BULLETIN 20 OF THE TEXAS MEMORIAL MUSEUM RED LIGHT LOCAL FAUNA (BLANCAN) OF THE LOVE FORMATION, SOUTHEASTERN HUDSPETH COUNTY, TEXAS William A. Akersten Submitted for publication February 23, 1970 THE TEXAS MEMORIAL MUSEUM/W. W. NEWCOMB, JR., DIRECTOR 24th & Trinity, Austin, Texas 78705/The University of Texas at Austin CONTENTS Page ABSTRACT 1 INTRODUCTION 1 Location Acknowledgments 1 Techniques 2 Abbreviations 2 Previous work 2 Geologic setting 2 STRATIGRAPHY 2 Bramblett Formation 4 Love Formation 4 Measured sections 7 9 STRUCTURE SYSTEMATIC PALEONTOLOGY 9 Locality descriptions 9 Aguila local fauna 11 Red Light local fauna 12 40 AGE AND CORRELATION PALEOECOLOGY 43 HISTORY OF THERED LIGHT BOLSON 47 SUMMARY AND CONCLUSIONS 50 REFERENCES CITED 51 ILLUSTRATIONS Figure Page 1. Physiography and mapped areas near Red Light Bolson 3 2. Geologic map of area studied 5 3. Cross-section of Love and Bramblett Formations 6 4. Outcrop of Love Formation, TMM 40664 10 5. Close-up of figure 4 10 6. Cross-bedded gravel, TMM 40664 11 7. Outcrop of Bramblett Formation 14 8. Edentata 15 9. Glyptotherium texanum 21 10. Geomys (Nerterogeomys) paenebursarius 22 11. Rodentia 23 12. Carnivora 28 13. Carnivora and Artiodactyla 32 14. Unnamed long-limbed camel 35 15. Mastodont, Leporidae, and Equus 37 16. Equus and Nannipus 38 TABLES Table Page la. Comparison, lower part of Red Light l.f. and Hudspeth l.f. of Fort Hancock Formation lb. Comparison,upperpart ofRedLightl.f. and Hudspethl.f. of Camp Rice Formation , 13 2. Measurements, mandibles of Geomys (Nerterogeomys) 17 3. Measurements,lowerdentitionsofSigmodon 19 4. Measurements, mandibles of Urocyon 20 5. Measurements, mandibles of Canis 20 6. Measurements, fibulae of large felids 26 7. Measurements, mandibles of Fells cf. rexroadensis 26 8. Measurements, radii-ulnae of long-limbed camels 30 9. Measurements, mandibles of Camelops 33 10. Measurements, upper deciduous dentition of Equus 33 11. Measurements, upper dentitions of Equus 34 12. Measurements, metacarpals 111 of Equus 40 13. Measurements, second phalanges ofEquus 41 14. Measurements, first phalanges ofEquus 42-43 15. Measurements, lower deciduous dentitions ofEquus 44 16. Measurements,radii-ulnae ofEquus 44 17. Measurements, ungual phalanges ofEquus 45 18. Measurements,lowerteethofEquus(Asinus)cumminsi 46 19. Measurements, second phalanges ofNannippus phlegon 47 20. Measurements,lowerteethofNannippusminor 48 21. Comparison, Red Light and Hudspeth local faunas 49 RED LIGHT LOCAL FAUNA (BLANCAN) OF THE LOVE FORMATION, SOUTHEASTERN HUDSPETH COUNTY, TEXAS BY WILLIAM A. AKERSTEN ABSTRACT Two new formations in the Red Light Bolson have yielded vertebrate fossils. The Red Light local fauna consists of thirty taxa of mammals and nu­merous lower vertebrates from fluvial deposits.The much smaller Aguila local fauna is from older playa deposits. The stratigraphic and faunal successions indicate a climatic change from arid playa conditions to moist fluvial conditions. This change is probably a result of the onset of pluvial conditions related to glaciation at higher latitudes. At the time that the youngest part of the Red Light local fauna lived, the bolson was occupied by a permanent,flowing drainage and three major terrestrial environ­ments probably existed in the area. Brush or succu­lent vegetation grew along the drainage, a savannah with scattered trees existed between the drainageand the mountains, and the mountains were largely covered with brush. The playa and fluvial deposits are correlated with the Fort Hancock and Camp Rice Formations,respectively, in the Hueco Bolson. The Red Lightlocal fauna correlates with the Hudspeth local fauna from the Hueco Bolson. Faunal evidence indicates that it lived during the Pleistocene por­tion of the Blancan and, from stratigraphic evi­dence, this is narrowed to the Nebraskan. INTRODUCTION Pleistocene and upper Pliocene vertebrates from the bolson deposits of the southwestern United States and adjacent Mexico are not well known. In Texas only the material from the Hueco Bolson has been described (Strain, 1966). The Mesilla Bol­son in New Mexico has yielded some Pleistocene material, mostly undescribed, Gidley (1922 and 1926), Gazin (1942), and many others have de­scribed vertebrates from the Benson and Curtis Ranch faunas from bolson deposits along the San Pedro River valley in Arizona. A few small col­lections have been described from other bolsons. Underwood (1963), mapping in the area of the Eagle and Indio Mountains in southeastern Hud­speth County, Texas, discovered Pleistocene verte­brate remains in the sediments of the Red LightBolson. In 1960, J. A. Wilson, J. Paulson, W, S. Strain, and J. R. Underwood made preliminary col­lections. I collected additional material during 1964 and 1965. The faunas represented by these collec­tions are herein designated the Red Light and Aguila local faunas. Location. Location.— The Red Light Bolson (Underwood,1963) is in a graben or half-graben trending north­west-southeast in southeastern Hudspeth County,Texas, between the Devil Ridge-Eagle Mountains-Indio Mountains chain to the northeast and the Quitman Mountains to the southwest (fig. 1). The partofthe bolson underconsideration inthispaperis bounded by the Rio Grande to the southwest and south, the front of the Indio Mountains to the northeast, and about latitude 31° 47' N. to the north (fig. 2). This area is included in Underwood’s map (1963). It is on parts of the Bramblett and Guerra ranches. Acknowledgments. Acknowledgments.— This problem was supervisedby Dr. E. L, Lundelius of the University ofTexas. His suggestions, advice, and patience are deeply ap­preciated. I am indebted to J. A. Wilson and Prof. R. K. DeFord of the same institution, and members of the committee, for their advice and constructive criticism. Dr.E. C.Jonas ofthe UniversityofTexas ran the X-rayclay analysis. J. G.Lundberg ofthe Universityof Michigan identified the fish. P. Brodkorb of the University of Florida provided a preliminary list of the birds. M. F, Skinner of the Frick Laboratory,American Museum of Natural History, generouslyprovided advice on the taxonomy of equids. J. G. Mead and K. Kutasi prepared the photographic il­lustrations. Mrs. M. S. Stevens, MSS, drew all but one ofthefaunal illustrations and contributed much advice on preparation and osteology. E. J. Deemer,EJD, drew one illustration. I am grateful for their assistance. Discussions with those just named, as well as with Dr. C. Hubbs, Dr. D. Reaser, Dr. P. E. Twiss, and Dr. J. B. Stevens, proved enlightening. The following people loaned specimens, providedcasts, or made collections available during personalvisits: Dr. T. Downs and Dr. J. R. Macdonald of the Los Angeles County Museum, Dr. D. E. SavageoftheUniversityofCaliforniaatBerkeley,Dr. D. S. Webb of the Florida State Museum, and Dr. C. W. Hibbard ofthe University ofMichigan, I should also like to acknowledge with thanks Dr. Hibbard’s ad­vice and critical reading of the manuscript. Permission to work on their land was graciouslygiven by Leo and Robert Guerra and Mr. and Mrs. John D. Bramblett. The hospitality of the Bramb­letts made life in the field more pleasant. Financial support was provided by the Depart­ment of Geological Sciences, University of Texas at Austin and an NDEA Title IV Fellowship. A Grant-in-Aid of Research from the Society of the Sigma Xi made possible the preparation and illustration of this report.Techniques. Techniques.—l collected larger vertebrate bones by walking and searching the outcrops. Only one of the more than 700 specimens collected included any associated material. No locality is rich enoughin the remains of large vertebrates to warrant quarrying. The micro-vertebrates were collected bysurface picking and, at locality TMM 40857, by dryscreening and washing. The horse teeth were prepared for sectioning in the following manner. They were vacuum impreg­nated with a thin solution of polyvinylacetate and alcohol, coated with a thin layer of petroleum jelly,imbedded in a dental casting plaster, and vacuum im­pregnated a second time. The petroleum jelly al­lowed the plaster to be easily removed after section­ing. Other preparatory techniques were standard. Large measurements were made with wooden calipers calibrated in millimeters. Small measure­ments were made with steel calipers calibrated in 0.1 mm. The measurements recorded in 0.01 mm were measured with a Gaertner measuring micro­scopewhich measured inincrementsof0.001mm. Abbreviations. Abbreviations.— For measurements, mm and cm are used for millimeters and centimeters. Abbrevia­tions referring to paleontological collections are: AM American Museum TMM Texas Memorial Museum CIT California Institute ofTechnology (now at the Los Angeles County Museum) MSS Margaret Skeels Stevens (privateRecent collection)UCMP University ofCalifornia Museum of PaleontologyUSNM United States National Museum Previous work. —The only publication which has much bearing on the deposits of the Red LightBolson is that of Underwood (1963). Albritton and Smith (1965) worked farther to the north where the older deposits of the Red Light Bolson do not crop out, Underwood (1963) also mapped the de­posits in the western part of the Green River and associated bolsons. Twiss (1959) mapped the de­posits in the eastern part of these bolsons. DeFord and Bridges (1959) described the Tarantula Gravel from the Green River Bolson. Much work, mostlyunpublished University of Texas theses, has been done in the bolson deposits farther down the Rio Grande. Dickerson (1966) is the only one of these which treated the bolson fill at great length.TotheeastofUnderwood’s area,Jones(inmanu­script) mapped the western portion of the Red Light Bolson. Strain (1966) mapped an area of the Hueco Bolson and summarized previous work on that bolson. Bell (1963) mapped an area of the Hueco Bolson to the east of the area mapped by Strain. In Mexico, Reaser (in manuscript) has mappedthe fill in the southern part of the Red Light Bol­son. Haenggi (1966) mapped in the El Cuervo Bol­son and the western part of the Presidio Bolson but only very young fill is exposed in the El Cuervo Bolson.Figure 1givesthelocationsofthesemapped areas. Geologic setting.setting.