stegomastodon. Linnaeus It is also noteworthy that Skinner (1942, EQUUS cf. E. CABALLUS Recent) p. 170) described a first phalanx from the (Wisconsin to Papago Springs Cave, Arizona, as too Fig. 17 small to to belong A. conversidens. He Referred specimens. Left P2-M3 referred this specimen to Equus tau, say- (SMUMP 60036) ; Upper Shuler, ing: "The single specimen of Equus tau is Hill. Isolated teeth (SMUMP 60490) ; very different from the larger Equus con- ton Lewisville site. versidens of the cave fauna. The phalanx Upper Shuler, is slender proportioned and resembles Discussion.—When M. F. Skinner ex- some of the large Pliocene Equidae." The amined the collection from the Lewisville Arizona phalanx of E. tau is 14 percent site in 1956, he discovered that two sizes longer than the same element of A. fran- of caballine-type horses were represented. 36 Report of Investigations —No. 48

FIG. 17. Equus cf. E. caballas Linnaeus, (a) Left P4 (SMUMP 60036). (b) Right P2 (SMUMP 60490). xl Skinner did not attempt to make specific Left mandible fragment less teeth identification because of the fragmentary (SMUMP 60110); Lower Shuler, Moore nature of the specimens. Better material pit. Right mandible fragment less teeth since collected from the Lewisville site (SMUMP 60572) ; Lower Shuler, Wood and Pemberton Hill cannot be separated pit. from the modern form. Discussion. —The Dallas tapir material Because radiocarbon dates indicate an is very fragmentary and is referred to this age in excess of 37,000 BP for the Lewis- genus on the basis of age of the deposits. ville site, this is a very early report for this The tapir cannot be separated from earlier species in America. With minor discrepan- genera by size or form. Tooth measure- cies, the ranges given for the other equine ments used in comparisons were taken species are in line with the age allocation from alveoli and could be misleading, but of the T-2 deposits. at least a rather general size can be deter- mined. All described species from the Family TAPIRIDAE Pleistocene of North America are near the TAPIRUS sp. upper size range of the Recent species T. specimens. or Referred —Right metacar- terrestris, larger. Because the three pal 111 (SMUMP 60093); Hill, Moore pit. Dallas specimens are quite small, al- TABLE 13. Measurements of teeth of small Pleistocene Equidae (in mm).

Equus littoralis

Equus? quinni , n.sp. (WFIS 4086) (SMUMP 60146) (SMUMP 60101) (SMUMP 60467) (Holotype) Anteroposterior diameter of Ml 18.7 21.5 Transverse diameter of Ml 18.5 22 Length of protocone 10.9 12 Antero-posterior diameter of M2 17.2 21 Transverse diameter of M2 17.9 20 Length of protocone 10 11 Antero-posterior diameter of M2 18.5 Transverse diameter of M2 11.8

Equus tau Asinus cotwersidens (UNAM; lost) (UNAM 403) (holotype) Asinus francesi (After Hay, (After Hibbard, 1955a, p. 56) (TAM; no number) 1915, p. 539) Right Left Antero-posterior diameter of Ml 21.7 22 19.5 21 Transverse diameter of Ml 22 21.7 22 20 Length of protocone 10.5 11.1 11 11 Antero-posterior diameter of M2 21.7 21.9 20 22 Transverse diameter of M2 20.8 19.8 21 18.5 Length of protocone 11.3 11 12 11.5 Antero-posterior diameter of M2 20 Transverse diameter of M2 12.5 Hill-ShulerLocal Faunas 37 though perhaps within the lower range of make specific identification possible, but T. terrestris, the Dallas tapir is closer to the present writers have attempted to make T. terrestris than to the fossil species de- only generic identifications based on scribed from the Pleistocene. single plates. Until quite recently, tropical, rain-for- est climate has been associated with the Family EMYDIDAE presence of tapir in fossil faunas, based on GRAPTEMYS GEOGRAPHICA (Le Sueur) the habitat of the living Central American specimens.-—Third, 4th, and species. Referred Martin (1958, p. 412) suggested sth nuchals; right 3rd, 4th, and sth pleu- that the presence range of T. terrestris may rals (SMUMP 60109); Lower Shuler, represent a refuge from early man and not Moore pit. a climatic preference. Also, one must con- Discussion. —lt is evident that the cara- sider the possibility that a fossil species may not be closely related to the geo- pace of this turtle was somewhat more flat- graphically nearest living representative. tened than that of the sawback turtle (G. The recent species T. roulini dwells in the pseudographica) present today in the area, mountain forest of Ecuador and Central and the tubercles on the nuchals were less America at elevations up to 8,000 feet. De- well developed. These are both characters sire Roulin, while accompanying Bous- of G. geographica. The present range of singault on his 1824 Andean expedition, this species is no closer than 150 miles came across a strange animal in the bare northeast of Dallas. Collection of this speci- mountain of the Andes in Colom- heights men and other single plates from theLower the South Ameri- bia, "resembling typical Shuler member indicates that the climate can tapir but has long, thick hair like a during deposition of the Lower Shuler bear's" (Goodwin, 1954, p. 669). was more like that of deep east Texas Varous elements of the Dallas tapir ma- of terial are compared (table 14) with those today. The antero-posterior diameter of the Recent tapir, T. terrestris. a sth nuchal of a Recent specimen with an over-all carapace length of 176 mm is Class REPITILIA only 18 mm. Because the largest reported Order TESTUDINATA Recent specimen has an over-all length of Single carapace plates of turtles are 253 mm, the turtle to which the Lower abundant in the T-2 deposits but seldom is Shuler sth nuchal belonged was probably more than one plate of a single individual equal to or slightly larger than the maxi- recovered. A more detailed study may mum size of recent forms.

TABLE 14. Measurements of Dallas Tapirus material and observed size ranges of T. terrestris (in mm).

Tapirus sp. T. terrestris

(SMUMP (SMUMP (SMUMP Observed range 60110) 60572) 60093) (Simpson, 1945, p. 44) Antero-posterior diameter of P4 20* 18.7-21.9 Transverse diameter of P4, anterior lobe 13* '16.0-18.5 Transverse diameter ofP4, posterior lobe 13.4* 16.5-19.1 Antero-posterior diameter of Ml 21 19.5-24.0 Transverse diameter of Ml, anterior lobe 14 14.8-17.9 Transverse diameter of Ml, posterior lobe 15 14.5-17.0 Length of metacarpal III 1114.5 126 Transverse diameter of metacarpal III at mid-shaft 2:2 29.5

* Measurement made from alveoli. 38 Report of Investigations—No. 48

TERRAPENE CANALICULATA Hay what smaller than the very large frag- Referred specimens.—Half carapace ments found in the Hill and Lower Shuler. and complete plastron (DPSC 1063) ; Unfortunately, these fragments are useless Upper Shuler, Lewisville site. Almost com- for specific identification. There is, how- plete carapace (SMUMP 60310) ; Upper ever, one well-preserved specimen in the Shuler, Hickory Creek locality. Dallas collection (SMUMP 60028) con- Discussion. —Holman (1958 p. 277) sisting of almost half of the plastron and and others have considered T. canaliculata one-fourth of the carapace. An attempt was a synonym of T. Carolina, but the speci- made to compare the Dallas specimen with mens of the T-2 deposits are so much larger the Geochelone material from the Ingle- than the Recent specimens reported that side occurrence, described by Pyburn the T-2 specimens are referred to the fossil (1956). The measurements considered di- species. Milstead (1956, p. 164) referred a agnostic by Pyburn are not available for number of specimens collected from the the Dallas specimen, but over-all size of the Friesenhahn Cavern in Bexar County to Dallas and Ingleside specimens is com-

T. canaliculata, saying: "The largest T. parable. Pyburn assigned the Ingleside Carolina carapace examined was 162.0mm material to G. crassiscutata Leidy. in length; the largest fossil carapace was A complete specimen was collected sev- 208.0 mm." Both specimens collected from eral years ago by Mr. Thomas Williams the Upper Shuler are almost 300 mm in from the Hill-Shuler contact and sent to length. Dr. Thomas Oelrich, University of Michi- This has not been identified, TERRAPENE cf. T. CAROLINA (Linnaeus) gan. specimen but after examining photographs and specimens.—Complete cara- Referred measurements of the Dallas specimen (SMUMP 60112) ; Lower pace Shuler, (SMUMP 60028), Oelrich stated that ap- Moore Isolated single plates. pit. parently both belong to the same species. Discussion.—The provisionally material Both specimens are almost 4 feet long. referred to this species falls within the size range of the living form, commonly very Order CROCODILIA near the maximum. It is particularly note- Family CROCODYLIDAE not a worthy that single specimen of the ALLIGATOR MISSISSIPPIENSIS (Daudin) extremely large size that occurs in the specimens.—Fragment of left Upper Shuler has been recovered from the Referred maxilla withfour teeth and alveoli of three Hill or Lower Shuler members. Because (SMUMP 60059) ; Lower Shuler, this is the most northern report of T. ca- more 60276); naliculata, it may be that this species Moore pit. Single tooth (SMUMP moved into the Dallas area from the south Hill, Coppel locality. only after the Dallas area became rela- Discussion. Specimen (SMUMP tively arid, a condition already suggested 60059) was found to be identical in size for Upper Shuler time. with the skull of a Recent specimen whose body length was almost 10 feet. Specimen Family TESTUDINIDAE (SMUMP 60276) is considerably larger GEOCHELONE sp. and, according to Dr. Walter Auffenburg PI. I, fig. 4 (personal communication to Slaughter, Although fragments of Geochelone cara- 1959), could have belonged to an individ- pace are found in the three lower members ual perhaps 13 feet long. of the T-2 fill, the fragments are much less The association of the aquatic snail abundant in the Upper Shuler, only two Physa with specimen (SMUMP 60059) in- specimenshaving been collected therefrom. dicates that the part of the Lower Shuler in These Upper Shuler fragments are some- which these specimens were found was de- Hill-Shuler Local Faunas 39 posited in a pond on the flood plain of the Order GALLIFORMES ancient Trinity River system. Family TETRAONIDAE Order SQUAMATA TYMPANUCHUS cf. T. PALLIDICINCTUS (Ridgeway) Family COLUBRIDAE Referred specimen.—Distal end of tar- sal-metatarsal (SMUMP 60294); Lower ?COLUBER CONSTRICTOR Linnaeus Shuler, Coppel locality. —Vertebrae (DPSC Referred specimens. Discussion.—Although this species is Upper 1020). Shuler, Lewisville site. not found in the area today, its local ex- Discussion. —These specimens were sub- tinction was probably caused by hunting mitted for identification to Dr. H. G. pressure of man. It has been recorded his- Dowling, University of Arkansas. Dowling torically less than 150miles west of Dallas. stated (personal communication to Slaugh- ter, 1959) that because the vertebrae were Order PASSERIFORMES somewhat damaged he could not say posi- Family CORVIDAE tively whether they belonged to Coluber constrictor or to Masticophis flagellum. CORVUS BRACHYRHNCHOS Brehm Referred specimen.—Tarsal-metatarsal Class AMPHIBIA (DPSC 475) ; Lower Shuler, Moore pit. Order ANURA Discussion. —The common crow is abun- Unidentifiable frog bones have been col- dant in the area today. lected from the Lower Shuler of the Moore pit and the Upper Shuler of the Lewisville Order STRIGIFORMES site and Hickory Creek locality. Family TYTONIDAE

TYTO ALBA (Scopoli) Class AVES —Distal end of tar- Only four identifiable specimens of birds Referred specimen. 620) Lower Shu- have been recovered from the T-2 depo- sal-metatarsal (DPSC ; sits; all are from the Lower Shuler mem- ler, Coppel locality. ber. Crow and prairie chicken were identi- Discussion. —This species is present in fied by Mr. Norman Ford under the direc- the area today. tion of Dr. H. B. Tordoff, University of Michigan, barn owl by Dr. Tordoff, and Class PISCES eagle by Dr. Pierce Brodkorb, University Order LEPISOSTEIFORMES of Florida. Family LEPISOSTEIDAE The presence of a turkey- or crane-size (Lacepede) bird in the Upper Shuler of the Lewisville LEPISOSTEUS cf. L. SPATULA site was indicated by burned egg shell ad- A number of scales of this genus were hering to fossil bone. collected from the Lower Shuler member in the Moore pit. Perhaps more than one Order FALCONIFORMES species is represented, but only one speci- Family ACCIPITRIDAE men is identified specifically. This was AQUILA CHRYSAETOS (Linnaeus) done on the basis of size, because the re- Referred specimen.—Tarsal-metatarsal ferred specimen is beyond the size range of but L. (SMUMP 60293) ; Lower Shuler, Wood all spatula. pit. A report has been made by Uyeno and Discussion.—This eagle is very rare in Miller (1962) on Lower Shuler fish ma- the area today. terial submitted to them for detailed study. 40 Report of Investigations—No. 48

INVERTEBRATE FOSSILS FROM depth of over 20 feet. Thenest was burned, THE T-2 DEPOSITS as were associated bones of several extinct vertebrates. Phylum ARTHROPODA Class INSECTA PLANT FOSSILS FROM THE T-2 DEPOSITS One of the many clay balls in the sand of the Lower Shuler member at the Roy Sincere gratitude is expressed to Dr. Moore pit was sliced quite by accident dur- John Grayson, Socony Mobil Research ing excavation of a fossiliferous zone; it Laboratory, Dallas, Texas, for his patience contained a coprolite as a nucleus. The in running four samples for pollen before clay ball was in situ and the material with- a satisfactory sample was found. Unfor- in did not smoke or char upon ignition. tunately, Dr. Grayson has not yet made a The coprolite appears to have been re- detailed analysis, but the general identifi- placed by non-organic material except for cations are most interesting. Pine pollen hard parts of insects. The specimen was is by far the most abundant, but hickory, submitted to Dr. J. F. Gates Clark of the oak, and evening primrose are also repre- Division of Insects, Smithsonian Institu- sented. Pine trees are not native to the tion. Because of the fragmentary condition area today, and one would have to travel of the individual insect specimens, identifi- approximately 100 miles downstream to cations were necessarily rather general and encounter native stands. These closest therefore offer no help in deducing cli- pines are located in an area averaging over matic conditions. The occurrence is quite 5 inches more annual rainfall than the interesting, however, because Pleistocene Dallas area. The sample was collected from

fossil insects are rare. the Lower Shuler (Wood pit) ; this rein- Dr. Clark's identifications follow: forces the hypothesis that Lower Shuler climate was more moist than the climate Class Insecta today. It is doubly that the Order Coleoptera significant pine Family Carabidae spores dominate the sample and therefore Family Elateridae probably do not represent remnant stands Beetles some to Family Chrysomelidae of pine. The reference of spores Order Hymenoptera the "hickory family" may also be indica- Ants, bees, wasps, etc. tive, because these trees have not been Order Hemiptera Stink bugs, leaf bugs, etc. observed in the Dallas area by the writers, Order Orthoptera but they do occur in some abundance east Cockroaches, crickets, etc. of Dallas in the pine-hardwood forests. Dr. Class Myriapoda Order Diplopoda Grayson stated (personal communication Millipedes, centipedes to Slaughter, 1959) that he also observed in the sample the representatives of grass Because millipedes contain stink glands, families and Compositae (daisy), which many Inseetivora do not eat them. Milli- make the bulk of the diet pedes are often included, however, in the up of grazers. diet of the armadillo. The apparent size of This flora, especially the pine, is similar the coprolite suggests that it belonged to to that of the Jinglebob flora in Kansas. one of the large fossil armadillos in the Pine trees are not present in the area of terrace deposits. the Jinglebob local fauna today but are In addition to its occurrence in the found in the Sangamon deposits. coprolite, the order Coleoptera also occurs The Upper Shuler flora is represented by in the Upper Shuler in a dirt dauber nest numerous fossilized hackberry seeds col- found in situ at the Lewisville site. The lected at the Lewisville site and Hickory nest was collected from a burned area at a Creek locality. Part of a pecan shell was Hill-ShulerLocal Faunas 41 found at Hickory Creek beneath an exca- drouth along draws and other waterways, vated carapace of Terrapene canaliculata. the presence of this element does not ne- Because hackberry trees are very hardy gate the indicated arid climate for the and are able to survive long periods of Upper Shuler. MOLLUSCAN FAUNA FROM THE LOWER SHULER MEMBER, T-2 DEPOSITS