—The following data are from Underwood (1963) and personal observation unless otherwise noted. East of the outcrop area of bolson fill under consideration inthis paper are the Indio Mountains. Lower Cretaceous conglomerate, sandstone, shale,and limestone and Tertiary volcanic rock, mostly trachyte and tuff, crop out in these mountains. The only outcrops contributing to the modern al­luvium in the Red Light Bolson in the area studied are the siliceous conglomerate, sandstone, and shale of the Lower Cretaceous Yucca Formation. The Yucca makes up most of the western Indio Mountains and is the only formation adjacent to the bolson fill in the area studied. It dips to the east so that the lower part of the section crops out closest to the bolson fill. The lower Yucca is mostlyconglomerate; it becomes progressively finer grainedhigher in the section, farther from the bolson. The outcrops at Eagle Mountain, to the north, are mostly Tertiary syenite, trachyte porphyry,and rhyolite. Northwest of the Eagle Mountains are outcrops of Lower Cretaceous conglomerate,sandstone, and limestone including those of Devil Ridge. The outcrops along the east side of the Quitman Mountains are of the same general nature as in the Eagle-Indio Mountains except that the Yucca is not exposed (D. Reaser, oral communica­tion, 1967). STRATIGRAPHY New names,, the Bramblett Formation and the Love Formation, are herein proposed for two litho­logic units within the Red Light Bolson. The names proposed by Strain (1966) for lithologic units of similar ages and similar gross appearances in a near­by area of the Hueco Bolson should not be appliedto the units in the Red Light Bolson for the fol­lowing reasons: The two bolsons represent differ­ent basins of deposition with different source areas for the sediments. Although the compositions of corresponding formations in the two bolsons are grosslysimilar,theydifferin detail.Thedifferences are a reflection of the source areas and slight differ­ences in the environments of deposition. If the formational names used in the Hueco Bolson were applied to the lithologic units in the Red LightBolson, it would be done because of similar geo­logic history and time equivalence. These are not validcriteriafor therecognition ofrock-stratigraphicunits (American Commission on Stratigraphic No­ Fig. I.—Physiography and mapped areas near the Red Light Bolson (after Bell, 1963). Index to mapped areas; 1, Strain (1966); 2, Albritton and Smith (1965); 3, Jones (in manuscript); 4, Underwood (1963); 5, Bell (1963);6,Reaser(inmanuscript); 7,Twiss(1959);8,Haenggi(1966);9,Dickerson(1966). menclature, 1961, Art. 4), The objection may be made that the naming of newformationsineachofthebolsonsofthe South­westwill lead to aproliferationofformalnamesin the geologic literature. This is unavoidable because oftherulesfordefiningandrecognizing aformation. Moreover, it is desirable from the standpoint of ac­curacy and clarity. Informal names are often simi­lar to each other and are difficult to keep distinct in a discussion that includes the informal names of several authors. Application of the same name to similar units in different basins of deposition leads to the same problem and is often a source of in­accuracies. An example is the use of “Santa Fe marls,” or “group,” or “formation” to include a number of different units of different ages in dif­ferent basins of deposition. This has led to much confusion in the interpretation and relationshipsof these units (see Strain, 1966). Bramblett Formation. —The Bramblett Formation (new name) is named for the John D. Bramblett Ranch. Most of the outcrops in the United States part of the bolson, the type section, and the refer­ence section are on this ranch. The designated type section, BR-1, is located about three-quarters of a mile north-northwest of the Bramblett ranch house and about one-quarterof amilenortheastofthecountyroadtotheranch house. It is the first good outcrop going northwest along the road from the ranch house. The reference section, BR-2, is about 200 yards northeast of Dome Tank. The localities for both sections are shown in figure 2. Descriptions of the sections are given under the heading “Measured Sections.” The main outcrop areas of the Bramblett Forma­tion are to the southeast of Aguila Draw and alongthe county road. The total exposed thickness is probably less than 250 feet in this area, but the base is not exposed. Figure 3 is a generalized cross-section. A facies of irregularly bedded silt, sand,and gravel crops out in a narrow belt along the mountain front. The gravel, which makes up the bulk of this unit, is angular to subround and is de­rived from adjacent outcrops of siliceous sandstone and conglomerate of the Lower Cretaceous Yucca Formation. This facies rarely extends more than 200 yards from the mountain front before inter-fingering with or grading laterally into a facies of calcareous clay, silt, and fine sand. This fine-grainedfaeies contains a greater percentage of clay and be­comes salty and gypsiferous toward the axis of the bolson. The clay is yellow, green, and reddish brown. The silt and fine sand are buff to light brown. Most beds are lenticular and of small lateral extent. Cor­relation of scattered outcrops is usually impossible. Toward the top of the formation the fine-grained facies contains more sand and silt and grades into overlying fluvial deposits. No significant break in sedimentation has been found although most silt and sand beds have channeled lower contacts. The top of the uppermost green clay in the section has been arbitrarily chosen as the top of the Bramblett Formation. This green clay is widespread, six inches to two feet thick, and is not bedded or fissile. An­other, similar but slightly fissile clay is usuallypresent about two feet lower in the section. An X-ray analysis of this bed showed strong peaks for montmorillonite, quartz, and calcite; and weaker peaks for kaolinite and illite (E. C. Jonas, oral communication). In most outcrops, there is a zone about one inch thick, of gastropod and plant frag­ments near the top of this lower clay. The Aguilalocal fauna was collected from this zone. In several small outcrops between Arroyo Escudo and AguilaDraw, the zone is underlain by a continuous layerof caliche about one-quarter ofan inch thick. The Bramblett Formation was deposited in a closed basin. Rock fragments from the surroundingmountains were deposited close to the mountain front as alluvial fans. The center of the basin was occupied by a playa lake in which clay, silt, and fine sand were deposited. Some of the silt and fine sand was probably deposited in feeder channels of the playa lake. The characteristics of the Bramblett Formation are those enumerated by Wright (1946)for sediments deposited in a closed basin. The cli­mate was probably quite arid during the depositionofthis formation. The Bramblett Formation is similar in strati­graphic position and type of deposition to the Fort Hancock Formation in the Hueco Bolson (Strain, 1966)tothe east.Thetwodifferinthattheywere deposited in different basins, and their sediments had different source areas. Also, the top of the Fort Hancock Formation appears to be younger, for it contains a fauna which is known only above the Bramblett Formation. The bolson fill described by Dickerson (1966) appears to have been deposited during an older partofthe sameclimaticintervalin whichtheBramblett Formation was deposited. Love Formation. —The Love Formation (newname) is named for the Love triangulation station located at lat. 30° 48' 27.253" and long. 105° 12'12.150"inthe RedLight Bolson. Thetypesection,RL-1,islocatedinTMM local­ity 40664, on the John D. Bramblett Ranch, along a narrow stream cut through the locality. The ref­erence section, RL-2, is on the John D. Bramblett Ranch onthe eastside ofArroyo Escudo about two and one-quarter miles upstream from the pointwhere the arroyo crosses the road. The localities of both sections are shown in figure 2 and their descrip­tions are given under the heading “Measured Sec­tions.” The main outcrop area is to the northwest of Aguila Draw. Probably less than 200 feet ofsection are exposed in this area. The base is arbitrarily de­fined as the top of the highest green clay. Figure 3 is a generalized cross-section. Two major facies Fig. 2.—Geologic map ofa portion of the Red Light Bolson (after Underwood, 1963). Formations. Bramblett and Love the of cross-section generalized 3—Very Fig. makeup the Love Formation. A facies ofmoderate­ ly consolidated reddish silty clay and gravel cropsout along the mountain front and up to a mile into the bolson. Gravel is dominant in this facies next to the mountain front and decreases toward the center of the bolson. Where this facies interfingerswith the other major facies, the gravel is in isolated pieces and small pods within the silty clay. This facies has minor amounts of sand throughout and some caliche as nodules or incipient nodules. I have not seen any sedimentary structures. Beds are usual­ly thick with indistinct boundaries. The reddish silty clay and gravel facies interfmgerswith a fluvial facies of silt, sand, and gravel with some silty clay (figs. 4,5). Most beds are lenticular and correlation between scattered outcropsis usual­ly not feasible. The lower part of the fluvial facies has little gravel. The gravel beds are thickest, up to five feet, near the reddish silty clay facies and thin toward the center of the bolson. The gravel is de­rived from Lower Cretaceous limestone, from the siliciclastic rocks of the Lower Cretaceous Yucca Formation, and from caliche. Many carbonate peb­bles are etched. Red and green clay balls up to three feet in diameter are common; they appear to be derived from the Bramblett Formation and in­dicate that this formation was being eroded up­stream, Most of the thinner beds are uncemented,and most of the thicker ones are cemented by cal­cite. Most of the gravel beds contain a single set of cross-bedding (fig. 6), which indicates that they were deposited by a strong current during a singleevent of short duration, such as a flood. The majorcross-bedding direction of all the sediments within this facies is to the southeast. Much of the silt and fine sand is cross-bedded. Cementation by calcite is sporadic. Some fine silt beds are weakly cemented by caliche. Caliche nod­ules are common in some beds. Lenses of reddish,unconsolidated, silty clay occur throughout this facies but are more common near the base. The upper part of the Love Formation has been removed by erosion. Most outcrops are capped bypediment or terrace deposits. The pediment surfaces dip toward the center of the bolson so that the up­permostpreservedpartsoftheformation are young­ernear themountainfront. The Love Formation was deposited in a basin with a through-flowing axial stream. The drainage appears to have developed gradually rather than catastrophically, for the transition from playa de­posits to fluvial deposits is gradational. Therefore,Iinterpretthischangetobetheresult ofan increase in rainfall rather than a shift of drainage patterns or stream piracy. The increase of rainfall was, in turn, probably the result of the onset of pluvialconditions. The gravel fraction was derived from Cretaceous outcrops along the bolson and caliche in penecontemporaneous deposits of the bolson. There are no volcanic rock fragments that came from outcrops near El Paso such as Strain (1966)found in gravel of the same age in the Camp Rice Formation in the Hueco Bolson. Therefore, the drainage of the Hueco Bolson did not contribute sediment to the Love Formation in the area studied. Thereddish silty clay facies appears to have formed by a mixture of alluvial fans and flood plain depos­its. The remainder of the formation is fluvial. The Love Formation is similar in age and grosscomposition to the Camp Rice Formation (Strain,1966) in the Hueco Bolson. They differ in the composition of the gravels, in that the base of the Love is younger than the Camp Rice, and in that the base of the Love is gradational, whereas the base of the Camp Rice is marked by an uncon­formity. Also, the Pearlette Ash which Strain found in the Camp Rice Formation has not been found in the Love Formation. Measured Sections.Sections.— Sections were measured with a five-foot Jacob’s staff and a Brunton compass.Descriptions are of hand specimens, not thin sec­tions. If a unit is heterogeneous, a single sampleis described and the limits ofvariation are defined. Where possible, the five-fold system of Folk (1954)and the composition triangle of Mcßride (1963) are used to describe the composition, and the chart of Goddard et al. (1963), to describe the colors. Thicknesses are given in feet. SECTION RL-1 Type Section, Love Formation JohnD. Bramblett Ranch, southeasternHudspethCountyTexas. Section measured along a narrow, almost vertical stream cut through TMM locality 40664, see figure 2. Unit Unit Cumulative Thickness Thickness Top of cut on north side Pediment gravel 9. Medium to fine sandstone, calcitic submature clay pellet sublitharenite. Pinkish gray (SYR 8/1). Moderatelyconsolidated, thin bedded, cross-bedded, several thin units of sand crystals. May be from youngercycle of sedimentation 3 118 8. Slightly gravelly muddy sand, im­mature quartzarenite. Pale reddish brown (10R 5/4). Slightly consoli­dated; bluff former; thick,indis­tinct beds 11 115 7. Slightly silty clay. Grayish red (10R 4/2). Slope former; numer­ous caliche nodules at base, decreas­ing upward 15 104 6. Same as Unit 4 but without clay balls and only lenses ofcaliche gravel . 17 89 5. Slightly sandy conglomerate,calcitic submature caliche litharenite. No colordetermination. Well consoli­ dated, cross-bedded 4 72 Unit Unit Cumulative Thickness Thickness 4. Medium sand, mature cherty quartz­ arenite. Grayish orange pink (10R 8/2). Very heterogeneous unit. Beds, thin to thick; grain size, fine to coasse sand withlenses up to 6 in. of conglomerate and thinclayball lenses, unconsolidated to consolidated, occasional caliche nodules 32 68 . 3. Slightly clayey fine to coarse sand,calcitic submature cherty quartz­arenite. Grayish orange pink (SYR 7/2). Partially indurated, scattered poorly developedcalichenodules . 3 36 2. Slightly sandy clayey siltstone, im­maturecherty quartzarenite.Palered­dish brown (10R 5/4). Moderately con­solidated,bluffformer,rare incipient calichification,thinto thickindistinct beds. Contains lenses, pods, and in­dividual particles of coarse sand and gravel . 21 33 I. Very fine to fine sand, mature cherty quartzarenite. Grayish orange pink (SYR 7/2). Unconsolidated, cross- bedded. Scattered caliche nodules, gravel, and clay balls 12 12 . The base of the section is at the intersection ofan arroyoand a tributary which forms a steep, narrow cut through an almost vertical bluff. The section then follows this cut. SECTION RL-2 Reference Section, Love Formation John D. Bramblett Ranch, southeastern Hudspeth Coun­ty, Texas. Section measured on the east side ofArroyo Es­cudo about two and one-quarter miles upstream from the point where the arroyo crosses the road, see figure 2 Unit Unit Cumulative Thickness Thickness Topofbluffon northside 6. Mediumtofinesand,maturecherty quartzarenite. Grayish pink (5R 8/2). Unconsolidated, contains thin, darker units of clay 8 62.5 5. Silty very finesand, mature quartz­ arenite. Grayish orangepink (10R 8/2). Very heterogeneous unit. Contains beds of silt to coarse sand, some indurated, from 1 in. to 2 ft. thick and lenses of caliche . gravel. Most units cross-bedded . . 30.5 54.5 4. Same as Unit 3 but not indurated, slope former 6.5 24 3. InterbeddedlithologiesofUnits 1 and 2. Beds less than 1 ft. thick, some indurated, bluff former 6 17.5 .... 2. Silty clay. Pale red (5R 6/2). Usuallyblocky but occasionally fissile, bluff former. Contains a few beds up to 1 in. thickof silt as in Unit 1 5.5 11.5 1. Silt, calcific mature quartzarenite. Grayish orangepink (SYR 7/2).Bluff former. Contains thin beds ofclayasinUnit2 6 6 The base of the section is at the intersection of ArroyoEscudo with a tributary which cuts through a bluff. The section follows this tributary. SECTION BR-1 Type Section, Bramblett Formation John D. Bramblett Ranch, southeastern Hudspeth Coun­ty. Measured in an “amphitheater” about three-quarters of a mile north-northwest of the Bramblett ranchhouse and about one-quarter of a mile northeast of the county road to the ranchhouse, see figure 2. Unit Unit Cumulative Thickness Thickness Top of hill Terrace gravel 8. Slightly fissile claystone. Pale grayish olive (10Y 5/2). Waxy luster 1.5 76.5 7. Same as Unit 4, slope former . 20 75 6. Silt, mature quartzarenite. Pale yellowishbrown (10YR 6/2).Slightly indurated, bluff former, cross-bedded. Contains conspicuouslayersof heavy mineralsand occasion­al clay laminae 9 55 5. Moderately fissile shale. Moderate brown (SYR 3/4). Salty 2 46 .... 4. Alternating lithologiesas in Units 1 and 2. Beds 1 to 4 ft. thick, slope former 33 44 3. Slightly silty clay. Grayish olive (10Y 4/2) and grayish red (10R 4/2) intimately mixed,not in layers 1 11 2. Alternating silt and clay. Pale yellowish brown (10YR 6/2) to moderate brown (SYR 4/4). Slope former, very salty . 4 10 1. Clay. Grayish red (10R 4/2). Slope former, very salty 6 6 The base of the section is in the arroyo at the base ofa hill which terminates at the arroyo. SECTION BR-2 Reference Section, Bramblett Formation John D. Bramblett Ranch, southeastern Hudspeth Coun­ty, Texas. Section measured about 200 yards northeastof Dome Tank, see figure 2. Unit Unit Cumulative Thickness Thickness Top ofbluff 6. Sandy and silty caliche. Grayish orange pink (SYR 7/2). Very re­sistant. Sand and silt decrease up­ward. Caliche probably much youngerthan rest of section 1 55.5 5. Slightly sandy silt, calcareous submature quartzarenite.Lightbrown (SYR 6/4). Slope former . 3 54.5 4. Sandy conglomerate to gravelly Unit Unit Cumulative Thickness Thickness coarse sandstone, calcareous submature quartzareniteto lith­arenite. Grayish orangepink(SYR 7/2). Thickindistinct beds,bluff former. Extremely variable between and within beds ... . . 10.5 51.5 3. Gravelly sandstone, calcareous submature quartzarenite tolith­arenite. Grayish pink (SR 8/2)to grayish orangepink (SYR 7/2).Bluff former, thin to thickbeds. Very heterogeneousunit, also contains gravelly clay and silt,silt, and sand . . 21 41 2. Covered . . 6 20 1. Coarse sandy silt to silt, calcareous submature cherty quartzarenite.Light brown (SYR 6/4).Slope former . . 