Prior to 1959, little was known of the from calcium-charged water from nearby molluscan fauna of the T-2 deposits. A native limestone. In screenings, organic small collection from the Lewisville site debris was meager and consisted primarily (Upper Shuler) was identified as follows of small plant rootlets. by Dr. Elmer P. Cheatum, Southern Meth- The contact between the snail-bearing odist University (Crook and Harris, 1957, clay fill and the laminated sands is quite pp. 57-58): sharp, both above and below the fill, with little intergradation. There is some local Class Gastropoda— discoloring caused by precipitation of iron (Rockett) Stenotrema monodon —abundant compounds. Lack of stratification be Stenotrema monodon aliciae (Pilsbry) = can Stenotrema leai (Binney)—abundant attributed to several causes, wave action Anguispira alternata (Say)—abundant and reworking by organisms probably Polygyra cf. texasiana (Moricand) —rare Mesomphix sp.—abundant being primary ones. Class Pelecypoda— Although the fossils contained in the Fusconia cf. undata (Barnes) snail-bearing fill their Quadrula frustnlosa (Lea) clay are abundant, Amblema [—Crenodonta] plicata (Ortmann tests are very fragile and many were and Walker) destroyed in collecting. In order to make the faunal list as In 1957, several pelecypods collected by as complete possible, samples of the were taken all Slaughter from the Lower Shuler of the matrix from Moore pit were submitted for identifica- parts of the exposure and were repeatedly washed and sifted. vertebrate tion to Dr. Joseph P. E. Morrison, Smith- Only two fossils the one sonian Institution. He identified the fol- appeared in collection: tooth lowing: of Lepus sp. and one of A sinus conver- sidens (Owen). Crenodonta perplicata (Conrad) In listing the molluscan fauna, an at- Quadrula speciosa (Lea) tempt has been made to shorten and sim- Truncilla truncata (Rafinesque) plify descriptions, yet to retain as much pertinent information as is readily avail- WOOD PIT POND DEPOSIT able. Descriptions include the following reference points: Early in 1959, an ancient pond clay fill containing fossil snails was located in the (!) Habitat, aquatic or terrestrial. T-2 deposits at Wood pit. Fortunately, (2) Relative abundance of species, indicated several feet by rare, sparse, average, or abundant. Be- of the typical laminated sands cause quantitative analysis has not been of the Lower Shuler are displayed both made, these indications must be con- above and below the snail-bearing clay fill. sidered truly general. (3) Stratigraphic range. The pond-fill matrix, averaging 6 feet in (4) Present distribution (North America). thickness, lacks stratification in its com- (5) Ecology. plete exposure. It has vertical and hori- (6) Remarks. zontal continuity with little variation in Phylum MOLLUSCA lithology and faunal distribution. The light Class GASTROPODA gray, highly calcareous fill is in- slightly Family PLANORBIDAE durated and is very fine sand, silt, and GYRAULUS PARVUS (Say) clay. Calcium carbonate amounts to ap- proximately 10 percent by volume. Much Habitat and relative abundance.— of the calcium carbonate is contributed by Aquatic, average. minute particles of snail tests and calcium Stratigraphic range.—Middle Pliocene- precipitates, no doubt partially derived to Recent. Hill—Shuler Local Faunas 43

Present distribution.—"Eastern North Stratigraphic range.—Lower Pliocene America east of the Rocky Mountains from to Recent. Helisoma anceps [=H. antrosa Florida north to Alaska and northern (Conrad)] is wide-ranging and has been British America'' (Baker, 1928, p. 377). identified in Lower Pliocene (Laverne As pointed out by Taylor (in Hibbard and formation, Kansas), Aftonian, and Yar- Taylor, 1960, p. 100), this distribution mouth deposits and, rarely, in lllinoian given by Baker is based upon a relatively (Crete-Loveland, Kansas) sediments (Leo- narrow concept of the species. Taylor also nard, 1952, p. 21). stated: ''Probably other forms are identi- Present distribution.—Maine to Oregon cal with G. parvus and the distribution and from Hudson's Bay to Mexico. This is includes almost all North America." This a prominent species in the Dallas County species is abundant in Dallas County today. Recent fauna. Ecology.—G. parvus commonly lives in Ecology. —Helisoma anceps inhabits quiet pools and stream beds and also in streams, ponds, and short-lived pools. This larger lakes on partly water-logged pieces species commonly does not live in large of driftwood that have been washed up lakes. along shore (Leonard, 1959, p. 61). Material.—SMUP 460. Material.—SMUP 469. 3 HELISOMA TRIVOLVIS (Say) GYRAULUS CIRCUMSTRIATUS (Tryon) Habitat and relative abundance.-— Habitat and relative abundance.— Aquatic, more abundant than H. anceps. Aquatic, more abundant than G. parvus. Stratigraphic range.—Nebraskan or Stratigraphic range.—Kansan to Re- Aftonian (Dixon local fauna, Kingman cent. County, Kansas) to Recent. Present distribution.—"G. circumstria- Present distribution.—Atlantic Coast tus is found in a broad belt across central and Mississippi River drainages, north to North America, in southern Canada and Arctic British America, south to Tennes- the northern United States between the see, Missouri, and Kansas. The southern Atlantic and Pacific Oceans, southward in and western limits of distribution are not the Rocky Mountains to northern Arizona" clearbecause of uncertain systematic status (Hibbard and Taylor, 1960, pp. 96-97). of the species with respect to its races Because of its confusion with similar (Leonard, 1950, p. 16). Helisoma trivolvis species, such as G. parvus, the exact north- lentum (Say) is a prominent member of ern and southern limits are not known. Dallas County Recent fauna. The relatively Ecology.—G. circumstriatus is more small number and fragile condition of T-2 limited in distribution and environment deposit specimens prevent a positive com- than G. parvus; in places they are associ- parison with Recent specimens. ated. G. circumstriatus is more common in trivolvis inhabits small seasonable bodies of water whereas Ecology.—Helisoma waters. Although G. parvus prefers rivers and lakes. An ex- quiet, somewhat stagnant cellent description of the two species and it does live in permanent bodies of water, even a comparison of their differences in eco- it flourishes in ponds or sloughs, with vegeta- logical preference are given by Hibbard though they may be choked and Taylor (1960, pp. 96—101). tion or polluted with decaying organic Material.—SMUP 474. materials. It is invariably absent in flowing streams (Leonard, 1950, p. 16). HELISOMA ANCEPS (Menke) Material.—SMUP 472. Habitat and relative abundance.— (Say) Aquatic, sparse. PLANORBULA ARMIGERA relative abundance.—- ³ Habitat and Catalog numbers in this section of the paper are those of Southern Methodist University Biology Department, where the Aquatic, average. T-2 molluscan collections are housed. 44 Report of Investigations—No. 48

Stratigraphic range.-—Pleistocene to Re- Pleistocene deposits in Hardeman, How- cent ; not perfectly known. ard, Hall, and Wichita counties, Texas. Present distribution.—"New England Present distribution.—Eastern Quebec, west to Nebraska, south to Georgia and Nova Scotia, and New Jersey, west to Van- Louisiana, north to Great Slave Lake" couver Island; Manitoba, south to Ala- (Baker, 1928, p. 359). bama and Texas and west to Arizona and Ecology.—Planorbula armigera lives in southern California. small stagnant bodies of water. Ecology. —This species lives in quiet Material.—SMUP 470. ponds and ephemeral pools. Material.—SMUP 465. Family LYMNAEIDAE LYMNAEA CAPERATA Say LYMNAEA OBRUSSA Say Habitat and relative abundance.— Habitat and relative abundance.— Aquatic, sparse. Aquatic, average. Stratigraphic range. —Middle Pliocene Stratigraphic range.—-Nebraskan to Re- to Recent. cent. Present distribution.—"From Quebec Present distribution.—"From the At- and Massachusetts west to California; lantic to the Pacific oceans, and from Mac- Yukon Territory and James Bay south to kenzie Territory, Canada, south to Arizona Maryland, Indiana, Colorado and Cali- and northern Mexico" (Baker, 1928, p. fornia" (Baker, 1928, p. 263). 296). Ecology.—Lymnaea caperata lives in Ecology.—Lymnaea obrussa lives in intermittent streams or small ponds which shallow water on mud bottoms and among often become seasonably dry. The living lake and pond vegetation. species is probably restricted to a northern Material.—SMUP 479. distribution because of the species' appar- Family PHYSIDAE ent inability to withstand southern tem- APLEXA HYPNORUM (Linnaeus) perature highs. Material.—SMUP 480. Habitat and relative abundance. — Aquatic, sparse. LYMNAEA DALLI Baker Stratigraphic range.—Nebraskan to Re- Habitat and relative abundance.— cent. Aquatic, rare. Present distribution.—. . from east of Stratigraphic range.—Early Pliocene to the Cascade Mountains to the Atlantic and Recent. from Alaska and Hudson Bay south to the Present distribution.—"Ohio to north- vicinity of the Ohio River" (Baker, 1928, ern Michigan and Montana, south to Kan- p. 474). Ecology.—"lt is especially abundant in sas and Arizona" (Baker, 1928, p. 288). woodland pools which become dry in sum- Ecology.—"Lymnaea dalli is often ." . mer . (Baker, 1928, p. 474). found in moist shaded places a few or Material.—SMUP 482. centimeters away from the edge of water on wet surfaces of mud, stones or sticks" PHYSA GYRINA Say (Leonard, 1959, p. 56). Habitat and relative abundance.— Material.—SMUP 481. Aquatic, most abundant species in the pond site. LYMNAEA HUMILIS MODICELLA (Say) Stratigraphic range.—Yarmouth to Re- Habitat and relative abundance.— cent. Physa gyrina lives in large numbers Aquatic, average. in sinkhole deposits of southwestern Kan- Stratigraphic range.—Pleistocene to Re- sas and northwestern Oklahoma. The sedi- cent. This species has been observed in ments, while not accurately dated, are ob- Hill—Shuler Local Faunas 45 viously younger than Yarmouth (Leonard, abundant in lower zones of the Peoria silt 1950, p. 22). in Kansas (Leonard, 1950, p. 23). Present distribution.—P. gyrina, ac- Present distribution.—S. avara has been cording to Baker (1928, p. 452), ranges reported from every State in the continen- from the Arctic region south to Alabama tal United States (excluding Alaska), but and Texas. A study of Recent collections in view of the difficulty of identifying in Texas (Cheatum, MS) has failed to species of Succinea from shells alone, are identify living specimens of P. gyrina, and many records undoubtedly erroneous. it is quite possible that Baker's Texas shells Ecology. —Habitat preferences of small came from Pleistocene deposits. Many succineids vary from lowland swampy north-central Texas river-terrace deposits areas to well-drained forest areas where ranging in age from Kansan to Sangamon, they live under leaves, logs, and debris. which are now being studied by the Material.—SMUP 458. authors, contain fossil shells of P. gyrina or P. gyrina form hildrethiana. Taylor (in Family BULIMULIDAE Hibbard and Taylor, 1960, p. 118) re- BULIMULUS DEALBATUS (Say) ported: "The species [P. gyrina form Habitat and relative abundance.—Ter- hildrethiana ] has not been found living in restrial, rare, one specimen. Meade County and may not occur in the Stratigraphic range.—Unknown. southern Great Plains except as fossils." Present distribution.—Illinois, Tennes- The distribution of this species today may see, Alabama, Kentucky, Missouri, Arkan- likely be east of the Mississippi River, sas, Texas, Oklahoma, and Louisiana. This north and east of the mid-continentregion, is one of the most abundant species in Dal- and as far south as Alabama. las County today. Ecology.—P. gyrina inhabits stagnant, Ecology. —This hardy species lives in ephemeral pools and also lives on mud or bushes and grass and can prosper in arid among stones in a stream. In the Wood pit regions. It often hibernates a few inches pond deposit, P. gyrina appears in such beneath ground surface. concentrated numbers that a shallow-pond Material.—SMUP 451. environment can easily be visualized. Remarks.—There is an unusual variety Family CARYCHIIDAE of sizes in the T-2 deposit specimens, which CARYCHIUM EXILE H. C. Lea range in length from to mm. 4 17 Although Habitat and relative abundance.—Ter- the shells are very fragile, the luster of the surface is well preserved; most of the restrial, sparse. —Pleistocene Re- specimens resemble polished ivory in tex- Stratigraphic range. to cent. ture. Among individual specimens there is some variation in length of the spire, but Present distribution.—Maine and On- most of the spires are very short, the tario to Manitoba, south to Mobile, Ala- sutures are not deeply impressed, and all bama, and Texas (Leonard, 1959, p. 195). of the specimens have the typical elliptical Ecology.—This species has a preference shape. for marshy areas around ponds and creeks Material.—SMUP 459. and lives under logs or sticks or partially submerged in water. Family SUCCINEIDAE Remarks.—Taylor (in Hibbard and SUCCINEA cf. S. AVARA Say Taylor, 1960, p. 85) has suggested that C. Habitat and relative abundance.- exile is a synonym of C. exiguum. The Terrestrial, average. specimens of C. exile from the T-2 de- Stratigraphic range.—Yarmouth to Re- posits, which are few in number probably cent. Shells referred to this species are because the minute size of the shell causes 46 Report of Investigations—No. 48 its loss through screen processes, have Mexico and in Central America (Leonard, radial striae. 1959, p. 122). Material.—SMUP 468 Ecology. I—This is a woodland species and although it is widely distributed Family ZONITIDAE throughout the United States, it is absent MINUSCULA (Binney) HAWAIIA from large areas not having protective tim- Habitat and relative abundance.—Ter- ber. Z. arboreus prefers moist situations restrial, sparse. and lives under logs or among dead leaves. Stratigraphic range.—Lower Pliocene Material.—SMUP 475. to Recent. Family PUPILLIDAE Present distribution.—From Alaska to Maine, Costa Rica, Cuba, Santo Domingo, GASTROCOPTA ARMIFERA (Say) Puerto Rico, Bermuda, and Japan. In the Habitat and relative abundance.—Ter- United States, this species is found from restrial, rare. Maine to Florida and Texas, west to the Stratigraphic range.—Lower Pliocene Rocky Mountains. (Laverne, Kansas) to Recent (Taylor, Ecology.-—H. minuscula lives in leaf 1954, p. 4). mold and under logs, stones, and the loose Present distribution.—Eastern United bark of trees. In spite of its preferred habi- States and Canada: Quebec to northern tat, this species can withstand drouth and Florida, west to Red Deer, Alberta, the high temperatures. Dakotas, Boulder, Colorado, Lincoln Material.—SMUP 461. County, , and the mouth of the Pecos River, Texas (Pilsbry, 1948, p. RETINELLA cf. R. IDENTATA (Say) 875). Habitat and relative abundance.—Ter- Ecology. —G. armifera "is limited to rare. restrial, timber or brushy situations along stream Stratigraphic range.—Unknown. courses or thick grasses nearby. It thrives Present distribution.—"North America best in limestone districts, where it may from Canada (46° N. Lat.) south to occur in dense populations. The species is northern Alabama; from Maine west to found in leaf mold, dead leaves and grass, Kansas" (Leonard, 113). 1959, p. or beneath stones" (Leonard, 1950, p. 29). Ecology.-—This species lives under and Material.—SMUP 466. in rotting logs, leaves, and other forest debris in both upland and flood plain situ- GASTROCOPTA CONTRACTA (Say) ations (Leonard, 1959, p. 114). Habitat and relative abundance.—Ter- Remarks.—The smaller size and more restrial, rare. narrow of the T-2 species were Stratigraphic range.—Yarmouth to Re- used to distinguish it from the locally cent (Leonard, 1950,p. 30). abundant Recent subspecies Retinella Present distribution.—Northern United identata paucilirata. Too few specimens States, Ontario, Canada, and eastern Mex- were recovered for positive identification. ico. Material.—SMUP 477. Ecology.—Gastrocopta contracta lives in large numbers under chips and stones ARBOREUS (Say) and in rotting logs and debris on the edge Habitat and relative abundance.-—Ter- of forests (Baker, 1902, p. 236). restrial, sparse. Material.—SMUP 473. Stratigraphic range.—Early Pliocene to Recent. GASTROCOPTA PENTODON (Say) Present distribution.—Z. arboreus is Habitat and relative abundance.—Ter- widely distributed over the North Ameri- restrial, abundant. can continent from northern Canada to Stratigraphic range.—Unknown. Hill—ShulerLocal Faunas 47