14 14 The base of the section is at the first outcrop below the first bluffnortheast ofDome Tank. STRUCTURE Underwood (1963) discussed the fault between the bolson fill and the Indio Mountains. The fill,which he described as dipping toward the fault, is the conglomeratic facies of the Bramblett Forma­tion. This dip is in the opposite direction from that which would be expected from drag associated with movement along a gravity fault. There are numerous normal faults in the Love Formation near the moun­tain front. They dip both toward and away from the mountains. The total displacement of the faults within the Love Formation appears to be less than 100 feet. Much of the Bramblett Formation is cut by nor­mal faults, each of which appears to have less than 50feetofdisplacement.Thenumber ofthesefaults increases toward the center of the bolson. Because the Love Formation has been eroded away, it can­not be determined whether or not the faults cut it also. A few normal faults with less than 50 feet of displacement and several reverse faults with about a foot of displacement do cut the Love Formation. Most of these are eitherroughly parallel orperpen­dicular to the axis of the bolson. The structure of the fill in the Washboard Hills described by Underwood (1963) resulted from faulting in the Love Formation, where arcuate normal faults are expressed as a series of arcuate ridges. These ridges are well developed in the WashboardHillsandtracesofthemare visiblenorth­east, north, and northwest of the hills. On the av­erage the faults dip south-southwest, and the strata they cut dip north-northeast. At several localities the faults cut what seem to be the remnants of a pediment gravel. Displacement appears to be on the order of tens of feet. Underwood suggestedthat the structure resulted from creep toward the south-southeast. SYSTEMATIC PALEONTOLOGY This report gives systematic descriptions only of the mammals from the Red Light local fauna. It lists the Aguila local fauna and the lower verte­brates from the Red Light local fauna. John G. Lundberg of the University of Michigan identified the fish from both faunas. Pierce Brodkorb of the University of Florida is studying the birds from both faunas. Terminology ofhorse teeth is after Quinn (1955)and terminology of artiodactyl postcranial mater­ial is after Webb (1965). The Red Light local fauna may be divisible into two faunal units, one from the stratigraphicallylower localities and one from the stratigraphicallyhigher localities. The two are quite different but only the microvertebrates are well known from the lower localities and only the larger vertebrates are well known from the higher localities. A similar di­vision ispresent in the Hudspeth local fauna (Strain, 1966). Tables la and lb are faunal lists of the lower and higher localities of the Red Light local fauna and comparisons with the Hudspeth local fauna. If further study shows that these faunas are different, I propose that the name Red Light local fauna be retained for the fauna from the strati-graphically higher localities. Locality descriptions. —The Texas Memorial Mu­seum identifies fossil vertebrate localities by the firstfive-digit numberofeachspecimen,e.g.,40664. The second number of each specimen, e.g., -35, is a serial listing of the specimens from that locality.Therefore, the locality for each TMM specimen,40664-35, is determined without further descrip­tion. The systematic section of this paper does not list the locality for each specimen. Instead, the read­er is referred to figure 2 for the geographic positionof each locality, to figure 3 for the approximatestratigraphic position of each locality, and to the following short descriptions of each locality. The locality descriptions are in numerical order. TMM 40664, Love Formation, is the most fossil­iferous locality. It is a continuous exposure of a thick section, mostly interbedded fluvial silt, sand,and gravel. Much of the gravel and some of the sand is indurated. In the northeast part of TMM 40664 these fluvial sediments are interbedded with reddish silty clay. Some poorly developed caliche is near the base of the section. To the southwest,the gravel beds thin and finer sediment predomi­nates. Measured Section RL-1 is at this locality. TMM 40855, Love Formation, is a small, isolated outcrop of sand and caliche-gravel. Some fossil ma­ terial came from it. TMM 40856, Love Formation, resembles TMM 40664 except that much less section is exposed and the outcrop is a series of small hills. Some fossil 10 Fig. 6.—Cross-bedded gravel at locality TMM 40664, Love Formation. material came from it. TMM 40857, Love Formation, yielded most of the microvertebrate material. Sediments here are fluvial sand and silt with minor amounts of granule-to pebble-size material. TMM 40858, Love Formation, is a small, isolat­ed outcrop of sand and silt. Little material was collected here. TMM 40859, 40860, and 40861, Love Forma­tion, are small outcrops of interbedded silt, sand,and gravel resembling those at TMM 40664. Little material was collected at thes£ localities. TMM 40866, Love Formation, is the stratigraph­ically lowest locality within the Love Formation. Little material was collected from the fine sand and silt here. TMM 40891 and 40892, Love Formation, are in extensively faulted fluvial silt, sand, and conglom­erate similar to TMM 40856 and 40664. Outcrops are mostly covered with a lag gravel derived from pediment deposits but a few fossils were collected Fig. 4. (opposite page, top)—View of locality TMM 40664,Love Formation, looking north. Fluvial facies interfingeringwith reddish silty clay (dark, bluff-forming unit). Dark bed at upper right is younger gravel. Eagle Mountains in back. Fig. 5. (opposite page, lower)—Closeup of figure 4. R= red­dish silty clay, G= gravel. from good exposures. TMM 40962, Love Formation, is in fluvial silt and sand with a few thin beds of gravel. The fossil material from this locality came from the lower part of the exposed section, which also bears a few caliche nodules. TMM 40963, Love Formation, is in clay, silt,and fine sand. It is a small, low outcrop which yielded a small amount of fossil material, mostlymicrovertebrates. TMM 40964, Love Formation, is a large area of excellentexposuresbutverylittlebone.It ismostlysilt and sand with some clay low in the section and some gravel high in the section. TMM 40965, Bramblett Formation, is the same area as TMM 40964 but is assigned a different num­ber because it is in a different stratigraphic unit. The section is interbedded clay and silt. Microverte­brates were collected from a zone about one inch thick containing numerous plant and gastropodfragments. This zone is within a green clay close to the top of the Bramblett Formation. Fragmentaryremains of small fish are common, but other verte­brates are rare. fauna. Aguila local fauna.—The Aguila local fauna was collected from the upper part of the Bramblett Formation. It is named for Aguila Draw near which the collections were made. Class OSTEICHTHYS Gila sp. Class AYES Unidentified birds Class MAMMALIA Order RODENTIA Gen et sp. indet. fauna. Red Light local fauna.—The Red Light local fauna was collected from the Love Formation. It is named for the Red Light Bolson in which the collections were made. Class OSTEICHTHYS Gila nigrescens or pandora Pylodictis olivaris Small catfish Class AMPHIBIA Unidentified anurans Class REPTILIA Trionyx sp. Pseudemys sp. Geochelone sp. Gophems sp. Terrapene sp. Phrynosoma sp. Unidentified lizards and snakes Class AYES Odontophorinae, Gen. undet. Accipitridae, Gen. indet. Ciconia sp. Passeriformes, indet.; Class MAMMALIA Order EDENTATA Family MEGALONYCHIDAE Megalonyx sp. Figure 8, A Material—Partial right femur, TMM 40856-47. Description.—Both the distal and proximal ends of the femur and the proximal part of the third trochanter are missing. The surface of the remain­ingpartofthethirdtrochanterisflatbut roughened.It appears to have been quite long, more than 75 mm. The greatest width of the preserved part is 22 mm. The surface of the lesser trochanter is round­ed and rough. It is set off from the body of the femur by a groove on the posterior and proximalsides. Maximum width is about 26 mm. A large,roughened muscle attachment is present on the posterior side of the femur at the level of the lesser trochanter. At the level of the third trochanter, the width of the femur is 108 mm and the antero­posterior diameter is 42 mm. The shape and pro­ portions of this specimen are close to the femur from the Hudspeth local fauna that Strain (1966)referred to Megalonyx. TMM 40856-47 differs from the specimen of Megalonyx jeffersoni figured byLeidy (1855) in that the third trochanter in the latter is a sharp ridge. The specimen is closer to Megalonyx than to the other genera of North American ground sloths. Probable habitat.—Megalonyx probably was a browsing, forest-dwelling form, for it is associated with Late Pleistocene forested areas and forest faunas (Stock, 1925; Semken, 1966). The skeleton of Megalonyx indicates a slow, ponderous animal. Family MYLODONTIDAE Paramylodon sp. Figure 8, B, C Material.—Two caniniform (Ist anterior) teeth,TMM 40855-12 and -33; fragmentary right mandi­ble, TMM 40664-295; ungual phalange, TMM 40664-10. Description.— The first listed caniniform tooth,TMM 40855-12, is more strongly curved and small­er than corresponding teeth from Ingleside Pit,Texas, referred to Paramylodon harlani by Lun­delius (in manuscript). It is similar to this material in that the cross-section is a very rounded triangleand in having a series of fine, parallel, undulatingtransverse lines on the external surface of the tooth. Thegreatesttransverse diameteris 12.8mm,the greatest anteroposterior diameter is 14.0 mm,and the length along the outside ofthe curve is 90 mm. The wear facet is flat and slants only slightlyposteriorrelative tothelongaxisofthetooth.The strong curvature and small size indicate that it is probably an upper tooth from a young animal. The other caniniform tooth, TMM 40855-33, is larger in diameter and less strongly curved. It is probably an upper tooth from an individual largerthan the one represented by the other caniniform tooth. The basal portion is missing. The cross-sec­tion has the same shape as the previously described tooth but the lines are not present on the external surface. They may have been removed by stream abrasion. The greatest anteroposterior diameter is 19.2 mm and the greatest transverse diameter is 14.8 mm. The wear facet slants posteriorly and is slightly concave. The fragmentary right mandible is severelyweathered and broken. Only the area around the al­veolus for the fourth lower tooth is preserved. The fourth lower tooth is highly variable in Paramylo­donharlani(Stock, 1925).Theshapeofthealveolus in TMM 40664-295 is within the range of variation for Paramylodon harlani but it is smaller than most. Accurate measurements are impossible. The mandi­ble fragment is smaller than material of Paramylo­don harlani from the Ingleside fauna. The small process on the lingual side posterior to the fourth TABLE la TABLE lb COMPARISON OF THE LOWER PART OF THE RED LIGHT LOCAL FAUNA AND THE HUDSPETH LOCAL FAUNA OF THE FORT HANCOCK FORMATION REDLIGHT HUDSPETHLOCALFAUNA LOCAL FA UNA OF FT! HANCOCK FORMA TION (lowerpart) (Strain, 1966) Class OSTEICHTHYS Gila nigrescens orpandoraSmall catfish Class AMPHIBIA Unidentified anurans Class REPTILIA Testudinidaegen. sp. undet. Gopherus huecoensis Unidentifiedturtle Phrynosoma sp. Unidentifiedlizards andsnakes Class AVES Odontophorinae, Gen. undet. Passeriformes, indet. Class MAMMALIA Scalopus sp. Geomys(Nerterogeomys) Geomyspaenebursarius paenebursarius Prodipodomys sp. Perognathussp. Onychomys sp. Sigmodon hUdsp6thensis Sigmodon hudspethensis Citellus mcgheei CitellusfinlayensisNannippus phlegon Nannippus cf. minor Equus sp. Mastodont indet. Large camel Small artiodactyl Leporidae, 2 types indet. Leporidae gen. sp. undet. tooth is much less developed in TMM 40664-295 than in Pammylodon harlani. The ungual phalange is broken just posterior to the subungual foramina and the sheath is missing. The shape and cross-section are similar to those of Paramylodon harlani described by Stock (1925). Length from the subungual foramen to the distal tip is 58 mm and the greatest width is 21.4 mm. Probable habitat.— Stock (1925) suggested that this genus inhabited open country. He based this interpretation on skeletal characteristics, the pres­ ence of dermal ossicles, and associated faunas. The simple dentition suggests that it was a browser. Family GLYPTODONTIDAE Glyptotherium texanum Osborn, 1903 Figure 8, D, E; figure 9, A, B Material— Partial carapace, TMM 40962-1; nu­ merous isolated scutes; distal end ofright humerus, TMM 40664-245; right anterior ungual phalange, TMM 40664-109. Description.—Only about one-third of the cara­ pace is preserved. It includes the anterior portion and the portion of the left side that has the iliac COMPARISON OF THE UPPER PART OF THE RED LIGHT LOCAL FAUNA AND THE HUDSPETH LOCAL FAUNA OF THE CAMP RICE FORMATION RED LIGHT HUDSPETHLOCAL FA UNA LOCAL FAUNA OF CAMP RICEFORMA TION (Strain, 1966) Class OSTEICHTHYS PyIodictis olivaris Class REPTILIA Trionyx sp. Pseudemys sp. Geochelonesp. Testudinidaegen.sp.undet. Gopherussp. Terrapene sp. Class AYES Ciconia sp. Accipitridae, Gen. indet. Class MAMMALIA Megalonyx sp. Megalonyx sp. Paramylodon sp. Glyptotherium texanum Glyptotherium sp. Cratogeomys sp. Unidentifiedrodents Urocyon cf. progressus Canis cf. lepophagus Borophagus sp. Ischyrosmilus sp. Felis cf. rexroadensis Taxidea sp. Mastodont indet. Leporidae indet. Platygonus bicalcaratus Tanupolamasp. Tanupolama sp. Tanupolamacf. blancoensis Undescribed long-limbed camel Camelops sp. Gigantocamelus sp. Odocoileus? sp. Odocoileus sp. Capromeryx? sp. Equus(Plesippus) idahoensis Equusaff. scotti Equus (Plesippus) cf. Equus (Plesippus) simplicidens simplicidens Equus (Asinus) cumminsi Asinus cf. cumminsi Nannippusphlegon Nannippusphlegon Tapirus cf. copei and ischiac attachments. The only marginal area preserved is in the nuchal region. Part of it was destroyed prior to burial and part during exhuma­tion along an arroyo. Scutes and calichified por­tions of the carapace were recovered from the ar­royo. The specimen was upright when found with the long axis of the carapace about perpendicularto the current direction which prevailed duringdeposition of the Red Light Formation. Sediments within the carapace dipped upstream. The right side of the carapace had folded under the specimen and is so extensively calichified that the scute patternis largely destroyed. On the external surface is a thick band of caliche, two to six inches wide, which almost completely encircles the preserved portion.Numerous cracks distort the specimen. At presentmost of the external surface is unprepared. The nonmarginal scutes are usually hexagonaland arranged in transverse rows across the carapace. Fig. 7.—Silt and clay of the Bramblett Formation at the type section The surface of the scutes is punctate. They are about 30 to 50 mm in diameter as measured across parallel sides of the scutes. The central figures of these scutes are round to faintly polygonal and the surfaces ofthese figures are flat or slightly depressed.The diameters of the central figures are 50 percent or more of the diameters of the scutes. There are seven to nine marginal figures on each scute, seven being rare. They are clearly separated from each other and from the central figure by grooves in each scute but are continuous between scutes. They are punctate and do not show radial sculpturing.There are usually two to four hair follicles perscute in the groove between the central and mar­ginal figures. The three nuchal scutes preserved are small. Their lengths are about 25 mm, widths are about 25 mm, and thicknesses are about 19 mm. They are five sided with one suture on each of the lateral and anterior edges and two sutures on the posterioredge of each scute. The anterior margin is almost straight. The central figure is flat and raised well above the very narrow marginal area. No marginalfigures are present and punctation is reduced. A distinct groove is anterior to the central figure. Be­tween the nuchal scutes and the main carapacescutes is a row of small, almost square scutes. The centers of these and the adjacent row of carapace scutes are deeply depressed. Lateral to the nuchal area, rows of small, square to irregular scutes are intercalcated between the nuchal and main cara­pace scutes. Several posterior marginal scutes were found in the arroyo which cut through the specimen. These are very thick, up to 30 mm, and large. The central figure occupies most of the scute except for a nar­row marginal area on the internal side. There are also several scutes from the main carapace rows next to the posterior marginals. These are similar except that the narrow marginal area extends com­pletely around the scute. The centers of the central figures are slightly depressed. The attachment for the left ilium is 20 cm long,about 7 cm in greatest width, and about 73 cm in a straight line from the nuchal edge of the carapace.It is oriented almost perpendicular to the midline of the carapace. A portion of the attachments for the right ilium and the neural ridge of the sacrum are present. The attachment for the left ischium is aligned almost parallel to the midline of the speci­men. It is about 3cm wide and 15 cm long but it may not be complete. The anterior end is about 94 cm in a straight line from the nuchal edge of the carapace. This specimen is assigned to Glyptotherium tex­anum Osborn because the scutes are arranged in Fig. B.—A. Megaionyx sp., partial right femur, TMM 40856-47, internal and anterior views (XV2). B.Paramylodon sp.,outlineofalveolusforfourthlowertooth,TMM40664-295(XI). C. Paramylodon sp., caniniform tooth, TMM 40855-12, side view and cross-section (XI), D.Glyptotheriumtexanum,posteriormarginalscutes,TMM40962-1,externalview(X V2). E.Glyptotheriumtexanum,maincarapacescutes,TMM40962-1,externalview (XV2). transverse rows across the dorsal area of the cara­pace, the small nuchal scutes have straight margin­al edges, the marginal figures of the scutes are not sculptured, and the diameter of the central figureis 50 percent or more of the diameter of the scute. Of these characters, the only one mentioned byOsborn (1903) is the size of the central figures. The presence of transverse rows of scutes is inferred from Osborn’s text and figures. The character of the scutes in the nuchal region is not very clear in Osborn’s illustration but they appear to be similar to those of TMM 40962-1. The posterior marginalscutes of the type also appear similar to those of TMM 40962-1 found in the arroyo. Glyptotherium arizonae has heavy, nobby nuchal scutes set off from the main part of the carapace.The posterior marginal scutes have large, externallydirected projections, the mid-dorsal scutes have rather small central figures, and all but the lateral scutes have depressed centers (Gidley, 1926; Gazin, 1942). Brachyostmcon has heavy, wide nuchal scutes with a convex external edge, there is no row of small scutes between the nuchal scutes and main carapace scutes, and the transverse rows of scutes do not appear to extend across the dorsal area of the carapace (Brown, 1912). Boreostracon has well defined radial sculpturing on the marginal figures of the scutes, the marginalfigures are often poorly defined, the central fig­ures are often less than 50 percent of the diameter of the scutes, and the transverse bands of scutes become random in the dorsal area of the carapace(Simpson, 1929; Holmes and Simpson, 1931). Glyptodon has large, heavy nuchal scutes, the transverse rows of scutes only occasionally extend across the dorsal portion of the carapace, the cen­tral figures of the scutes are usually depressed and often smaller than the marginal figures, and there are usually less than nine marginal figures (Bur­meister, 1870-1874; Castellanos, 1953; Melton, (1964). The isolated scutes are from all portions of the carapaceplussomefromthecaudalarmorand, pos­sibly, the cranial shield. Onlyabout6cmremainsofthe distalendofthe humerus. It is 109 mm wide; the articulation is 70 mm wide and 40 mm in greatest anteroposteriordiameter. No difference can be seen between it and the humerus of Glyptotherium arizonae figuredby Gidley (1926). The ungual phalange is broad and flattened. It is 80 mm long. The proximal end is 43 mm wide and 37 mm in anteroposterior diameter. — Probable habitat. The dentition of glyptodontsiscomplex andhypsodont asinmany grazing mam­mals. However, there is no enamel on the teeth and they appear to lack a mechanism to crop grass. If they had grazed, the cheek teeth would have had to grow extremely fast to make up for wear since there is no enamel to resist abrasion. For these reasons I consider that glyptodonts probably were browsing forms. The restricted head movement,short stature, and ponderous body would not al­low them to feed on high vegetation. It seems im­probable that they could live in hilly terrain. Order RODENTIA Family GEOMYIDAE