Present distribution.—Eastern United Stratigraphic range.—Unknown. States and Canada, west to New Mexico Present distribution.—East Texas, Okla- and Arizona, south to Texas, and Florida. homa, Louisiana, and Arkansas. Ecology.—This species lives on wooded Ecology.—This species lives among leaf hillsides, in well-drained groves, and mold and on decaying logs and thrives in among leaves and grass in underbrush. moderately humid forest environment. Remarks.—The specimens of G. pento- Remarks.—The first specimens collected don from the T-2 deposits measure almost from the Wood pit pond site are difficult to uniformly 1.8 mm in length. distinguish from Strobilops sparsicosta Material.—SMUP 457. Baker. The bases of the shells are nearly smooth, GASTROCOPTA PROCERA (Gould) with but few faint riblets. The parietal lamella emerge to the edge of the Habitat and relative abundance.—Ter- callus and extend far into the shell, where restrial, sparse. there are other complex lamella. The Stro- —Early Pleistocene Stratigraphic range. bilops collected from the stream site (p. to Recent. The geologically oldest occur- 48) bear the typical lamella and oblique the Nebraskan or Aftonian rence is in ribbing of S. texasiana. Dixon local County, Kan- fauna, Kingman Material.—SMUP 456. sas (Hibbard and Taylor, 1960, p. 123). Present distribution.—Eastern United Family ENDODONTIDAE States from Maryland to South Carolina, HELICODISCUS PARALLELUS (Say) west to and Texas Kansas, east (Leonard, Habitat and relative abundance.—Ter- 1959, p. 179). restrial, average. Ecology. —This lives near species Stratigraphic range.—Kansan (Cu- streams, under the loose bark of trees, and dahy) to Recent (Taylor, 1954, p. 5). under and and can withstand logs stones, Present distribution.—Eastern Mexico periods of drouth. to Newfoundland. Material.—SMUP 476. Ecology—H. parallelus lives upon de- PUPOIDES ALBILABRIS (C. B. Adams) caying wood in shady or humid places and Habitat and relative abundance.—Ter- among damp leaves (Pilsbry, 1948, p. restrial, sparse. 627). Stratigraphic range.—Lower Pliocene to Remarks.—The distribution in Texas of

Recent (Leonard, 1950, p. 29) ; through- H. parallelus is not perfectly known. Sub- out the late Kansan of the mid-continent Recent shells have been collected from region, Kansas, Oklahoma, Texas, and Foard County by Dr. Walter Dalquest, Nebraska. Midwestern University. The species oc- Present distribution.—Maine to On- curs sparsely in a sub-Recent deposit in tario, Canada, south to the Gulf of Mexico, Dallas County. west to the Dakotas and Arizona; West Material.—SMUP 464. Indies (Leonard, 1950,p. 29). Ecology. —This hardy species thrives in Family HELICINIDAE under woodlands, leaf mold, or in the HELICINA ORBICULATA TROPICA Pfeiffer loose bark of trees, but it also inhabits un- Habitat and relative abundance.—Ter- shaded fields of short grass. Material.—SMUP 453. restrial, rare, two specimens. Stratigraphic range.—Unknown. This Family STROBILOPSIDAE is an abundant species in east Texas today STROBILOPS cf. S. TEXASIANA (Pilsbry and Ferris) and was reported by Sellards (1940, p. Habitat and relative abundance.—Ter- 1637) from the Berclair terrace deposit in restrial, sparse. Bee County. 48 Report of Investigations—No. 48

Present distribution.—Arkansas, Okla- Present distribution.—lowa, Kansas, homa, Texas, and Mexico. Missouri, Arkansas, Oklahoma, Texas, Ecology.—Helicina orbiculata tropica Louisiana, Mississippi, Alabama, Tennes- apparently has great resistance to drouth, see, Illinois, Indiana, Virginia, Maryland, for it lives at great distance from water. It District of Columbia (Pilsbry, 1940, p. has been observed on trees at heights of 5 680). to 6 feet (Cheatum, MS; Pilsbry, 1948, pp. Ecology.—Stenotrema leai is a wood- 1084, 1085). This is one of the most abun- land or woodland border species; it thrives dant species of the Dallas area today; it under leaf mold, fallen trees, and stones. thrives in open fields among sparse vege- Material.—SMUP 454. tation. Material.-—SMUP 452. Class PELECYPODA Family SPHAERIIDAE Family POLYGYRIDAE SPHAERIUM STRIATINUM (Lamarck) MESODON THYROIDUS (Say) Habitat and relative abundance.—- Habitat and relative abundance.—Ter- Aquatic, sparse. restrial, sparse. Stratigraphic range.—Middle Pliocene Stratigraphic range.—Unknown. to Recent. Present distribution.—Eastern United Present distribution.—New England States and Canada, west to eastern Nebras- west to California, Canada south to Loui- ka and Texas; distributed throughout the siana, and Florida. wooded areas of Dallas County. Ecology.—S. striatinum inhabits small Ecology.—Mesodon thyroidus is a true creeks, ponds, and rivers where it may be woodland species, which prefers humid buried in black slimy mud to a depth of localities. several inches (Baker, 1928, p. 115). Tay- Material.—SMUP 455. lor {in Hibbard and Taylor, 1960, p. 76) POLYGYRA TEXASIANA (Moricand) stated: "Perennial water bodies with Habitat and relative abundance.—Ter- some current action are suitable habitats. restrial, rare, only two specimens. The species lives in large lakes or small Stratigraphic range.—Yarmouth to Re- ones, in rivers or small streams, in a bot- cent. tom of gravel, sand or mud." Present distribution.—Illinois, Tennes- Material.—-SMUP 467. see, Alabama, Kentucky, Missouri, Arkan- sas, Texas, Oklahoma, and Louisiana. This WOOD PIT CHANNEL DEPOSIT is one of the most abundant species of Late in a second area, apparently Dallas County today. 1959, an ancient stream bed, was excavated ap- Ecology.—This hardy species lives in proximately 200 yards to the northwest bushes and grass and can prosper in arid of the first locality. Although the new area areas. It often hibernates a few inches deep is vertically aligned with the pond de- in the ground. and the same laminated Material.—SMUP 463. posit displays sands of the Lower Shuler above and be- STENOTREMA LEAI (Binney) low the newly exposed clay, at no place is Habitat and relative abundance.—Ter- the snail-bearing clay fill thicker than 18 restrial, abundant. inches. The exposure was made by com- Stratigraphic range. Middle Pleisto- mercial gravel companies and is very ir- be cene (Kansan) to Recent. The first re- regular; the over-all dimensions cannot corded appearance is in the Cudahy fauna, determined at this time. At one place, 20+ Kansas and Oklahoma (Hibbard and Tay- feet exposure with little variation in thick- lor, 1960, p. 150). ness of the clay can be seen. The silty and Hill—Shuler Local Faunas 49 sandy clay is gray and highly calcareous. ence in the pond of permanent water spe- The snail fauna of the Wood pit channel cies, such as Gyraulus parvus, Planorbula deposit, collected by bulk sampling and armigera, Aplexa hypnorum, Lymnaea washing, is as follows: caperata, L. humilis modicella, Physa gy- trivolvis ruia, and Helisoma , suggests that Gastrocopta procera (Gould) —rare, two speci- mens (SMUP 208) they lived in a flood-plain pond, near a Bulimulus dealbatus (Say)—sparse, more wooded area, which was temporarily out of abundant than at pond site (SMUP 201) Hawaiia minuscula (Binney)—more abundant reach of river deposition. Physa gyrina is than at pond site (SMUP'2O2) found abundantly in stagnant-water en- Pupoides albilabris (C. B. Adams) (SMUP 203) vironments. The heavy concentration of Polygyra texasiana (Moricand) —rare, three this species in the pond deposit along with specimens (SMUP 204) the supporting species that thrive under Strobilops cf. S. texasiana (Pilsbry and Ferris) (SMUP 205) semi-stagnant water conditions, indicates Retinella cf. R. identata (Say)—rare, one that the pond at times was reduced to a specimen (SMUP 20T) shallow pool which may have become dry Helicina orbiculata tropica (Pfeiffer) —rare, one specimen ('SMUP 206) seasonally. Although the pond was short Pupisoma cf. P. dioscoricola (C. B. Adams) lived, or at least did not last long enough rare (SMUP 209) to become a swamp or marsh, it was in ex- The only species that occurs in the chan- istence long enough for many generations nel but not the pond deposit is described of snails to accumulate. The sand, silt, and below. land-snail tests that are a large part of the pond-clay deposit probably were brought Family PUPILLIDAE in by surface runoff from higher ground PUPISOMA cf. P. DIOSCORICOLA (C. Adams) B. and by small streams. Vertical range.—Unknown. The narrow band of clay comprising the Present distribution.—Florida, south channel deposit contains much more sand Texas (Brownsville), West Indies, Central and silt than does the pond fill and con- and South America (Pilsbry, 1948, p. tains none of the aquatic Mollusca. The 1008). channel deposit is the record of a stream Ecology.—P. dioscoricola lives on the that apparently supplied the pond with bark of trees and on plants. Its tropical runoff water only during moist seasons. habitat today denotes its preference for The shifting meander belt of the Trinity humid environment. River eventually filled the pond and buried Material.—SMUP 209. it beneath flood-plain deposits of sand and silt. DISCUSSION OF MOLLUSCA FROM T-2 DEPOSITS SUB-RECENT COMPARATIVE MOLLUSCAN Most of the molluscs collected from the FAUNA T-2 deposits were species normally found To illustrate the difference in climate in ponds and streams in moist, woodland between sub-Recent time (deposition of the areas. The variation of the collections last 6,000 years) and the time of deposi- from the two Wood pit sites, only 200 yards tion of the T-2 deposits, a faunal compari- an apart, is excellent demonstration of son is made with sub-Recent small creek the importance of considering the ecology terrace (p. 51) molluscan fauna in Dallas of any fossil locality. The channel deposit County. contained one species not found in the The small creek terrace molluscan pond site; relative abundance of species fauna of Dallas is as follows: was markedly different in the two sites. Screenings from the channel deposit Gyraulus parvus (Say) (SMUR 113) Helisoma anceps (Menke) —abundant (SMUR yielded many fewer specimens. The pres- 101) 50 Report of Investigations—No. 48

Helisoma trivolvis (Say) (SMUR 102) Early Wisconsin must have been more 110) Lymnaea bulimoides (Lea) (SMUR stable and lacked the extreme highs Amnicola intrega (Say)—abundant (SMUR than, 112) in temperature of, the climate of north- Physa anatina (Lea) —abundant (SMUR'IO9) central Texas today. The Sangamon inter- Ferrissia rivularis (Say) (SMUR 120) Bulimulus dealbatus (Say)—abundant (SMUR val of the Great Plains region was de- 104) scribed by Frye and Leonard (1957, p. (Binney) (SMUR 118) Huwaiia minuscula 10) as more moist than at present in that Succinea cf. S. grosvenori (Lea) (SMUR 117) Gastrocopta procera cf. G. p. sterkiana (Pils- region. Faunal evidence shows this to be bry) (SMUR 111) equally true during the deposition of at Gastrocopta contracta (Say) (SMUR 114) at Pupoides albilabris (C. B. Adams) (SMUR least the first part of the T-2 deposits 107) Dallas, because the T-2 molluscan fauna Strobilops texasiana (Pilsbry and Ferris) indicates a more moist climate at the time (SMUR 106) Euconulus fulvus (Miiller) (SMUR 116) of deposition. Helicodiscws parallelus (Say) (SMUR 115) Frye and Leonard also pointed out that Helicina orbiculata tropica Pfeiffer—abundant many molluscan species found today at (SMUR 105) Mesodon thyroidus (Say) (SMUR 103) higher and cooler latitudes or altitudes Polygyra texasiana (Moricand) —abundant were widespread throughout the Great (SMUR 100) Plains regions and in the southern High Pisidium nitidum Jenyns (SMUR 119) Sphaerium striatinum (Lamarck) —abundant Plains from Kansan to Early Wisconsin, (SMUR 108) only to disappear from these regions dur- ing Late Wisconsin and Recent. Many of Wherever the sub-Recent and Recent the wide-ranging species identified from Dallas County molluscan faunas are en- the T-2 deposits live today in north-central countered, by far the most abundant ter- Texas under varied ecological conditions. restrial species are Bulimulus dealbatus, Some, however, are Recent inhabitants of Polygyra texasiana, Helicina orbiculata, areas quite distinct from the area under and Mesodon thyroidus. These species are study. Physa gyrina, Planorbula armigera, among the most rare in the T-2 deposits; Lymnaea caperata, and Aplexa hypnorum because they are hardy species, it is quite are now confined to an area north and east easy to see how they could have become of the mid-continent region. Pupisoma relatively more abundant in Recent time as dioscoricola has been recorded in Texas summer temperatures became more severe in Recent time only from the Brownsville and the climate less stable. area at the mouth of the Rio Grande, The climatic interval between the San- where the relative humidity exceeds that of gamon (last major interglacial) and the Dallas County. FOSSIL MAN IN ALLUVIAL TERRACES OF THE UPPER TRINITY RIVER