Geomys(Nerterogeomys) paenebursarius (Strain) 1966 Figure 10, A, B, C Material—Partial left mandible with incisor,TMM 40857-3; partial left mandible with incisor and P 4 through M2, TMM 40857-4; partial left mandible with Mj and M2, TMM 40857-5; rightmandible fragment with P4, TMM 40857-7; dia­stemal region ofmandible, TMM 40963-8; associat­ed upper incisors and skull fragments, TMM 40857­9; upper left incisor, TMM 40857-10; damagedP4, TMM 40857-16; numerous fragmentary ma­terial. Description. —TMM 40857-3 lacks the articular portion of the ascending ramus, TMM 40857-4 lacks almost all of the ascending ramus, TMM 40857-5 lacks everything posterior to the alveolus for the M3, and TMM 40857-7 lacks everythingposterior to the P4. All mandible fragments have the mental foramen ventral to the anterior portionof the masseteric crest. This is the criterion for assigning the specimens to the subgenus Nertero­geomys Gazin (1942) as redefined and changed in rank from genus to subgenus by Hibbard (1967).Themassetericfossae,labialto M3,inTMM 40857-3 and -4 are shallower and slightly wider than in speci­mens of Recent Geomys bursarius of similar size. The P4’s have dentine tracts on the internal and external surfaces of both the anterior and posteriorlophs. The re-entrants on the P4’s are squared. The lower molars have enamel only on the posteriorsurfaces. Measurements of these specimens and of mandibles of other species of Nerterogeomys are given in table 2. The material from the Red Light local fauna dif­fers from Geomys (Nerterogeomys) smithi Hibbard (1967) in having a much greater diastemal lengthand a smaller occlusal length of P 4 to M2. The oc­clusal length of the paratype of G. (N.) smithi is actually smaller than that of the material from the Red Light local fauna, but it is from a very im­mature individual. The Red Light specimens are larger than G. (N.) minor (Gidley) (1922) in all measurements. Geomys (N.) paenebursarius (Strain)(1966) is very close to the Red Light specimensexcept that the length of the diastema in the for­mer is slightly smaller. The holotype and paratypespecimens of G. (N.) paenebursarius from which this measurement can be taken appear to be young­er individuals than those from the Red Light local TABLE 2 MEASUREMENTS IN MM OF MANDIBLES OF VARIOUS SPECIES OF GEOMYS (NERTEROGEOMYS) LENGTH LENGTH WIDTH, WIDTH WIDTH LENGTH GREATEST POSTERIOR INCISOR DIASTEMA WIDTH PjtoA?2 P4 Mi LOBE P4 DIASTEMA G. (N.) smithi Holotype, UMMP 25095 from Hibbard (1967) 5.9 2.1 2.4 6.0 3.9 Rexroad fauna G. (N.) smithi Paratype, UMMP 28258 from Hibbard (1967) 5.0 - Rexroad fauna G. {N.} minor Holotype, USNM 10498 from Hibbard (1967)Benson fauna 5.0 - 1.9 2.0 2.1 5.4 3.5 G. (N.) minor UMMP 29156 from Hibbard (1967)Rexroad fauna - - - 1.7 1.9 5.0 - G. (N.) paenebursarius Holotype, TMM 40240-10 5.0 2.7 2.0 2.4 2.6 7.3 3.7 Hudspethlocal fauna G. (N.) paenebursariusParatype, TMM 40240-14 5.0 2.5 2.0 2.3 7.2 3.8 - Hudspethlocal fauna G. (N.) paenebursarius Paratype, TMM 40240-19 5.2 2.4 Hudspeth local fauna G. (N.) paenebursariusTMM 40857-3 Red Light local fauna 2.3 9.6 4.3 G. (N.) paenebursariusTMM 40857-4 Red Light local fauna 5.1 2.5 2.0 2.1 2.4 8.0 4.2 fauna. No significant qualitative difference can be ing. The greatest anteroposterior diameter of the seen between these two groups of specimens. preserved portion is 4.7 mm and the greatest width The upper incisors are bisulcate as in Recent is 4.4 mm. Ithas a single groove on the anterior sur-Geomys. TMM 40857-10 is much smaller than the face as in Recent Cratogeomys. The enamel on the incisors from TMM 40857-9. external side of the tooth is a little wider than in The P 4 lacks enamel on the posterior surface as available specimens of Cratogeomys castenops. The in Recent Geomys. Both the internal and external specimen did not char or give off a burned bone sides of the anterior and posterior lobes appear to odor when held in the flame of a Bunsen burner. have dentine tracts but this may have resulted from TMM 40856-101 is only the ventral portion of a the damage to the tooth. The re-entrants are mandible, broken through the alveolus for the P4squared as in the P4’s. and about 7 mm posterior to the alveolus for the Probable habitat. —Recent Geomys prefer sandy M3. It is too fragmentary for definite identification soils at least four inches deep to burrow in. They but is assigned to Cratogeomys on the bases that eat a wide variety of vegetable matter and a few it is from a gopher about the same size as that insects (Davis, 1960). indicated by the incisor discussed above and is from the same stratigraphic interval as the incisor. Cratogeomys sp. Probable habitat. —Recent Cratogeomys caste-Figure 11, A nops feed on roots, stems, and bark, and preferdeep, rock-free soils. According to Davis (1960),Material—Upper left incisor, TMM 40664-290; this animal can live in soils containing considerable left mandible fragment, TMM 40856-101. clay. Davis also remarked that Geomys and Crato­Description.—Therootareaoftheincisorismiss-geomysinhabitmutuallyexclusiveenvironments with Geomys living in soil containing little clay. Family HETEROMYIDAE Prodipodomys sp. Figure 11, B Material-Left Mj, TMM 40857-13. Description. —The tooth is unworn except for slightabrasionofthe cusps,probablyresultingfrom the screening procedure. The greatest length of the tooth is 1.11 mm and the greatest width is 1.83 mm. There is no sign of dentine tracts along the sides of the tooth. Absence of a dentine tract is the primaryfeature which distinguishes Prodipodomys from Dipodomys (Hibbard, 1962). Cusp nomenclature is from Wood and Wilson (1936). The hypolophid is sharp, thin, and nearlystraight. The hypostylid, hypoconid, and entoconid are well defined, sharp cusps on the hypolophid.The hypoconid is much closer to the hypostylidthan to the entoconid. The metalophid is sharp,higher than the hypolophid, and convex posterior­ly. The three anterior cusps-metaconid, proto­conid, and protostylid--are on the metalophid and are well developed. Anterior to the metalophid is another sharp but lower loph connecting the meta­conid and protostylid. It is concave posteriorly.This loph and the metalophid enclose a depressionwhich is pointed at the metaconid and rounded at the protostylid. The hypoconid is connected to the protoconid by a low ridge and the hypostylid is connected to the protostylid by a higher ridge.These four cusps enclose a small depression, part of the median valley. This specimen is larger than Prodipodomys rex­roadensis Hibbard (1954) and about the size of a Prodipodomys sp. from the Sanders local fauna (Hibbard, 1956). The configuration of the occlusal surface is similar to that of a molar in the Sanders material. Specific assignment cannot be made until additional material is collected. Probable habitat.— The Recent kangaroo rats,Dipodomys, are primarily desert and wasteland in­habitants. They eat mostly seeds but sometimes also take green vegetation (Davis, 1960).Most spe­cies do not require water but are able to make use of metabolic water. Of all the rodents in North America, they are the best adapted to desert life. These data do not necessarily mean that Prodi­podomys lived in a desert. The association of moist-climate mammals and mollusks with Prodipodomysin the Sanders and Rexroad local faunas shows that this genus was not restricted to arid environments. Perognathus sp. Figure 11, C Material-Left P4, TMM 40857-14. Description.— The tooth is only slightly worn. Measurements are: greatest length, 0.94 mm and greatest width, 1.08 mm. It is intermediate in size between Perognathus merriami and P. hispidus and about the size ofP. penincillatus. Cusp morphologydoesnot differfromthatofRecent species ofPerog­nathus. The morphology of the P 4 is not diagnosticin distinguishing the species of this genus. Probable habitat.—Pocket mice feed almost ex­clusively on seeds and low woody growth (Davis, 1960). They are widespread in the western United States and Mexico but usually prefer arid to semi­arid lands (Hall and Kelson, 1959). Family CRICETIDAE Onychomys sp. Figure 11, D Material—Left Mj, TMM 40857-12. — Description. The specimen is moderately worn;the greatest length is 1.83 mm and the greatestwidth is 1.15 mm. It is about the size of the Recent Onychomys leucogastor but differs from it in beinglower crowned. Also, the anteroexternal re-entrant between the anteroconid and the protoconid is narrow and has a straight anterior border oriented at right angles to the long axis of the tooth. This re-entrant in the O. leucogastor is wider and has a rounded anterior border directed posterointernally.TMM 40857-12 differs from O. fossilis Hibbard (1941a), from the Borchers fauna, in these same respects and in having the re-entrant between the entoconid and metaconid almost exactly oppositethe protoconid rather than anterior to it as in O. fossilis. TMM 40857-12 is almost identical to a specimen of O. gidleyi Hibbard (1941b) from the Rexroad fauna but since the M 3 is the most diag­nostic tooth in determining species of Onychomysit cannot be referred with any confidence. , Probable habitat.