This section summarizes evidence of The Carrollton focus is in the top 24 to ancient man associated with the various 30 inches of the red Albritton clay and the terrace deposits of the Trinity River at basal 6 to 8 inches of the overlying Pattillo Dallas; most of the attention is focused on sand. Artifacts collected from this focus the Pemberton Hill-Lewisville, or T-2, are dart points, about 5 percent of which deposits. are Paleo-American types: Plainview, Me- All small creeks in the Dallas area that serve, Dalton, Scottsbluff, and Angostura. are currently flowing on Austin chalk The rest are stemmed or shouldered dart display evidence of two earlier fills. The points of the Carrollton, Trinity, Edge- youngest and lowest of these is generally wood, Yarbrough, and Wells types and a called simply the "small creek" terrace. few central Texas Archaic types. About 70 The older and higher has not been dated percent of the scrapers collected are with any confidence and remains un- gouges of the Clear Fork type, although named. somewhat smaller ; the remainder are side, The small creek terrace materials were Hake, and end scrapers. Other types of locally derivedfrom the Cretaceous forma- tools found include drills, choppers, knives, tions and are lithologically different from gravers, the Carrollton axe, and the so- those of the main river terraces. Arche- called net sinkers, which are probably ac- ological correlation is possible, however, tually atl-atl weights. A few small manos between the T-l deposits of the Trinity and milling stones were also collected. The River and the small creek terrace deposits. Wheeler site, near Carrollton, has pro- Fortunately, the small creek terrace de- duced two rather fragmentary Carrollton posits are not acidic as are those of the T-l focus skulls. Both are keel-vaulted. terrace and therefore better preserve shell In June 1959 a mussel shell collected and bone. from a Carrollton midden was radiocarbon dated at 5945±200 The writers are T-0 DEPOSITS BP. indebted to Dr. C. I. Alexander, Socony A modern fauna occurs in the T-0 de- Mobil Research Laboratory, Dallas, Texas, posits which also contain Neo-American fer determination of this date. Stage materials (pottery, agriculture, and The Elam focus usually occurs at depths arrow points) at the base of the alluvium ranging from 15 to 30 inches below the and Historic Stage artifacts in the upper- surface of the Pattillo sand (Crook and most 6 inches. The same cultures are found Harris, 1952, p. 24). This focus is wholly on the surface of the T-l deposits. Archaic. The artifacts resemble those of the Carrollton focus except for their T-l DEPOSITS smaller size. The net sinkers and Paleo- American points which occur in the Car- The T-l deposits are rises or natural rollton focus have not been found in the levees the Recent flood ve- on plain and later Elam focus. neers upon the eroded slopes of the next Neo-American Stage—Sites of this higher terrace deposits near stream valleys. Stage are usually found within the top 6 to Three cultural stages are associated with these deposits: Archaic Stage, Neo-Ameri- 8 inches of the T-l deposits as well as on can Stage, and Historic Stage. the T-0 terrace. The sites are divided into Archaic Stage—This Stage is divided pre-Gibson, Gibson, and Fulton aspects into the Carrollton focus and the Elam and in turn into Alto, Sanders, Wylie, and focus (Crook, 1952; Crook and Harris, Henrietta foci. This sequence is estimated 1952). to have begun approximately 1,000 A.D. 52 Report of Investigations—No. 48

Historic Stage—Historic sites are un- numbers in local areas, such as the Lewis- common in the Dallas area. One interest- ville site. Interestingly enough, all ma- ing find, however, was a relatively small terial at Lewisville appears to be confined battle-axe discovered on a bluff in southern to the general hearth area of presumed Dallas County. After the rust was removed, human habitation. the blade was found to be covered with a The Lewisville site hearths, artifacts, very ornate design of silver inlays. The and fossil faunal materials occur only in handle concealed another weapon, a screw- the Upper Shuler yellow sandy clay and out dagger. Another interesting find, made are well encrusted with caliche. Heaviest after a flood had scoured the low-lying, concentration of caliche is in the lower locally derived alluvium of the T-0 terrace zones. near Carrollton, was a silver chain-mail Large numbers of hackberry seeds, both gauntlet. European gun-flints and a few burned and unburned, occur within the lead balls have also been collected from hearths. The charcoal from Hearth 1 used Dallas County sites. One site near Kauf- for radiocarbon dating was described man, 30 miles south of Dallas, produced a (Crook and Harris, 1957, p. 69) as dress ornament of tinkler made from sheet "charred, fibrous, vegetable material"— copper. probably grass or sedges of some type. Also radiocarbon the charcoal from T-2 DEPOSITS dated, Hearth 8 was so degenerated that the cellu- The T-2 deposits, the primary concern lar structure necessary for identification of this paper, contain a few traces of early was obscured. The charcoal gave the im- man which have been uncovered during pression of possibly having been cedar. excavation for building materials such as Hickory Creek—This site was dis- gravel, sand, and clay. These Paleo-Ameri- covered after a large pit was opened for can Stage sites are usually in the Upper fill and gravel. Although to date no arti- Shuler about 15 to 20 feet below the sur- facts have been found, two flint flakes and face of the terrace. a large collection of faunal material, in- Lewisville sit—The remains of early cluding some burned bone, have been col- man were uncovered after about 20 feet of miles on T-2 deposits had been removed for fill for lected. The site, 10 upstream Garza Dam. Twenty-one hearths and Hickory Creek from the Lewisville site, is hearth areas were located and excavated in the same stratum of the T-2 deposits as by the Dallas Archeological Society before that site. Evidence of human activity is the new lake covered the site. A Clovis scanty, yet the concentration of faunal ma- point, a scraper, a chopper, a hammer terial around such evidence is reminiscent stone, a few flint flakes, and a large faunal of Lewisville; no fossils have been found collection were obtained. This apparently in the remainder of the pit. represents a camp site of the Llano Cul- The specimen of Elephas columbi exca- ture. A complete description of this site is vated from the Upper Shuler of this site by given by Crook and Harris (1957). North Texas State College personnel is The Lewisville site has been known pri- intriguing in that the skull with tusks was marily an archeological radio- as locality; found lying upside down, although associ- carbon dating gives its age as more than ated with the remainder of the skeleton 37.000 BP (Brannon et al., 1957, p. 149). which its side. Many split It is also one of the major concentra- was lying on and mussel tions of faunal material known from the bone fragments, teeth, even T-2 deposits. It is noteworthy that while shell appear to be burned—as do a few most exposures of the Upper Shuler por- caliche nodules—even though no definite tion of the T-2 deposits lack fossil remains, hearths, like those at the Lewisville site, where fossils do occur they are in great have yet appeared. Hill—ShulerLocal Faunas 53

In addition, a carbonized, possibly was deposited but before deposition of the burned, pecan was found immediately be- veneer of Albritton red sandy clay. neath a carapace fragment of Terrapene Whether this occurred by gully cutting or canaliculata which, in turn, had an appar- intentional burial into the exposed and ently burned hackberry seed cemented to eroded-off slope of the T-2 fill, then com- it by caliche. posed of residual Upper Shuler yellow Lagow pit—A small faunal collection sandy clays, is immaterial; a yellow sandy including part of a human skeleton was ob- clay fill might easily escape detection after tained from this pit. The human and other some thousands of years. animal bones are mineralized. The T-2 The age of Lagow man, therefore, must deposits of the Lagow pit are practically be between 28,000 and 10,000 BP or prior identical with those of the Lewisville site, to the commencement of deposition of the although the two localities are 25 miles Albritton. Most probably the age is closer apart. to 10,000 BP because the erosion and Unfortunately, in Lagow pit no artifacts slope-off must have been essentially com- were found, and the skeletal remains are pleted before emplacement of the remains. incomplete. Results of a recent study, made This is corroborated by the remains of by Harvard University and the British Archaic man that have been recovered else- Museum Fluorine Laboratory, on the La- where from the Albritton clay. It is the gow pit human and animal remains are opinion of Crook (1961) that the remains contained in a paper by Oakley and How- of some type of pre-Archaic Man, acci- ells (1961). It is demonstrated that the dentally or intentionally, were buried in a Lagow pit man is not the same age as the residual surface of eroded Upper Shuler animals found in the T-2 deposits; how- about 10,000 years ago. ever, the human remains are somewhat Seagoville site—What appeared to be a mineralized and seem to be old. The Lagow large Plainview point was discovered in pit human remains are fragmentary but the Richards formation exposed in a gar- do not differ from remains of American den in the city of Seagoville. The point was Indians, except for a very thick skull and about 8 inches below the surface and thick walls of the long bones. Certainly partly covered by a thin caliche coating. there is every evidence that this is Homo No other artifacts have been found at this sapiens. place; this is the only known possible evi- The thickness of the skull of Lagow man dence of man in the Richards. is similar to that of a fragmentary, miner- Pemberton Hill site—The mid-section alized human skull found in 1959 by Tom of a projectile point and a blade or knife Padgitt and Richard Johnson of the Tar- have been collected from the Upper Shu- rant County Archeological Society in a ler member of this exposure, which is but site near Eagle Mountain Lake, Tarrant 4 miles downstream from the Lagow pit. County, Texas. Provenance of the two sites is identical Shuler (1923, p. 334) described his and geology and paleontology of the two fruitless search of the Lagow pit walls for sites are similar. signs of a burial pit. Certainly any burial Human evidence is very scarce, but it or intrusion from an overlying surface, seems indisputably associated with the once the Albritton red sandy clays were other faunal materials. The broken mid- present, should be obvious. Any admixture section of a projectile point could well be of these red sandy clays and the yellow from a typical Clovis-type point, based on sandy clays of the Upper Shuler would be technique, size, and flaking. The flint most striking. Therefore, it seems that the blade, or knife, was found in situ very bones of Lagow man must have reached close to caliche-encrusted fossil remains of their discovery position after the T-2 fill Elephas columbi and Equus caballus. 54 Report of Investigations-—No. 48

Boatwright pit—Sellards (1952, pp. dropped by man on his way to the T-4 ter- 99-105) described the remarkable occur- race to obtain material for artifacts, have rence of crudely carved stone heads in been found relatively recently. human image from near Trinidad, Texas, about 80 miles downstreamfrom the Dallas T-5 DEPOSITS sites. These stones apparently were col- lected from the "Trinity terrace" of Stovall T-5 deposits have a capping of cobbles of petrified ferruginous and McAnulty (1941, pp. 216-217) which quartzite, wood, man made is correlated in this paper with the T-2 de- sandstone, and some flint. Early trips to these cobble fields to obtain ma- posits at Dallas. The stones were collected terial for tools. Because all local cultures at the contact of the basal gravels and the clean cross-bedded sands. Three carved used, these materials, one may find in the T-5 deposits small of most of the stone heads were found; these are at the sites ages Texas Memorial Museum, Austin. A fos- represented on lower terraces. silized human femur, of disputable origin, is also known from this same area. CRETACEOUS UPLANDS Althugh few artifacts have been found Carrollton focus artifacts are found in in the T-2 deposits, the artifacts that have situ in shallow alluvial deposits accom- been recovered from the several sites in- panying small tributary creeks of the vestigated indicate that the T-2 deposits Trinity River, which have their head- offer possibilities where they are exposed waters on the Cretaceous uplands. Through to a considerable depth. these artifacts, these deposits are corre- lated with T-l terrace of the main river T-3 AND DEPOSITS T-4 where the Carrollton focus is also found. T-3 and T-4 deposits have produced no Bison bison predominates in this essen- artifacts; a few flint flakes, no doubt tially modern fauna. COMPARISON WITH OTHER LOCALITIES AND LOCAL FAUNAS