—Onychomys feeds primarily on insects but will occasionally eat plant material (Davis, 1960) and other rodents (Hall and Kelson, 1959). Most species inhabit arid and semiarid re­ gions of the western United States and Mexico. Sigmodon hudspethensis Strain, 1966 Figure 11, E, F Material —Partial left mandible with partial in­cisor and Mi through M3, TMM 40857-10; unworn left Mi, TMM 40857-11. Description. The mandible is broken through — the diastema and several millimeters posterior to the M3, The portion ventral to the incisor is also missing. The isolated left M\ shows slight abrasion of the highest points. Both specimens have typicalSigmodon cusp patterns, although the folds are not as tight as in Recent species. They fit completelyStrain’s (1966) description of Sigmodon hudspeth­ensis and comparison of the specimens shows that they are most probably conspecific. In the type description of Sigmodon hudspeth­ensis, one of the criteria used to separate it from S. intermedins is that the anteroexternal re-entrant of the Mj in the former is deeper and not closed to the outside. The M\ of the type of S. inter­medins (Hibbard, 1938) does have a shallower an­teroexternal re-entrant which is closed to the out­side but this specimen is heavily worn. Topotypematerial of S. intermedins is very similar to S. hud­spethensis in these two characteristics if not heavilyworn. There is also a low enamel ridge across the opening of this re-entrant in all specimens of S. hudspethensis examined. This would close the re­entrant at a very extreme state of wear. The re­entrant in S. intermedins closes at an earlier state of wear. \ There are several other differences between Sig­modon hudspethensis and S. intermedins. The enamel folds in the former are tighter and their sides are more nearly parallel. This tighter foldingis an advanced characteristic. No mandibles are in the type series described by Strain (1966). The mandible from the Red Light local fauna has the alveolar region elevated considerably above the mas­sateric valley whereas the mandibles of S. inter­medins do not. More material is necessary to de­termine if this characteristic is consistent in S. hudspethensis. Comparative measurements of the lower dentitions of S. hudspethensis and S. inter­medins are given in table 3. The sample is small but the measurements of the length of Mj, width of M3, and length of M 3 are slightly larger in S. hudspethensis. The most diagnostic measurement is the width of the incisor which is much greater in S. hudspethensis. Probable habitat.—Recent species of Sigmodon are grazers and prefer warm climates. Order CARNIVORA Family CANIDAE Urocyon cf. progressus Stevens, 1965 Figure 12,A Material—Part of a left mandible with alveoli of P 4 through M3,TMM 40855-15. Description. —The specimen is broken anterior to the P 4 and posterior to the subangular notch. It is more massive than Urocyon cinereoargenteus and the subangular notch is more posterior. The depth of the mandible below the molars increases more markedly in a posterior direction than in the modem species. No other qualitative differences be­tween the fossil and Recent specimens could be de­termined. Table 4 gives the measurements of the fossil specimen and comparative measurements for a small sample of Recent U. cinereoargenteus from Texas and an unusually large Recent specimen from Indiana. The alveolar lengths of-each tooth in TMM 40855-15, except the M3, are within the range of Recent specimens. The fossil is larger in all other measurements except that the alveolar length from the P 4 to the M 3 is equaled by the large specimenfrom Indiana. The type material of Urocyon progressus Stevens,from the upper Pliocene Rexroad fauna of Meade County, Kansas, does not include a mandible (Stevens, 1965).Thismaterial-aleftparietal, aleft Ml, and an incomplete left tibia--is also from a grayfox larger than U. cinereoargenteus. This is the basis for tentatively referring the mandible to U. progres­sus. Urocyon atwaterensis Getz (1960) is a smaller animal. Probable habitat.—The Recent gray fox prefersbrushy or wooded area and rarely ventures into open country. He is omnivorous, eating mostlybirds and small animals, but he occasionally in­cludes fruits, nuts, insects, and crayfish in his diet (Davis, 1960). Urocyon cf. progressus probably had similar habits. Canis cf. lepophagus Johnston, 1938 Figure 12, B Material—Partial right mandible with alveoli or roots of C through P4, TMM 40664-3; fragment of TABLE 3 MEASUREMENTS IN MM OF LOWER DENTITIONS OF SIGMODON LENGTH Mi WIDTH Mi LENGTH m2 WIDTH m2 LENGTH m3 WIDTH m3 WIDTH LOWER INCISOR Sigmodon hudspethensisTMM 40240-1, Holotype Hudspethlocal fauna 2.21 1.44 1.58 1.54 Sigmodon hudspethensisTMM 40240-2, ParatypeHudspethlocal fauna 2.22 1.47 1.67 1.82 1.65 1.36 Sigmodon hudspethensisTMM 40857-10 Red Light local fauna 2.16 1.42 1.67 1.62 1.93 1.68 1.34 Sigmodon hudspethensisTMM 40857-11 RedLight local fauna 2.32 1.42 Sigmodon intermedius UMMP 41193, 3 spec.Rexroad Loc. No. 3 2.05 _ 1.32 1.16 1.74 1.58 " 1.63 1.32 1.68 1.58 1.42 1.32 1.05 1.00 1.00 left mandible with roots of Pj 2, TMM 40664-9;partial right radius, TMM 40664-11; right meta­carpal 11, TMM 40664-11. Description. —The partial right mandible is brok­en at the anterior alveolus of the M j and at the an­terior part of the canine. The roots of the C, ?2,and P 4 are present. Some traces of the alveoli for the incisors are visible. The fragment of left mandible is broken throughthe canine alveolus and the anterior alveolus for the P3. The ventral portion of the specimen is also missing. Urocyon cf. progressusTMM 40855-15 Red Light local fauna Urocyon cinereoargenteusObserved range of ten TMM recent specimensfrom Texas Urocyon cinereoargenteus Large specimen from Indiana. MSS 20 Canis cf. lepophagusTMM 40664-3 Red Light local fauna Canis cf. lepophagusTMM 40664-9 Red Light local fauna Canis lepophagus Observed range of 7 specimensCita Canyon fauna Canis latrans Observed range of 20 Recent speci­mens from Texas Measurements of these specimens, a sample of Canis lepophagus from the Cita Canyon fauna, and a Recent sample of C. latrans from Texas are givenin table 5. Measurements taken are only those which could be obtained for the material from the Red Light local fauna. The alveolar lengths of the individual premolars and of the premolar series are generally larger in C. lepophagus than in C. latrans as is the depth ofthe mandible. The measurements of the material from the Red Light local fauna are closer to C. lepophagus. In TMM 40664-3 and four out of the five C. lepophagus in which both TABLE 4 MEASUREMENTS IN MM FROM THE LEFT MANDIBLES OF SEVERAL SPECIES OF UROCYON AL VEOLAR ALVEOLAR ALVEOLAR AL VEOLAR ALVEOLAR DEPTH OF LENGTH LENGTH LENGTH LENGTH LENGTH MANDIBLE P4toMj P4 Mi m2 m3 UNDER M2 (lingual) 31.5 6.6 11.4 6.7 3.0 14.4 26.4 to 6.0 to 10.4 to 5.7 to 1.5 to 9.9 to 30.2 7.7 11.3 7.4 2.4 12.2 31.6 12.0 6.3 2.3 13.2 TABLE 5 MEASUREMENTS IN MM OF MANDIBLES OF CANIS. NUMBERS IN PARENTHESES SHOW SIZE OF SAMPLE WHERE SMALLER THAN TOTAL SAMPLE ALVEOLAR AL VEOLAR ALVEOLAR ALVEOLAR ALVEOLAR DEPTH OFLENGTH LENGTH LENGTH LENGTH LENGTH MANDIBLEP1 P2 P3 P4 P1toP4 BETWEEN Pj and 4.4 9.5 11.3 11.7 43.9 15.8 5.3 9.1 - - - - 4.1-5.3 8.7-10.8 10.5-12.1 11.0-13.4 41.2-45.0 16.0-17.6 (6) (6) (5) (6) 3.5-4.9 7.9-10.0 9.3-11.5 10.4-12.1 37.6-44.8 13.7-16.7 LENGTHPOST. ALV. OF M3 TO SUB­ANGULARNOTCH (parallel to base of mandible) 20.5 12.8 to 18.4 16.8 DEPTH OF DEPTH OF MANDIBLE MANDIBLE BETWEEN BETWEEN andPj P$ and P^ 16.0 18.3 - - 15.5-18-8 16.6-18.8 (6) (6) 13.2-17.3 14.3-18.2 Fig,9.—A. Glyptotheriumtexanum,partial carapace,TMM40962-1,internalview(aboutX0.12), B. Glyptotheriumtexanum,partialcarapace,TMM40962-1,internalviewofnuchalarea(aboutX0.17), = c caliche nodule = il iliac attachment = is ischiac attachment = n nuchal scutes Fig. 10.—A. Geomys (Nerterogeomys) paenebursarius, partial left mandible with incisor and P 4 through M2, TMM 40857-4,externaland occlusal views(X5), B. Geomys (Nertergeomys) paenebursarius, partial left mandible with incisor, TMM 40857-3, external view (X5). C. Geomys(Nertergeomys)paenebursarius, damagedP 4,TMM40857-16,occlusalview(X10). Fig. 11.—A. Cratogeomys sp., upper left incisor, TMM 40664-20, external and anterior views (X5). B. Prodipodomys sp., left Mj, TMM 40857-13, occlusal and internal views (X10). C. Perognathus sp., left TMM 40857-13, occlusal view (XI0). D. Onychomys sp., left Mj, TMM 40857-12, occlusal and external views (X10). E. Sigmodon hudspethensis, left Mj, TMM 40857-11, occlusal and internal views (X10). F. Sigmodon hudspethensis, partial left mandible with partial incisor and Mj through M3,TMM 40857-10, external and occlusal views (XI0). measurements could be taken, the depth of the mandible between the Pj and ?2 was greater than the depth between the ?2 and P3. In 80 percent of the C. latrans in which both measurements could be taken, the reverse was true. In the original description of Canis lepophagus,Johnston (1938) characterized the mandible as hav­ing relatively long teeth, in comparison to C. latrans, with the spaces between them correspond­ingly reduced. This is true in the samples measured but there is some overlap. Both the proximal and distal ends of the rightradius arepresentbutenoughofthe shaftismissing so that the length cannot be determined. There is no major quantitative or qualitative difference be­tween it and the same element from a Recent coy­ote. The right metacarpal II has the same configura­tion as that of a Recent coyote but it is slightlysmaller. The length is 51.0 mm. Probable habitat.—The coyote ranges from arid desertstohumidmoderately timbered areas.It isan opportunistic feeder, eating anything it can catch or find. Borophagus sp. Figure 12, C Material. —Proximal portion of left radius, TMM 40664-221. Description.—About two-thirdsofthetotallengthof the radius is preserved. The greatest anteropos­teriordiameteroftheproximalarticulation isabout 19mmandthe greatesttransverse diameterisabout 26 mm. The specimen is larger than the material of Osteoborus cyonoides figured by Matthew and Stirton (1930). It is very similar in shape and size to undescribed material of Borophagus sp. from the Cita Canyon fauna of Texas. Probable habitat.—Borophagus appears to have been a scavenger; its dentition is adapted more to crushing bone than to shearing flesh. It is usuallyconsidered an ecological equivalent of the hyaena. Family MACHAIRODONTIDAE Ischyrosmilus sp. Figure 12, D, E Material—Mostofan edentulousrightmaxillarywithafragmentofapalatine,TMM 40664-1;almost complete left fibula, TMM 40856-22. Description.