TRINIDAD, TEXAS, LOCALITY est terrace deposits that contain extinct fauna. Because the contained faunas are Comparison between the Dallas terrace similar and the lithologic composition of deposits and those at Trinidad, Texas, de- these two terraces is nearly identical, di- scribed by Stovall and McAnulty (1941) rect correlation is probably between the deserves serious because the ter- attention, paleontologic materials studied for this races in the two areas were deposited by paper and those studied by Stovall and the same river and are but 80 miles apart. McAnulty (1941). The authors correlate the Trinidad fos- It is probable that the variation be- sil material, and the Trinity terrace de- very tween the two areas in numbers of terraces posits of Stovall and McAnulty (1941) in is a result of geography. Dallas lies in the which it occurred, with the Dallas dis- Cretaceous upland zone where the river coveries and the T-2 deposits. This corre- gradient today is still 12 inches per mile. lation is based on independent paleonto- Trinidad lies below the contact line with logic study of the Trinidad region. the Gulf Coastal Plain Tertiary surface; Although Stovall and McAnulty (1941, the gradient there is 8 inches per mile. In p. 216) designated the Tiinity terrace de- addition, Dallas lies in the constricted zone posits as T-3, this merely reflects the indi- where the Trinity River has cut a notch vidual expression at Trinidad of the dual through the resistant Austin chalk; Trini- nature of the T-l deposits at Dallas. At dad lies below the point of debouchment Dallas, the T-l deposits are composed of into the wide, low-lying Tertiary plain. the basal red sandy Albritton clay and Relatively minor alluvial events of late overlain by the gray Pattillo sand. The date might, therefore, have expressed Trinidad T-2, or Trinidad terrace, deposits themselves topographically in a slightly are almost lithologically identical to the Al- different manner in the two areas, whereas britton red sandy clay which is capped by topographic expressions of earlier, major a thin layer of what appears to be gray events would be essentially the same. Pattillo sand; the Trinidad T-l, or Power In 1941, before the advent of radio- Plant terrace, deposits are composed essen- carbon datings, Stovall and McAnulty tially of gray Pattillo sand. Trinidad is tentatively assigned their T-3 deposits to only 80 miles downstream from Dallas, and an interstade of the last glaciation. Thus, such lithologic continuity is very possible. they recognized the salient point: the T-3 At Dallas, Trinity aspect Archaic arche- deposits are non-glacial, i.e., interglacial ological materials occur in place in the or interstadial. This agrees with the au- upper zone of the Albritton, along the Al- thors' views on the age of the Dallas ter- britton-Pattillo contact line, and in the races. In light of recent radiocarbon dates lower half of the Pattillo; Neo-American for the interglacial Beaumont formation archeological remains commonly occur as offshore (C. I. Alexander, personal com- surface deposits on top of the Pattillo. munication, 1959) and the T-2 deposits at This sequence is repeated perfectly in the Stovall and McAnulty's possible corresponding members at Trinidad. Dallas, correlation of their T-3 deposits at Trini- Thus, the Trinidad T-l and T-2 fill was dad with the Lissie is subject to revision. deposited within the last 10,000 years, or essentially within the Recent epoch, based The T-2 deposits at Dallas and the T-3 de- at Trinidad to be correlative on correlation with Dallas deposits, on the posits appear presence of modern fauna, and on radio- with the Beaumont. carbon dates. The Trinidad T-3 deposits, Although the Dallas area fossil material like the Dallas T-2 deposits, are the young- is far more extensive than that described 56 Report of Investigations-—No. 48 from Trinidad, exposures at the two locali- A probable and paleontologic correla- ties are greatly disproportionate. The Dal- tion exists between this late Pleistocene ter- las pits are a result of much greater eco- race fill and the T-2 deposits on the Trinity nomic exploitation of sand and gravel. River at Dallas. Allowing for the differ- Those fossils present at Trinidad are, in ence in upstream drainage, lithologic simi- the main, almost directly comparable to larities are notable. An excess of calcium the Dallas specimens. carbonate in the terrace materials is re- It is perhaps noteworthy that the new sponsible for the good preservation of the species Megalonyx brachycephalus (Mc- Iron Bridge fossils, which occur in the Anulty) from Trinidad may be present in most extensive vertebrate fossil bed in the Dallas collection. This cannot be dem- Texas that is near the moist, destructive onstrated positively because of the frag- humid climate of the wide Mississippi mentary nature of the Dallas specimens, basin. but the size range is similar. Likewise, ex- Miss Hazel Peterson, East Texas State amination of the deer jaw in the Trinidad College; Dr. E. L. Lundelius, The Univer- collection proved that the jaw belongs to sity of Texas; and the writers collected the the new subspecies from Dallas, Dama vir- following Pleistocene fossils: Mammut ginianus aplodon (Slaughter). americanus, Elephas columbi, Camelops The only problem that the arose in com- sp. (although not directly comparable parison of contained materials was the with the Dallas specimens, the size is identification in the Trinidad collection of similar), Bison alleni or B. latifrons, Tetra- a nearly complete skull of Bison occi- meryx shuleri, Breameryx sp., Equus com- dentalis. plicatus or E. fraternus, Holmesina sep- According to measurements published tentrionalis, and Alligator mississippiensis. the by Stovall and McAnulty (1941, p. 241), All of these types are represented in this specimen's horn-core length is well be- Dallas collections; there seems to be noth- yond the for this as with maximum species ing to negate the proposed correlation given in Skinner and Kaisen's (1947, p, the Hill-Shuler local faunas. The presence 170) definitive work on the Bison. Thus, of T. shuleri at Iron Bridge may be most despite the configuration, the B. occiden- significant, because the holotype of this Dallas oc- talis assignment is somewhat in question. genus and species is from the Added to this is the lack of positive geo- currence. logic provenance for this particular speci- The predominance of mastodon at Iron men at Trinidad. Unlike the other de- Bridge is in strong contrast to the approxi- scribed fossils, this specimen is attributed mately 50 to 1 ratio of mammoth-mastodon the . by . from near Cayuga, Anderson County at Dallas. This is easily explained ." struc- . . It may or may not be an example from grazing habits inferred from tooth Ele- the T-3 deposits at Trinidad. The measure- ture and geographic distribution of ments of this specimen seem closer to those phas and by the browsing-tooth structure is of B. alleni, the species to which the Dallas of Mammut. Even today the delineation bison material is assigned. fairly sharp between west Texas grass- of lands and east Texas timber. The town IRON BRIDGE DAM LOCALITY Edgewood, 90 miles east of Dallas, is well boundary At Iron Bridge Dam, approximately 50 named. This grassland-forest in the past and miles east of Dallas on the Sabine River most certainly existed climatic drainage in Van Zandt and Hunt counties, fluctuated east and west, with Dallas earth-fill excavations exposed another in- changes. In upper Pleistocene time, teresting bone bed that bears comparison was in grassland country and mammoth with the bones from the T-2 deposits at predominated; Iron Bridge was in a for- Dallas. ested area and mastodon was the more Hill—Shuler Local Faunas 57 prominent. Other forms of animal life T-2 deposits have a greater total of forms intermingled between the grassland and over Ingleside-Berclair than Ingleside-Ber- forest habitats. clair in turn have over Trinidad. Stovall Even today, relict pines can be found on and McAnulty (1941, p. 247) also con- the upper Sabine River drainage in south- sidered the age of the Ingleside-Berclair west Hunt County, and relict palmetto and faunas lo be similar to that of Rancho La swamp birch can be found on the middle Brea; they considered Trinidad and Mc- Sabine near Mineola. These are indicative Kittrick local faunas to be slightly older. of a past climate similar to the more hu- No further faunal comparison can be made mid, sub-tropical climate now existing east with the Ingleside-Berclair deposits until and southeastward toward Louisiana. The material from them has been identified to Dallas area of the upper Trinity River sup- the specific level. ports a sub-arid flora of pecan and hack- Sellards (1940, p. 1646) related the berry and other hardwoods in the bottoms Berclair terrace formation to the period of and draws, whereas the uplands basically high sea stand during which the marine are grasslands. Beaumont formation was deposited. In the Because the upper Sabine River and the Berclair area, theBeaumont is recognizable Elm Fork of Trinity River are so close, near Refugio, some 80 miles inland and at both must have reacted to the same cli- an elevation at least 80 feet above sea level. matic stimuli even in ancient times. The flood-plain alluvium that later became the Berclair terrace merges with delta and INGLESIDE AND BERCLAIR LOCALITIES beach deposits of this high sea stand. It is believed that the Beaumont was deposit- The famous Ingleside fauna from the ed during a late interglaciation and prob- Texas Coastal Plain has not been thor- ably the last interglaciation or Sangamon. oughly described in the literature; results Radiocarbon dates indicate that the now- of future studies will be important for submerged offshore Beaumont formation comparison with the results of study of the was subjected to subaerial weathering and Dallas material. erosion between about 28,000 and 10,000 The Ingleside deposits were considered BP. The latter date was the beginning of by Sellards (1940, p. 1651) to be con- the present transgression and the deposi- temporary with the Berclair terrace de- tion of Recent marine sediments. This sug- posits. Descriptions by Sellards and subse- gests that the Berclair terrace deposits are quent references by other writers to un- roughly contemporary with the T-3 depos- published knowledge of the collections its at Trinidad. These deposits all seem lo have been largely in terms of genera be last interglacial alluvial equivalents of rather than species. Despite this handicap, the marine Beaumont formation. it is obvious that the fauna is late Pleisto- On the other hand, as described by Sel- cene. Stovall and McAnulty (1941, p. lards (1940, p. 1651) and confirmed by the 247) after comparing the Ingleside local writers' studies of the site, the Ingeside pit fauna to the Trinidad fauna, concluded was a lake deposit, situated on the exposed that the Trinidad material was slightly surface of the Beaumont. Thus, at the time older than that of Ingleside-Berclair; cer- of the Ingleside lake fossil accumulation, tainly no younger. The absence of a num- the sea had withdrawn at least as far as it ber of genera at Trinidad may not be sig- is today and perhaps farther. There are nificant. because of the wider variety of shallow somewhat brackish lakes immedi- faunal material found in Ingleside and ately adjacent to the Ingleside site, but the Berclair terrace deposits. This is especially water in these lakes is not too brackish for true because the Dallas and Trinidad frogs to live in and cattle to drink. The faunas are equated herein, and the Dallas present vegetation is being overrun by 58 Report of Investigations—No. 48 scrub live-oak, but some sixty years ago, scraper, indicative of early man, was found according to the memory of local residents, beneath the Dinobastis skeleton. the area was coastal prairie, perhaps as it The fauna is not nearly as similar to the was in the day of Ingleside's fossil lake. Dallas material as are the Trinidad, Iron It is suggested that the Ingleside deposits Bridge, and Berclair faunas. Age and prob- must be somewhat later than those of Ber- ably climatic differences are suggested. clair, perhaps laid down during the early Even today the areas are different physio- stages of the last glaciation. The Ingleside graphically. fauna, however, cannot have been greatly Pending possible radiocarbon dating of different from that of the immediately the Friesenhahn charcoal, the writers sug- preceding interglaciation, because intergla- gest that the cavern deposits may represent cial forms may have lingered on even after early last glaciation. Perhaps the turtles a new glacial advance had commenced in were hibernating regularly in this favor- the north and had lowered sea level 80 feet able underground retreat as the weather or more. in the area cooled. As winter cold periods The age of the Ingleside-Berclair faunas, lengthened, fewer and fewer turtles would therefore, should be between the last inter- survive, which would account for the large glaciation and early last glaciation. Be- collection of fossil turtles. This could also cause most of this material occurs in the explain the presence of Geochelone (re- Ingleside lake bed, much of the material stricted in its final appearence to the San- may be slightly later than the Trinidad and gamon interglaciation) and the large num- Dallas faunas. bers of Terrapene canaliculata which, to date, have not been reported north of the FRIESENHAHN CAVERN LOCALITY Lewisville site (T-2, Upper Shuler) in Denton County. Friesenhahn Cavern, Bexar County, is It has also been suggested that the ex- another famous and important but not of these turtles was caused by in- fully described late Pleistocene locality. tinction creasing aridity. If so, the cave deposits This site is in the Edwards Plateau region necessarily be allocated to an inter- of Texas, which is higher today than either must stadial within the last glaciation or the gla- the Coastal Plain or the Dallas-Trinidad itself must be considered arid. Be- region and characterized by dissected ciation are generally believed to limestone hills covered with brush and cause glaciations be accompanied by pluvials, only the first cedar. Publication of more complete stud- will be considered. Arid condi- ies of this material will be most welcome. possibility seem unlikely, however, because of The only detailed publication on this tions the number of mastodon remains site is by Milstead (1956), who described large Friesenhahn Cavern. The brows- numerous turtle specimens. Prior to the found at of the mastodon is in- discovery of fauna from the Upper Shuler ing tooth structure of forested conditions which would member of the Lewisville T-2 deposits, the dicative be available during times arid most prolific and northernmost occurrence hardly to annihilate the vast numbers of of Terrapene canaliculata was probably in enough Friesenhahn Cavern. In addition, Friesen- turtles. The presence of Sigmodon hispidus at hahn contained a good hit of Geochelone seemingly would negate the material, including a new species not yet Friesenhahn of this local fauna representing recognized elsewhere in Texas. These ex- possibility (Illinoian) cavations also produced a complete skele- the next earlier glacial stage of ton on the great cat Dinobastis serus; a because the earliest reported occurrence complete skeleton of the peccary Mylohyus; this species in the United States is from and remains of Mammut, Elephas, Pan- the Lower Shuler member of the T-2 thera, and many other genera. A flint deposits. Hill—Shuler Local Faunas 59

If, then, the Friesenhahn assemblage along that Arkansas drainage from eastern generally suggests a glaciation, the scanty Oklahoma, especially when interglacials but definite evidence of human presence are normally considered as having been almost certainly places the period in the drier. Now, however, a somewhat compara- last glaciation of circa 25,000 to 10,000 ble situation is known on the upper Trinity BP. Turtle remains suggest that the Frie- River near Dallas. Here the upstream ad- senhahn fauna should be assigned to the vance of native pine from the vicinity of earlier half of this period. Anderson County is a minimum of 100 miles. Further, because the Dallas region JINGLEBOB LOCALITY has blackland on the divides and uplands, in which pines cannot grow and propagate, Hibbard's (1955b) Jinglebob local the fossil pine flora could only have existed fauna of interglacial (Sangamon?) in age in the sand and sandy clay deposits along south-central Kansas, north of the just the stream valley. Oklahoma border, resembles a slightly This may suggest the cause of some of more northern version of the Dallas collec- the heavy iron cementation at the top of the tion. The question mark following Sanga- Hill gravels and of the yellow limonite in mon indicates the problem in referring to the Lower Shuler sand at Dallas. Conifers the last interglaciation, whatever it may concentrate iron in the soil, whereas oaks eventually be named or numbered. concentrate calcium carbonate. Hibbard (1955b) interpreted the Jingle- During the earlier phases of the Dallas bob local fauna as non-glacial. Combining T-2 deposition, the climate appears to have material from several localities, he was been more moist than that characteristic of handicapped by the preponderance of small the area today; during later phases of T-2 vertebrate forms over the more definitive deposition progressive drying-up occurred, larger species; the Dallas collections offer culminating in greater aridity than today. no such handicap. Hibbard is the outstand- The evidence from Jinglebob agrees with ing advocate of the idea thatBison arrived this sequence of events. in America during the Illinoian glacial and the presence of B. latifrons in an inter- RANCHO LA BREA LOCALITY glacial fauna suggests Sangamon age. Also indicative of an interglacial age is the The term "Rancholabrean" is the de- presence of Terrapene llanensis, related to scriptive name of the general late Pleisto- T. canaliculata, which is the turtle present cene fauna of North America, based upon in the Dallas T-2 deposits. Obviously, giant the magnificent collections from Rancho bison and T. canaliculata in the Hill-Shuler La Brea tar pits in California. The faunas local faunas, although slightly different of the T-2 deposits at Dallas and all other from those in the Jinglebob, suggest a faunas which are discussed herein are quite general correlation with it. Additional cor- similar in a rather general way with the relation is indicated by the similarities of Rancho La Brea local fauna; therefore, Arctodus collected from the different all of the above mentioned fossil faunas faunas. should be correlated with the ""Ranchola- The land snails of Hill-Shuler and Jin- brean." glebob faunas offer suggestive compari- Superb as the California material is, it sons, and the identification of pine pollen does occur at the far western edge of a vast in the deposits at both localities is signifi- may be incomplete cant. continent and possibly Considering the semi-arid climate cur- or not entirely representative of forms pres- at rent in the Jinglebob area (Meade County, ent elsewhere, especially those consider- Kansas), it is surprising to think of an able distance. Such discrepancies, how- upstream advance of a flora containing pine ever, should not distort the basic premise 60 Report of Investigations—No. 48 of a generally similar, widely distributed tainties concerning contamination by the late Pleistocene fauna. tar, these tests are certainly worth consid- Again, although upper Pleistocene ering. The radiocarbon tests indicate ages roughly includes the Illinoian-Sangamon- of about 14,000 to 17,000 BP, which fall Wisconsin (certainly at least the last two near the middle of the last glaciation. glaciations and the intervening intergla- The Rancho La Brea fauna is somewhat ciation), it is not certain that all of the late, e.g., the Bison present is antiquus Rancho La Brea materials represent this rather than the giant species ( B. latifrons, span. The period of accumulation involved B. alleni) currently thought to be represen- is not positively known, nor, for that mat- tative of the last interglaciation. The horse ter, is the precise provenance of any given fauna is rather small, with but one species, specimen from the various tar deposits. Equus occidentalism in contrast to the sev- Several clues suggest, however, that per- eral equine groups in the terrace deposits at haps the bulk of the fossil material from Dallas. Breameryx is the extinct antilo- the Californiatar pits is last glacial-pluvial caprid present, rather than Tetrameryx, in age. The Rancho La Brea local fauna Stockoceras, etc.; from the Scharbauer ar- would, therefore, belong to late upper cheological site near Midland, Texas, it is Pleistocene. known from radiocarbon dating that this The fantastic but easily explainable con- tiny antelope lasted until 12,000 to 20,000 centration of carnivores in the tar pits im- BP. The absence of large land tortoises plies an even more vast population of may not be significant because of geogra- herbivores in the immediate area to support phy, but southern California enjoys a very these predators. Observing theLos Angeles mild climate today and has ecological con- basin today, with its aridity and brush tact with Mexico. Because Tapirus is pres- fires, it is difficult to imagine the presence ent at Rancho La Brea, the absence of turtle there of the forage necessary to support may indicate a time slightly later than the herds of such large animals, unless it were last interglaciation, and from the presence connected with during a pluvial period of sub-Recent and Recent fauna in the tar During such glacio-pluvial glaciation. pits, it is apparent that accumulation con- times, even the desert country of the inte- tinued into post-glacial time. rior portions of southern California, Ari- While definitely correlative with the zona, Nevada, and Utah contained large fauna of the upper Pleis- lakes and enough vegetation to support a "Rancholabrean" Dallas T-2 assemblage is consid- great Pleistocene fauna. tocene, the and re- A few radiocarbon tests have been run ered to be somewhat earlier in age on wood and plant material from the tar gionally different in a number of species pits, and although there are many uncer- and even genera. CONCLUSIONS

In summing up the Pemberton Hill— A number of climatic implications re- Lewisville (or T-2) terrace deposits and garding this non-glacial period of the upper the faunal, floral, and archeological ma- Pleistocene are of interest. Two basic cir- terial contained therein, it is concluded that cumstances must be understood. First, if they belong to a non-glacial period of the valley cutting is correlated with glacial Upper Pleistocene. Whether this is ulti- periods and valley filling with interglacial mately correlated with an interglaciation periods for Gulf Coastal Plain streams today called Sangamon, or a new inter- (Fisk, 1938, 1940) the following may be glaciation which may become the last inter- visualized. The maximum glacial advance glaciation, or be assigned to a pronounced was attained during lowest sea level. As and major interstade yet to be established the glaciers retreated and sea level rose, within the last or Wisconsin glaciation, river flow and cutting action slowed. Fil- remains to be seen. ling began, at first with coarser material, Radiocarbon dates now indicate that the then as stream flow was further reduced classic concept of the Wisconsin glaciation (and sea level rose higher and higher), encompasses approximately the period progressively finer sands and silts were de- 25,000 to 10,000 BP, a period during posited until the flood plain reached its which sea level was significantly lowered. highest level. Equilibrium was achieved Along the Texas coast, the Beaumont for- sometime during the ensuing interglacia- mation is the product of late interglacial tion. As a new glaciation began, sea level stream increased, valley cut- high sea stand, during which the rise in sea fell, flow and began again, eventually leaving the level was as much as 80 feet above present ting former flood plain as remnants and ter- sea level and the coastline advanced about races. a terrace fill should represent 30 miles inland from today's coastline. Thus, a period extending from a glacial retreat Radiocarbon dates of offshore cores taken into the following interglaciation, until the Beaumont from formation have shown either equilibrium is reached or cutting is that the sea had withdrawn and subjected renewed. portions of today's Continental Shelf to Second, the Dallas region is an area of subaerial weathering by about 28,000 BP overlap between the east Texas-Missis- when deposition of Recent marine deposits sippi Valley timbered country and the commenced. These events agree rather grassland prairies to the west. Timber closely with the radiocarbon brackets occurs in the draws and stream valleys, (25,000 to 10,000 BP), mentioned at the whereas adjacent divides and uplands are beginning of this paragraph, placed upon prairies. As climate varies between moist the classic Wisconsin. and dry, the timbered bottoms and grassy uplands and relative to each The interglacial and related high sea wax wane, This caused stand during which the Beaumont forma- other, up and downstream. is by the drainage heading in more arid tion was deposited should, therefore, be areas to the northwest but flowing south- earlier than about 28,000 BP. The two eastward into more moist zones. Presum- radiocarbon dates from different samples ably, flora and fauna of these two different at the Lewisville site in the Dallas T-2 de- environments would fluctuate in like indicate an posits age greater than 37,000 manner. BP for these deposits. These deposits ap- The beginning of deposition of the Dal- parently pre-date the classic Wisconsin and las T-2 fill presumably took place in a may some day be shown to correspond period of greater annual rainfall, and cer- with the Beaumont marine deposits. tainly stream flow, than today. The basal 62 Report of Investigations—No. 48

Hill gravels represent the strongest flow; climate types. A few—perhaps the large these gravels are followed by the overlying alligator, the giant Geochelone, the tapir, Lower Shuler sand, the Upper Shuler the peccary, and the armadillos—suggest sandy clay, and the locally derived black- a slightly warmer climate than today. Cer- land alluvium of the Richards, all of which tainly winters could have been no colder appear to represent progressively reduced than those occurring in the area today and carrying power. Iron cement and deposits possibly were milder, with resultant cooler of limonite, present respectively in the Hill summers; in other words, a more equable gravel near the top and in the immediately year-round climate must have existed. If overlying Lower Shuler sand, suggest a the interglacial high sea stand extended more humid climate than exists today in the shoreline 50 miles inland from the the region. The pine pollen from the Lower present coastline, that may have contrib- Shuler implies upstream migration of uted both to a milder climate and an in- flora; it indicates an annual rainfall, or a creased rainfall in the earlier stages, hav- continuous water equivalent in the flood ing the comparable climatic effect of mov- plain, of 40 or .more inches as compared ing the region that distance southeast with today's 36-inch average at Dallas. The today. snail fauna in particular, both from the Most of the fossils are directly compara- Lower Shuler and even from the lower part ble with material described from other of the Upper Shuler, strongly indicates a localities. Differences, such as noted here- heavily timbered environment, at least in in for new forms, are no more than region- the bottoms of the flood plain at the time. al variations to be expected in thorough The presence of alligator, beaver, musk- investigations of an area not previously rat, vole, and a number of other forms studied in detail. which live in timber and in moist climate Certain evolutionary trends exist which reinforce this assumption. may have some bearing on extinctions or On the other hand, the presence of cal- faunal changes, at least in the Dallas re- careous nodules, such as those found in gion. At Dallas, mastodon fossils are vir- the Upper Shuler, is often considered in- tually unknown above the Hill gravels, dicative of aridity. These concretions in- whereas mammoth bones are abundant. crease in size from bottom to top of this Giant land tortoises are common in the member. The sharp reduction in number Hill and Lower Shuler members but are of species of large animals and the pres- extremely rare above these members. Ter- ence of prairie animals, such as prairie rapene canaliculata seems to replace a near dogs and the turtle Gopherus, lend sup- smaller carolina-Wke Terrapene the port to the hypothesis of an increasingly Lower Shuler-Upper Shuler contact. In arid climate. the lower members, fossils of the equine The existence of timbered valleys sur- group are abundant and varied, with bones rounded by grassy uplands, however, made of horses, , and asses present; in it possible for both woodland and prairie the Upper Shuler bones of caballine horses animals to exist contemporaneously within appear and are, to date, the sole repre- a short distance of each other. It appears sentatives. There are at least two caballine that conditions were more favorable to the types, one draft-horse size and one pony woodland, moist climate biota in the ear- size. Remains of the large carnivores are, lier phases of terrace deposition, and less so far, restricted mostly to the Hill and so in the latter phases. Lower Shuler; only the bones of coyote None of the fossils, with the possible ex- and a small black bear have been found in ception of Synatomys from the Upper Shu- the Upper Shuler. ler, indicates a cold climate by today's Some of these apparent faunal changes, standards. The vast majority are temperate especially the absences, could be the result Hill-ShulerLocal Faunas 63 of chance collecting. Many or all could be The period of the classic Wisconsin, connected with the change to a drier cli- of circa 25,000 to 10,000 BP, is not well mate. Allowing for all of these, there still represented in the terrace deposits at Dal- remains a subtle suggestion of faunal las because the river was cutting rather change, perhaps not of types but than depositing. After this period of valley reduction in numbers so that the fauna cutting, the T-l fill of Recent age was de- may have become highly vulnerable to posited, and it contains a modern mam- climatic change during the last glacial malian fauna, including Bison bison, and advance. a modern snail fauna. Radiocarbon dates of Many these animals lived on into the suggest that deposition of these deposits last and early glaciation some even into began at least 6,000 BP and possibly one or post-glacial time. Based upon the southern two thousand years earlier. While the High Plains archeological sites, however, smaller forms continued into the upper it is known that by essentially the close of the larger such as the the last glaciation, in Folsom-Yuma time of Pleistocene, ones, deer and the black about 8,000 to 11,000 BP, the bison re- excepting possibly bear, mained in extinct form but in smaller had almost completely changed. Man was species. A few remnant camels, sloth, and common in the Archaic Stage, which be- horses lingered in the west and the last of gan as early as 10,000 BP and lasted until the mastodons in the east. about 2,000 to 4,000 BP. BIBLIOGRAPHY

BAKER, F. C. (1902) The Mollusca of the Chi- Gegenwart, vol. 9, pp. 10-17, Oehringen, Ger- cago area: Chicago Acad. Sci. Bull. no. 3, pt. 2, many. Gastropoda, pp. 137-410, 9 pis., 124 figs. DOERING, J. A. (1956) Review of Quaternary * (1928) The fresh-water Mollusca of surface formations of Gulf Coast region: Bull. Wisconsin, pt. 1, Gastropoda: Wisconsin Geol. Amer. Assoc. Petrol. Geol., vol. 40, pp. 1816- Nat. 'Hist. Survey Bull. 70, 507 pp., 20'2 figs., 1862, 11 figs., 7 tbls. 28 pis. FISIC, H. N. (1938) Geology of Grant and LaSalle BARBOUR, E. H., and SCHULTZ, C. B. (1937) An Parishes: Louisiana Geol. Survey Bull. 10, pp. early Pleistocene fauna from Nebraska: Amer. 149-181, 5 figs. Mus. Nat. Hist. Novitates no. 942, pp. 1-10, (1940) Geology of Avoyelles and 4 figs. Rapides Parishes: Louisiana Geol. Survey Bull. BRANNON, H. R., JR., DAUGHTRY, A. C., PERRY, 18, pp. 53-113. D., SIMONS, L. H., WHITAKER, W. W., and FRYE, J. C., and LEONARD, A. B. (1957) Eco- WILLIAMS, MILTON (1957) Humble radiocar- logical interpretations of Pliocene and Pleisto- bon dates I: Science, new ser., vol. 125, no. cene stratigraphy in the Great Plains region: 3239, pp. 147-150. Univ. Texas, Bur. Econ. Geol. Rept. Inv. 29, BRAY, E. E., and NELSON, H. F. (1956) Radio- 11 pp., 3 figs. carbon age determination of Recent-Pleisto- GIDLEY, J. W. (1901) Tooth characters and cene contact in Block 126 Field, Eugene Is- revision of the North American species of the land, Gulf of Mexico: Bull. Amer. Assoc. genus Equus: Bull. Amer. Mus. Nat. Hist., Petrol. Geol., vol. 40, no. 1, pp. 173-177, 2 figs. vol. 14, art. 9, pp. 91-142, 27 figs., 4 pis., 2 this. BROWN, BARNUM (1907) The Conard fissure, a (1921) Pleistocene peccaries from Pleistocene bone deposit in northern Arkansas: the Cumberland Cave deposit: Proc. U.S. Nat. with descriptions of two new genera and Mus., vol. 57, no. 23'24', pp. 651-678, 13 figs., twenty new species of : Amer. Mus. 2 pis. Nat. Hist. Mem., vol. 9, pt. 4, pp. 155-208, GOODWIN, G. G. (1954) The animal kingdom, 3 figs., 12 pis. vol. 1, Mammals: Doubleday and Co., Inc., CHANDLER, A. C. (1916) A study of the skull and Garden City, N.Y., 680 pp. dentition of Bison antiquus Leidy, with special HARRIS, R. K. (1942) The Gilkey site: The reference to material from the Pacific Coast: Record, vol. 6, no. 4, pp. 18—19. Univ. California Geol. Pub., vol. 9, no. 11, pp. HAY, O. P. (1913) Notes on some fossil horses, 121^135,12 figs., 3 tbls. with descriptions of four new species: Proc. CHEATUM, E. P. (MS) Monograph of the fresh- U.S. Nat. Mus., vol. 44, no. '1969, pp. 569-594, water and land snails of Texas. (Unpublished.) 28 figs., 5 pis. COPE, E. D. (1884) The extinct Mammalia of the (1914a) The extinct bisons of North Valley of Mexico: Proc. Amer. Philos. Soc. America; with description of one new species, Philadelphia, vol. 22, pp. 'l-21. •Bison regius : Proc. U.S. Nat. Mus., vol. 46, (1893) A preliminary report on the no. 2021, pp. 161-200, 10 figs., 12 pis. vertebrate paleontology of the Llano Estacado: (1914b) Camels of the fossil genus Texas Geol. Survey, 4th Ann. Rept. (1892), Camelops: Proc. U.S. Nat. Mus., vol. 46, no. pt. 2, pp. 1-137, 23 pis. 202'5, pp. 267-277,1 fig., '2 pis. CROOK, W. W., JR. (1932) The Wheeler site: A (1915) Contributions to the knowl- 3500-year old culture in Dallas County, Texas: edge of the mammals of the Pleistocene of Field and Laboratory, Vol. XX, no. 2, pp. 43- North America: Proc. U.S. Nat. Mus., vol. 48, 65, 3 figs., 6 pis., 1 chart. no. '2086, pp. 515-575, 5 figs., 8 pis.

(1961) A revised interpretation of — (1917) Description of a new species the Lagow discovery, Texas: Amer. Antiquity, of extinct horse, , from the vol. 26, no. 4 : , pp. 545-548, 1 fig. Pleistocene of Yukon Territory: Proc. U.S. - and HARRIS, R. K. (1952) Trinity Nat. Mus., vol. '53, no. '2212, pp. 435-443, 3 pis. aspect of the Archaic horizon: The Carrollton (1921) Descriptions of some Pleisto- and Elam foci: Bull. Texas Arch. Paleont. Soc., cene vertebrates found in the United States: vol. '23, pp. 7-3 ! 8, 2 figs., 5 pis. Proc. U.S. Nat. Mus., vol. 58, no. 2328, pp. —, (19570 Hearths and arti- 83-146, 4 figs., 9 pis. facts of early man near Lewisville, Texas, and HENDERSON, JUNIUS (1935) Fossil non-marine associated faunal material: Bull. Texas Arch. Mollusca of North America: Geol. Soc. Amer. Paleont. Soc., vol. 28, pp. 7-97, 11 figs. Spec. Paper 3, 313 pp. ,(1958) A Pleistocene HIBBARD, C. W. (1955a) Pleistocene vertebrates campsite near Lewisville, Texas: Amer. An- from the upper Becerra (Becerra Superior) tiquity, vol. 23, no. 3, pp. 233-246, 6 figs. formation, Valley of Tequixquiac, Mexico, with DALQUEST, W. W. (1957) First record of Bison notes on other Pleistocene forms: Univ. alleni from a late Pleistocene deposit in Texas: Michigan Contr. Mus. Paleont., vol. 12, no. 5, Texas Jour. Sci., vol. 9, no. 3, pp. 346-354, pp. 47-96, 5 figs., 9 pis., 12 tbls. 1 % (1955b) The Jinglebob interglacial DEVRIES, H. (1958) Radiocarbon dates for Eem (Sangamon?) fauna from Kansas and its cli- and Wurm-interstadial samples: Eiszeitalter u. matic significance: Univ. Michigan Contr. Hill-Shuler Local Faunas 65

Mus. Paleont., vol. 12, no. 10, pp. 179-228, PATTILLO, L. F., JR. (1940) River terraces in the 8 figs., 2 pis., 1 chart. Carrollton area, Dallas County, Texas: Field no. 27-32, 1 pi. -— (1958) Summary of North American and Laboratory, vol. 8, 1, pp. Pleistocene local faunas: Papers Michigan 2 figs. _ __ .. „„. (MS) Acad. Sci., vol. 43, pp. 3-32. [1957 meeting.] PETERSON, HAZEL, and SLAUGHTER, B. H. Bridge of — and TAYLOR, D. W. (I960) Two late The Iron local fauna ('Sangamon?), Pleistocene faunas from southwestern Kansas: Texas. Mollusca of North Univ. Michigan Contr. Mus. Paleont., vol. 16, PILSRRY, H. A. (1940) Land : Acad. Nat. no. 1, pp. '1-223, 18' figs., 16 pis., 9' this. America (north of Mexico) Sci. Philadelphia Mon. 3, vol. 1, pt. 2, pp. 575-994, HOLMAN, J. ALAN (1958) The Pleistocene Herpetofauna of Sabertooth Cave, Citrus '203 figs. — (1946) Land Mollusca of North County, Florida: Copeia, Dec. no. 4, pp. '22, (north Mexico) : 276-280, 3 America of Acad. Nat. Sci. tbls. Philadelphia 1—520, (1931) Mon. 3, vol. 2, pt. 1, pp. HOLMES, W. W., and SIMPSON, G. G. figs. Pleistocene exploration and the edentates of 281 Florida: Bull. Amer. Mus. Nat. Hist., vol. 59, (1948) Land Mollusca of North America (north of Mexico) : Acad. Nat. Sci. art. 7, pp. 383-418, '2l figs. Philadelphia Mon. 3, vol. 2, pt. 2, pp. 521- KRIEGER, A. D. (1946) Culture complexes and 1113, 304 figs. northern Texas: Univ. Texas chronology in POTZGER, ,T. E., and TIIARP, B. C. (1947) Pollen Pub. 4640, 366 pp., 26 figs., 35 pis., 6 tbls. profile from a Texas bog: Ecology, vol. 28, (1950) LEONARD, A. B. A Yarmouthian mollus- no. 3, pp. 274-280, 2 figs., 1 tbl. can fauna in the mid-continent region of the PYBURN, WILLIAM (1956) A study of fossil United States: Univ. Kansas Paleont. Contr., Testudo material from Ingleside, Texas: report Mollusca, art. 3, pp. 1-48, 4 figs., 6 pis. on research problem made for course work in — (1952) Illinoian and Wisconsin Department of Geology, Univ. Texas (unpub- molluscan faunas of Kansas: Univ. Kansas lished) . Paleont. Contr., Mollusca, art. 4, pp. 1-38, 15 QUINN, J. H. (1957) Pleistocene Equidae of figs., 5 pis. Texas: Univ. Texas, Bur. Econ. Geol. Rept. ■ (1959) Handbook of gastropods in Inv. 33, 51 pp., 1 fig., 7 pis., 10 tbls. Kansas: Univ. Kansas, Mus. Nat. Hist, Misc. RAY, C. N. (1938) The Clear Fork culture com- Pub. 20, 224 pp., 87 figs., 11 pis. plex: Bull. Texas Arch. Paleont. Soc., vol. 10, LULL, R. S. (1921) Fauna of the Dallas sand art. 7, pp. 193—207, 2 pis. pits: Amer. Jour. Sci., 51th ser., vol. 2, pp. 159- RINKER, G. C. (1949) Tremarctotherium from '176, 5 figs. the Pleistocene of Meade County, Kansas: MARTIN, P. S. (1958) Pleistocene ecology and Univ. Michigan Contr. Mus. Paleont., vol. 7, biogeography of North America, in Zoogeog- no. 6, pp. 107-112, 1 pi. raphy (Symposium of Indianapolis meeting:, SELLARDS, E. H. (1940) Pleistocene artifacts and County, Dec. 28, 1957) : Amer. Assoc. Adv. 'Sci. Pub. associated fossils from Bee Texas: 51, pt. '2, art. 15, pp. 375-420, 6 figs., 2 tbls. Bull. Geol. Soc. Amer., vol. 51, pp. 1627-1657, MATTI-lEW, W. D. (1920) A new specimen of the 7 figs., 2 pis., 4 tbls. Pleistocene bear, Arctotherium, from Texas (1952) Early man in America: Uni- versity 47 figs., (abst.) : Bull. Geol. Soc. Amer., vol. 31, pp. of Texas Press, Austin, 211 pp., '224-2'25. 8 1 pis. (1918) The Geology of Dallas MERRIAM, C. (1913) The skull and dentition SHULER, E. W. J. Bull. 1818, 1-54, of a camel from the Pleistocene of Rancho La County: Univ. Texas pp. 2 Brea: Univ. California Geo!. Pub., vol. 7, no. figs., 21 pis. 14, pp. 305-323,11 figs. — (1923) Occurrence of human re- mains with Pleistocene fossils, sand pit, (1925) Relationships and structure Lagow of the short-faced Arctotherium, Dallas, Texas: Science, vol. 57, no. 1472, pp. bear from the 333-334. Pleistocene of California: Contr. to Paleont., Carnegie Inst. Washington (1935) Terraces of the Trinity River, Pub. 347, pp. 1-35, County, Laboratory, figs., pis., Dallas Texas: Field and 5 10 18 this. figs. (1956) vol. 3, no. 2, pp. 44—53, 2 MILSTEAD, W. W. Fossil turtles of Fries- G. G. (1929) Pleistocene mammalian enhahn SIMPSON, Cave, Texas, with the description of a fauna of the Seminole Field, Pinellas County, new species of Testudo: Copeia, Aug. 29, no. 3, Florida: Bull. Amer. Mus. Nat. Hist., vol. 56, pp. 3 tbls. 162—'171, art. 8, pp. 561-599, 22 figs. OAKLEY, K. P., and HOWELLS, W. W. (1961) (1945) Notes on Pleistocene and Age of the skeleton from the Lagow sand pit, Recent tapirs: Bull. Amer. Mus. Nat. Hist., Texas: Amer. Antiquity, vol. 26, no. 4, pp. vol. 86, art. 2, pp. 33-82, II figs., 6 pis. 18 this. L 54'3 -545, 1 table. ■ (1949) A fossil deposit in a cave in OSBURN, H. F. (1942) Proboscidea, Vol. II: St. Louis: Amer. Mus. Nat. Hist. Novitates no. Amer. Mus. Nat. Hist. Press, pp. 805-1673, 1408, pp. 1-46, 6 figs., 9 this. 563 figs., 30 pis. SKINNER, M. F. (1942) The fauna of Papago OWEN, R. (1869) On fossil remains of equines Springs Cave, Arizona, and a study of Stocko- from Central and South America referrable to ceras, with three new antilocaprines from Equus conversidens Ow., Equus tau Ow., and Nebraska and Arizona: Bull. Amer. Mus. Nat. Equus orcidens Ow.: Philos. Trans. Royal Soc. Hist., vol. 80, art. 6, pp. 143-220, 19 figs., 26 London, vol. 159, pp. 559-573, 2 figs., 2 pis. tbls. 66 Report of Investigations—-No. 48

and KAISEN, 0. C. (1947) The fossil STOCK, CHESTER (1928) A peccary from the Bison of Alaska and preliminary revision of McKittrick, Pleistocene, California: Carnegie the genus: Bull. Amer. Mus. Nat. Hist., vol. Inst. Washington Pub. 393, art. 3, pp. 25—27, 89, art. 3, pp. 123-256, 5 figs., 19 pis., 25 this., 7 figs. 3 maps. STOVALL, J. W, and MCANULTY, W. N. (1941) The vertebrate fauna and geologic age of SLAUGHTER, B. H. (1959) The first noted occur- Trinity River terraces in Henderson County, rence Dasypus bellus of in Texas: Field and Texas: Amer. Midland Nat, vol. 44, no. 1, Laboratory, vol. 1 pi. 27, no. 2, pp. 77-80, pp. 21'1—250, 2 figs, 2 pis, 7 tbls.

— ('1960) A new species of Smilodon TAYLOR, D. W. (1954) A new Pleistocene fauna from a late Pleistocene alluvial terrace deposit and new species of fossil snails from the High of the Trinity River: Jour. Paleont., vol. 34, Plains: Univ. Michigan Mus. Zool. Occas. no. 3, pp. 486-492, 1 fig., 1 pi, 1 tbl. Papers no. 557, pp. 1-16. UYENO, TERUYA, and MILLER, R. R. (1962) Late STEPHENSON, R. L. (1949) Archeological survey Pleistocene fishes from a Trinity River terrace, of Lavon and Garza-Little Elm reservoirs; a preliminary Texas: Copeia, 1962 no. 2, pp. 338-345, 5 text report: Bull. Texas Arch. Paleont. figs. Soc., vol. 20, pp. 2T-62, 8 pis. WENDORF, FRED, KRIEGER, A. D., ALBRITTON, (195'2) The Hogge Bridge site and C. C., JR., and STEWART, T. D. (1955) The the Wylie focus: Amer. Antiquity, vol. 17, no. Midland Discovery: University of Texas Press, 4, pp. 299-312, 6 figs. Austin, 139 pp., 36 figs., 7 tbls. Plate I PLATE I

Vertebrate fossils from T-2 terrace deposits

1, 2. Occiput of Camelops huerfanensis dallasi Lull. x% 24 1. Holotype (SMUMP 60005). 2. Referred paratype (SMUMP 60029). 3. Occiput of ?Camelops sp. (SMUMP 60237). x% 25 4. Fragmentary carapace of Geochelone sp. (SMUMP 60028). Xl/14 38 5-8. Bison alien i Marsh, x '/i 27 5. Horn core (SMUMP 60003). 6. Horn core (SMUMP 60019). 7. Female horn core (SMUMP 60067).

8. Horn core (SMUMP 60317) (approximately l/!). Hill-Shuler Local Faunas Plate I

Index

abbreviations and terms: 4 Bison : 12, 59 : Aenocyon ayersi 12, 17 alleni: 1,12,27-29,56, 60 dirus: 17 antiquus : 27, 60 alba, Tyto : 13, 39 frsso/i: 54, 63 albilabris, Pupoides: 47, 50 latifrons: 27, 56, 59, 60 Albritton clay: 8 occidentalis: 56 Alexander, C. I.: 51, 55 bison, Bison: 54 aliciae, Stentotrema monodon : 42 Boatwright pit: 2, 54 alleni, Bison : 1,12, 27-29, 56, 60 brachycephalus, Megalonyx: 12, 14, 56 Alligator mississippiensis: 13, 38, 56 brachyrhynchos, Corbus: 13, 39 alternata, Anguispira: 42 Brazos River : 8 Alto focus: 51 Breameryx: 12, 56, 60 Amblema [=zCrenodonta] plicata: 42 minimus: 24 Americana, Tanupolama: 27 Brisson: 30 American Museum of Natural History: 4 British Museum Flourine Laboratory: 53 americanus, Mammut: 12, 20, 56 Broadwater, Nebraska: 35 Ursus: 12, 18 Brodkorb, Pierce: 39 Amnicola Integra : 50 Buckner Home—Hackberry Creek terrace (T-5) : anatina, Physa: 50 5 anceps, Helisoma-. 43, 49 bulimoides, Lymnaea: 50 Anderson County: 56 Bulimulus dealbatus: 45, 50 Anglo-American materials: 9 burned hackberry seed: 53 Angostura type points: 51 bursarius, Geomys: 12, 16 Anguispira alternata: 42 Antilocapra : 23 caballus, Equus: 13, 31, 35-36, 53 antiquus, Bison : 27, 60 caliche: 7 antrosa, Helisoma: 43 California: 59 Anura: 13 Institute of Technology Vertebrate Paleon- Aplexa hypnorum: 44, 50 tology Collection: 4 aplodon, Dama virginianus : 21-24, 56 californicum, Arctodus: 18 aquaticus, Scalopus : 12, 13 Smilodon: 19 Aquila chrysaetos: 13, 39 calobatus, Equus: 34 arboreus, Zonitoides: 46 Calvert, E. D., farm: 11 Archaic Stage: 9, 51 Camelops hesternus: 24, 25 Arctodus: 59 huerfanensis dallasi: 12, 24-25, 26 calif ornicum: 18 kansanus: 24 simus: 12, 17-18 sp.: 12, 25-26, 56 Arizona, Papago Springs Cave: 35 canaliculata, Terrapene: 13, 38, 41, 53, 59, 62 armifera, Gastrocopta: 46 canadensis, Castor: 12,16 armigera, Planorbula: 43-44, 50 Lynx: 20 Asinus conversidens: 13, 29-30, 31, 32-33, 34, 35, caneloensis, Canis: 17 36, 42 Canis caneloensis: 17 latrans : 16, 17 francesi: 33, 34, 35, 36 lupus: 17 hydruntius: 33 niger: 16, 17 onager: 30 sp.: 12, 16, 17 asinus, Equus : 30 caperata, Lymnaea: 44, 50 AufTenburg, Walter: 14, 38 Carolina, Terrapene : 13, 38, 62 Austin chalk: 6 carolinensis, Sciurus: 12, 15 australis, Synaptomys : 16 Carrollton: 5 avara, Succinea: 45 focus: 51 Aves: 39 artifacts: 54 carved stone heads: 54 ayersi, Aenocyon: 17 Carychium exiguum : 45 exile : 45 Barbour and Schultz: 35 clay: Castor canadensis: 12, 16 Beaumont 8, 9, 55, 57 Cayuga, Anderson County: 56 Bee County: 9, 47 charcoal: 52 bellus, Dasypus: 12, 14 Cheatum, Elmer P.: 42 Berclair local fauna: 2 chrysaetos, Aquila: 13, 39 locality: 57 circumstriatus, Gyraulis: 43 terrace deposits: 9, 47 Clark, J. F. Gates: 40 Bexar County: 9, 38, 58 clay balls: 40 72 Report of Investigations—No. 48

Clear Fork gouge type scrapers: 51 quinni: 13, 31, 32, 33, 35 climatic implications: 9, 50, 59 scotti: 30 Clovis point: 52 semiplicatus: 34 cobble fields: 9 sp.: 13, 32 Coluber constrictor: 13, 39 tau: 33, 34, 35, 36 columbi, Elephas: 12, 20, 52, 53, 56 Euconulus fulvus: 50 complicatus, Equus: 56 exiguum, Carychium: 45 compressus, Platygonus: 21 exile, Carychium: 45 Connolly, J. L.: 2 Coluber: 39 constrictor, 13, Smilodon: contracta, fatalis, 19 Gastrocopta: 46, 50 Ferrissia rivularis: conversidens, Asinus: 50 13, 29-30, 31, 32-33, 34, 35, flagellum, Masticophis: 39 36, 42 flood plain (T-0) : 9 Equus: 34, 35 5, cooperi, Synaptomys: floridanus, Neotoma: 15 16 Sylvilagus: Coppel 3, 12, 21 locality: 11 tests: 6 coprolite: 40 fluorine-uranium-nitrogen 1, 5, Foard County: 47 Corvus : 39 brachyrhynchos 13, Norman: 39 couesi, Dama virginianus: 23 Ford, fossil man: 51-54 coyote: 16 francesi, Asinus: 33, 34, 35, 36 crassiscutata, Geochelone : 38 fraternus, Equus: 13, 29, 30, 31-32, 33, 56 Crenodonta perplicata: 42 Friesenhahn Cretaceous uplands: 5 Cavern: 38 Crook and Harris: 52 local fauna: 2, 9 Cynomys 14-15 locality: 58-59 ludovicianus: 12, frog bones: 39 frustulosa, Quadrula: 42 Dallas— Frye and Leonard: 50 Museum of Natural History: 4 Fulton aspect: 51 Prehistorical Society Collections : 4, 13 fulvus, Euconulus: 50 dallasi, Camelops huerfanensis : 24-25, 26 Fusconia undata: 42 dalli, Lymnaea: 44 Dal quest, Walter: 23, 47 Dal ton type points: 51 Garza Dam: 52 Dama couesi : 23 —Little Elm Dam: 11 texanus: 22, 23 Gastrocopta armifera: 46 virginianus aplodon: 12, 21-24, 56 contracta: 46, $0 pentodon: Dasypus bellus : 12, 14 46-47 : dealbatus, Bulimulus : 45, 50 procera 47, 50 Denton County: 5 sterkinia: 50 Creek: 11 Geochelone: 13, 38, 58, 62 Didelphis marsupialis : 12, 13 crassiscutata: 38 Dinobastis serus: 58 geographica, Graptemys: 13, 37 dioscoricola, Pupisoma: 49, 50 Geomys bursarius: 12,16 dims, Aenocyon: 17 Germany, Rixdorf: 8 Gibson aspect: 51 Eagle Ford Gifford-Hill pit: 3,10 shale: 6 Glyptodon: Eagle Mountain Lake: 53 12, 13-14 Good farm: East Texas State College: 56 11 —~ Gopherus: 62 Ecuador: 37 gossipinus, Peromyscus: 12, 15 Edgewood type points: 51 Grand Prairie: Elephas: 58 11 columbi: 12, 20, 52, 53, 56 Grapevine Creek: 11 imperator: 20 Graptemys geographical 13, 37 Elam focus: 51 pseudographica : 37 Ellerman and Morris-Scott: Grayson, John: 40 30 region: Elliott, Harry W.: 4 Great Plains 50 Elm Fork, Trinity River: 5 grosvenori, Succinea: 50 Equidae: 29—36 Gulf of Mexico: 8 Equus asinus: 30 Gyraulus circumstriatus: 43 parvus-. 42-43, 49 caballus: 13, 31, 35-36, 53 gyrina, 50 calobatus: 34 Physa: 44-45, complicatus: 56 conversidens: 34, 35 Harvard University: 53 fraternus: 13, 29, 30, 31-32,33, 56 Hawaiia minuscula: 46, 50 lambei: 32 Haymarket pit: 11 littoralis : 33, 34, 36 Hay, 0. P.: 24 midlandensis: 13, 30 31, 33 hearths and hearth areas: 52 occidentalis : 60 Helicina orbiculata tropica: 47-48, 50 onager-. 30 Helicodiscus parallelus: 47, 50 Index 73

Helisoma anceps: 43, 49 Llano Culture site: 52 antrosa : 43 Lonsdale, John T.: 3 trivolvis: 43, 50 Los Angeles County Museum: 4 lentum : 43 lotor, Procyon: 12, 18—19 Hemionus: 30 Love Field—Bethel terrace (T-4) : 5 Henderson County: 1, 5, 9 ludovicianus, Cynomys : 12, 14-15 Henrietta focus: 51 Lull, R. S.: 24 hesternus, Camelops : 24, 25 Lundelius, E. L.: 3, 56 Hibbard, C. W.: 13, 21,33, 34,35, 59 lupus, Canis : 17 Hibbard and Taylor: 30, 33 Lymnaea bulimoides: 50 Hickory Creek: 41 caperata: 44, 50 locality: 3, 11 dalli: 44

site: 52-53 humilis modicella : 44 Hickory terrace: 11 obrussa: 44 hildrethiana, Physa gyrina: 45 Lynx canadensis: 20 Hill member: 6, 7 rufus : 12, 20 Hill-Shulerlocal 56 faunas: 2,10, 29, Magnolia Petroleum Company: 8 hispidus, Sigmodon: 58 12, 15-16, Malakoff: 9 Historic Stage: 52 51, Mammut: 58 Holloman gravel pit: 34 septentrionalis: americanus: 12, 20, 56 Holmesina 12,14, 56 Peromyscus: sapiens: maniculatus, 12, 15 Homo 53 marsupialis, Didelphis: Hudson, Virgil: 2 12, 13 Masticophis flagellum : 39 huerfanensis dallasi, Camelops : 12, 24-25, 26 humilis modicella, Lymnaea: Mastodon americanus: 12, 20 44 Masurian Interstadial deposits: 8 Hunt County: 56 9, McKittrick local fauna: 57 hydruntius, Asinus: 33 Megalonyx brachycephalus: 56 hypnorum, Aplexa: 44, 50 12, 14, Mephitis mephitis : 12, 19 identata, Retinella: 46 Meserve type points: 51 Illinoian glacial stage: 58 Mesomphix : 42 imperator, Elephas: 20 Mesodon thyroidus : 48, 50 Indian village (Tawakoni) : 9 Mexico: 35 Ingleside local fauna: 2 Microtus (Pitymys) pinetorum: 12,16 locality: 57 Midland: 60 pond deposits: 9 midlandensis, Equus: 13, 30-31, 33 insects: 40 Midwestern University: 23, 47 Integra, Amnicola: 50 Milton pit: 11 Iron Bridge Dam: 9 minimus, Breameryx: 24 local fauna: 2 miniscula, Hawaiia: 46, 50 locality: 56-57 mirifica, Tanupolama: 27 mississippiensis, Alligator: 13, 38, 56 Jinglebob— modicella, Lymnaea humilis : 44 flora: 40 monodon, Stenotrema: 42 local fauna: 18 Moore pit: 3,10—11, 40, 42 locality: 59 Moore, Roy: 2 Johnson, Richard: 53 Morrison, JosephP. E.: 42 Kansan glacial: 34 Mustela vison : 12, 19 kansanus, Camelops: 24 Mylohyus: 21, 58 Kansas: 35, 40 Nannippus: 35 Kaufman: 52 phlegon: 34, 35 Kerrville: 16 Nebraska, Broadwater: 35 Lagow man: 53 Neo-American Stage: 51 pit: 3, 10, 53 Neotoma floridanus: 12, 15 lambei, Equus: 32 Niger, Canis: 16, 17 Sciuris: latifrons, Bison : 27, 56, 59, 60 12, 15 latrans, Canis: 16, 17 nitidum, Pisidium: 50 leai, Stenotrema: 42, 48 North Texas State College: 52 lentum, Helisoma trivolvis: 43 obrussa, Lymnaea: 44 Leonard, A. Byron: 3 occidentalis, Bison: 56 Leon River: 8 Equus: 60 Lepus: 12, 21, 42 Oelrich, Thomas: 38 Lepisoteus spatula: 13, 39 Onager: 30, 32 leucopus, Peromyscus: 12, 15 littoralis: 34 Lewisville site: 3,11, 40, 41, 42, 52, 53 semiplicatus: 34 Lissie formation: 55 onager, Asinus: 30 littoralis, Equus: 33, 34, 36 Equus: 30 Onager: 34 Ondatra zibethicus: 12, 16 llanensis, Terrapene: 59 orbiculata tropica, Helicina: 47-48, 50 74 Report of Investigations—No. 48

Padgitt, Tom: 53 Sangamon deposits: 35, 41 Paleo-American points: 51 interglacial: 1, 50 pallidicinctus, Tympanuchus: 13, 39 San Patricio County: 9 Panthera: 58 Scalopus aquaticus: 12, 13 Papago Springs Cave, Arizona: 35 Scharbauer site: 60 parallelus, Helicodiscus: 47, 50 Sciurus carolinensis: 12, 15 parvus, Gyraulis: 42-43, 49 niger: 12, 15 Paramylodon: 12, 14 Scott, Charles: 3 Pattillo, L. F., Jr.: 5 scotti, Equus : 30 Pattillo sand: 8 Scottsbluff type points: 51 paucilirata, Retinella identata: 46 Seagoville site: 53 Pemberton Hill: 3,10 Seale pit: 3, 11 site: 53 Seale, W. Q.: 11,17 Pemberton-Hill—Lewisville terrace (T-2) : 1, semiplicatus, Equus: 34 5, 6 Onager: 34 deposits: 51 Sellards, E. H.: 54, 57 pentodon, Gastrocopta: 46-47 septentrionalis, Holmesina: 12,14, 56 Peromyscus gossipinus: 12, 15 serus, Dinobastis: 58 leucopus: 12, 15 Shuler, E. W.: 5,10, 53 maniculatus: 12, 15 Shuler member, Upper and Lower: 6, 7 perplicata, Crenodonta: 42 shuleri, Tetrameryx: 9, 12, 24, 56 Peterson, Miss Hazel: 56 Sigmodon hispidus: 12,15—16, 58 phlegon, Nannippus: 34, 35 simus, Arctodus: 17-18 Physa: 38 skeletal remains: 53 anatina: 50 Skinner and Kaisen: 56 gyrina: 44-45, 50 Skinner, M. F.: 35 hildrethiana: 45 small creek terraces: 49, 51 pinetorum, Microtus (Pitymys) : 16 Smilodon trinitiensis : 12, 19-20 Pisidium nitidum: 50 californicus: 19 Platygonus compressus: 12, 21 fatalis: 19 Plainview points: 51, 53 floridanus: 20 Planorbula armigera: 43-44, 50 gracilis : 19' plant fossils: 40-41 troglodytes: 19, 20 plicata, Amblema: 42 Smithsonian Institution: 40, 42 pollen: 40 Socony Mobil Company: 8 Polygyra texasiana: 42, 48, 50 Research Laboratory: 40, 51 Power Plant terrace: 55 Southern Methodist University: 42 procera, Gastrocopta: 47, 50 Biology Department: 4 Procyon lotor: 12, 18-19 Museum of Paleontology: 4,13 pseudographica, Graptemys: 37 Spanish explorations: 9 Pupoides albilabris: 47, 50 sparsicosta, Strobilops: 47 Pupisoma dioscoricola : 49, 50 spatula, Lepisosteus: 13, 39 Pyburn, William: 38 Sphaerium striatinum : 48, 50 Quadrula frustulosa: 42 speciosa, Qiiadriila: 42 speciosa: 42 Spermophilus tridecemlineatus : 12, 15 Quaternary deposits: 6 Starrett, L. P., Jr.: 2 Quinn, J. H.: 17, 29, 33 Stenomastodon: 35 quinni, Equus? : 13, 31, 32, 33-35 Stenotrema aliciae: 42 radiocarbon dates: 1, 8, 9, 36, 51, 52, 53, 57, 60, 62 leai: 42, 48 rainfall: 40 monodoii: 42 map: 2 sterfcinia, Gastrocopta procera : 50 Rancho La Brea: 19 stevensi, Tanupolama: 27 locality: 59-60 Stockoceras: 23, 60 Rancholabrean: 59 Stovall and McAnulty: 9,13, 54, 55, 56, 57 Red River: 8 striatinum, Sphaerium: 48, 50 Retinella identata: 46 Strobilops sparsicosta: 47 paucilirata: 46 Richards texasiana : 47, 50 member: 6, 7 Sulphur River Basin Survey Party: 13 River: 8 rivularis, Ferrissia: 50 Succinea avara: 45 Rixdorf, Germany: 8 grosvenori : 50 Roulin, Desire: 37 Sylvilagus floridanus : 12, 21 roulini, Tapirus: 36, 37 Synatomys : 12, 6'2 rufus, Lynx: 12, 20 australis : 16 Sabine River: 8, 9, 56 cooperi : 16 Saldiver, Theodore: 2 sapiens, Homo: 53 T-0 deposits: 51 Sanders focus: 51 T-0 terrace. See flood plain. Index 75

T-l deposits: 51-52 trivolvis, Helisoma: 43, 50 terrace. See Union Terminal—Carrollton ter- troglodytes, Smilodon : 19, '2O race. floridanus, Smilodon : 20 T-2 deposits: 52-54 gracilis, Smilodon-. 19 See Pemberton Hill—Lewisville terrace. ter- tropica, Helicina orbiculata : 47-48, 50 race. Truncilla vertebrate fossils from, list of: 12-13 truncata: 42 37 T-3, T-4, T-5 deposits : 54 turtles: T-3 terrace. See Travis School—Farmers Branch Tympanuchus pallidicinctus : 13, 39 Tyto alba: T-4 terrace.terrace. See Love Field—Bethel terrace. 13, 39 T-4 terrace. See Love Field—Bethel terrace. T-5 terrace. See Buckner Home—Hackberry undata, Fusconia: 42 T anupolama:Creek terrace.12, 26-27 University of California: 4 Tanupolama: 12, 26-27 of Florida: 14, 39 americana : 27 of London: 6 mirifica : 27 of Michigan: 38, 39 stevensi: 27 Union Terminal—Carrollton terrace (T-l) : 5, 9 Tapirus : 13, 36-37, 60 Ursus : roulini: 37 americanus 12, 18 U. S. National Museum: 13 terrestris : 36, 37 Tarrant County: 5, 53 Uyeno and Miller: 39 Archeological Society: 53 tau, Equus : 33, 34, 35, 36 Van Zandt County: 9, 56 Tawakoni Village (Indian) : 9 vertebrate fossils, list of from T-2 deposits: 12-13 Terrapene canaliculata: 13, 38, 41, 53, 58, 59, 62 virginianus aplodon, Dama: 12, 21-24, 56 Carolina-. 13, 38, 62 vison, Mustela: 12, 19 : 59 llanensis Wagner Free Institute of Science: 4 terrestris, Tapirus: 36, 37 Wells type points: 51 Tetrameryx : 60 Wentworth shuleri: 9,12,24, 56 grain size classification: 4 West Fork of Trinity texanus, Dama virginianus: 22, 23 River: 5 Texas Memorial Museum: 4, 54 Wheeler site: 51 texasiana, Polygyra: 42, 48, 50 White, Theodore: 11,13 Strobilops: 47, 50 Williams, Thomas: 38 thyroidus, Mesodon: 48, 50 Wisconsin interglacial: 1, 50 Tordoff, H. B.: 39 wolf: 17 Travis School—Farmers Branch terrace (T-3) : 5 Wood pit: 3, 11, 40, 42-50 tridecemlineatus, Spermophilus: 12,15 Wurm of Fourth Glacial: 8 Trinidad: 1, 2, 9, 54 Wylie 51 locality: 55 focus: terrace: 55 trinitiensis, Smilodon-. 12, 19-20 Yarbrough type points: 51 Trinity— Yarmouth interglacial: 35 River: 1 terrace: 2, 54, 55 Zonitoideszibethicus, Ondatraarboreus: : 12,1646 type points: 51 Zonitoides arboreus : 46