— Much of the maxillary posterior to the infraorbital canal is missing and the occlusal area is severely damaged. The alveolus for the ca­nine is present but damaged. Only traces of several alveoli posterior to the canine are present. Mawby(1965) described a partial maxillary and premaxil­lary of Ischyrosmilus johnstoni from the Cita Can­yonfauna ofRandall County, Texas. TMM 40664-1 is similar to his specimen on two points. The suture between the maxillary and premaxillary on the pa­latal surface extends to near the anterior edge of the canine alveolus rather than near the middle as in Smilodon. The palatal part of the maxillary is smoother than in Smilodon. There is one point of difference; Mawby believed that the infraorbital foramen in Ischyrosmilus is smaller than in Smilo­don. However, the roof of the infraorbital canal is missing in his specimen, UCMP 66485. TMM 40664-1 has a large, oval infraorbital foramen with the long axis of the opening slanting posterodorsal­ly. The opening in Smilodon is a rounded trianglewith the base ventral and the long axis almost verti­cal. If enough of the roof to the opening in TMM 40664-1 werebrokenaway,itwouldappearsmaller because of the difference in shape. The long axis of the opening to the infraorbital foramen in TMM 40664-1 is 30.0 mm and the greatest transverse axis is 16.3 mm. Estimated alveolar length of the canine in TMM 40664-1 is about 40 mm and esti­mated width is about 16 mm, compared to 36.5 mm and 16.5 mm, respectively, for the Cita Canyonspecimen. The specimen from the Red Light local fauna is slightly larger than the specimen from Cita Canyon but both are so fragmentary that corres­ponding measurements cannot be accurately made. The left fibula is only slightly damaged. Measure­ments of it and the fibulae of other large Pleisto­cene felids are given in table 6. This fibula is slightly longer than that ofDinobastis serus and of very similar proportions. The fibulae of Smilodon californicus are shorter and stouter. The fibula of Felis atrox is both longer and stouter. The proximal end of the fibula from the Red Light local fauna has an anterior tuberosity but no posterior tuberosity. In the place of the posteriortuberosity there is a sinuous ridge with a concave lateral surface. The proximal articulation for the tibia is roughly circular and almost flat. It faces dorsally and slightly internally. The ridge for attach­ment ofthe interosseus membrane is sharp and well defined. The distalarticulationfor thetibiaislunate, concave in the proximal portion and convex in the distalportion.Itfaces internodorsally.Thearticular surface for the astragalus faces internally. The proxi­mal portion is slightly concave and the distal, slight­ly convex. The characteristics of this fibula do not compare favorably with any of the North American Pleistocene felids for which fibulae are known. The fibula of Ischyrosmilus is unknown, but Mawby(1965) described a right tibia which he referred to Ischyrosmilus. It is long and slender. The articula­tions for the fibula are complementary to the ar­ticulations for the tibia of TMM 40856-22 and they are only about 5 mm farther apart. On the bases of proportions and the shape and position of the ar­ticular facets, TMM 40856-22 is best referred to Ischyrosmilus. Probable habitat.—The accumulations of probo­scidian remains associated with machairodonts (Evans, 1961) and the specialization of the denti­tion (Matthew, 1910) indicate that they fed pri­marily on proboscidians. The long-legged, slender build of Ischyrosmilus (Mawby, 1965) is probably an adaptation for a more cursorial habit than in other machairodonts. Family FELIDAE Felis cf. rexroadensis Stephens, 1959 Figure 12,F, G Material.—Partial left mandible with roots ofP3and P4, TMM 40664-4; partial right mandible with rootsofP 3 andanteriorrootofP4,TMM40664-5;partial left mandible with alveoli ofP 3 and anterior alveolus ofP4, TMM 40856-36; partial right squa­mosal with partial tympanic bulla, TMM 40664-6;severely damaged proximal end of humerus, TMM 40664-8;right metacarpal 111,TMM 40664-7; water-worn left astragalus, TMM 40664-18. Description.—TMM 40664-4 is broken throughthe canine alveolus and the anterior alveolus of the Mj. TMM 40664-5 and -36 are broken through the canine alveolus and the posterior alveolus of the P4.The ventral portion of both specimens is missing.TMM 40664-36 is waterworn. Measurements of these specimens are given in table 7. The mandible of Felis rexroadensis Stephens (1959) is not known. The mandibles from the Red Light local fauna were compared with the type maxillaries of F. rexroad­ensis in the University of Michigan Museum of Pa­leontology, The apparent occlusion was very goodand the size matched. The mandibles are larger than Lynx rufus or L. canadensis. They are slightly largerand have a shorter diastema than F. longignathusShotwell (1956). They are much smaller than F. lacustris Gazin (1933). The partial squamosal and tympanic bulla is the onlyspecimen which has acounterpartinthe known material of Felis rexroadensis. Unfortunately, the squamosalofUMMP 34195(holotype) isso crushed and incomplete that comparison is difficult. The two squamosals are of about the same size and shape. The only area well preserved in both is the notch or valley between the glenoid fossa and the tympanic bulla. This notch is broadly U-shaped in both specimens. In all specimens of Lynx rufus and L. canadensis examined, this notch is V-shapedand narrower. The proximal end of a humerus is larger than that of Lynx rufus but it is too severely damagedto be accurately measured. The right metacarpal 111 does not differ quali­tatively from that of Lynx rufus but it is larger.The total length is 31.3 mm. The left astragalus also does not differ from that of Lynx rufus except in being larger. The greatestlength is 30 mm and the greatest width is 19mm. Probable habitat. —The modern bobcat is a largely nocturnal “stalk and pounce” predator that occu­ pies awiderange ofhabitats exceptopen grasslands.It feeds primarily on birds, small mammals, and carrion (Davis, 1960) but is occasionally able to kill weak or young deer. Felis cf. rexroadensis prob­ably hadsimilarhabitsexceptthatitmay havebeen large enough to prey more efficiently on largeranimals. Family MUSTELIDAE Taxidea sp. Figure 13, A Material. —Partial left mandible with alveoli for M 2 and posterior alveolus for Mj, TMM 40855-14. Description. —The specimen is broken just an­terior to the posterior alveolus for the Mj and a little posterior to the alveoli for the M2. It is largerthan but morphologically identical to Recent speci­mens of Taxidea taxus from Texas. Itmay represent a large, fossil species or a large individual of the Recent species. Probable habitat. —Taxidea taxus is a burrowingmustelid. It feeds primarily on small burrowingmammals but will eat almost any vertebrate flesh including carrion (Hall and Kelson, 1959). It pre­fers areas of grass or low brush. Order PROBOSCIDEA Mastodont, gen. et sp. indet. Figure 15, A Material—Damaged right calcaneum, TMM 40664-27; unworn cone from cheek tooth, TMM 40664-26; two partial vertebrae, TMM 40664-28 and -29; numerous enamel and ivory fragments. Description.—The calcaneum appearsto be from a sub-adult animal since the epiphysis at the proxi­mal end is not completely fused. The greatestlength of the specimen is 207 mm and the approxi­mate width across the articular surface for the as­tragalus is 160mm. No assignment to taxon can be made becauseofthelackofidentifiedcomparativematerial. The cone from a cheek tooth is 28 mm high.Thereareamajoraccessoryconidand severalminor ones as buttresses to the cone. It may be from a Stegomastodon, but it is too incomplete for defi­nite assignment. The partial vertebrae are too incomplete to be measured and have no important qualitativecharacteristics. Probable habitat.— Mastodonts were probablybrowsing animals. Order LAGOMORPHA Family LEPORIDAE Two types, gen. et sp. indet Figure 15, B, C, D Material—Upper incisor, typeA,TMM 40857-72; TABLE 6 MEASUREMENTS IN MM OF THE FIBULAE OF ISCHYROSMILUS AND OTHER LARGE PLEISTOCENE FELIDS TMM 40856-22 DINOBASTIS SMILODON FELIS ATROX ISCHYROSMILUS RED LIGHTL.F. SERUS (MEADE, 1961)FRIESENHAHN CAVE CALIFORNICUS (MERRIAM & STOCK, 1932)RANCHO LA (MERRIAM & STOCK, 1932)RANCHO LA BREA BREA Greatest length 317 287 212.7-284.7 369 Greatest ant.­ post. diameter proximal end 24.6 25-26 32.5-45.9 35.4 Greatest ant.­ post. diameter distal end 28.9 29 17.2-26.7 Transverse diameter distal end 15.9 14 14.4-34.0 Greatest diameter,middle of shaft 12.9 10.5 9.4-13.7 16.1 upper incisor fragments, type B, TMM 40857-73; partial right mandible with P4, TMM 40856-45; left femur, TMM 40855-16; partial sacrum, TMM 40962-2; numerous other skeletal fragments. Description.—The two types of incisors are differentiated by the presence or absence of fold­ing of the enamel along the groove of the incisor. Type A, TMM 40857-72, has a deep fold in the enamel extending posterior from the external ex­pression of the groove. The enamel in type B,TMM 40857-73, is not folded and simply follows the external expression of the groove. It is not possible at this time to assign these incisors to genera. The partial mandible is from a small animal, but it might not have been an adult. Only the portionanterior to the M| remains. The P3, diagnostic in rabbits, is missing. The P 4 is 1.9 mm long and 2.6 mm wide. The femur is only slightly damaged, mostly from stream abrasion. It is 98 mm long and about 15 mm wide at the distal end. Only the first two vertebrae are preserved in the sacrum. The greatest width across the pleurapophy­ses is 17 mm. Probable habitat. —Rabbits are both grazers and browsers. They inhabit a wide range of environ­ments. TABLE 7 MEASUREMENTS IN MM OF MANDIBLES OF FELIS CF. REXROADENSIS FROM THE RED LIGHT LOCAL FAUNA LENGTH ALVEOLAR ALVEOLAR DEPTH DEPTH DIASTEMA LENGTH P3 LENGTH P4 MANDIBLE MANDIBLE UNDER P3 UNDERP4 TMM 40664-4 app. 9.4 8.9 11.7 16.8 17.5 TMM 40664-5 app. 9.0 9.7 - - - TMM 40856-36 - 8.7 - - - Order ARTIODACTYLA Family TAYASSUIDAE Platygonus bicalcaratus Cope, 1893 Figure 13, B, C Material-Left Ml, TMM 40664-17; right Cl,TMM 40856-23; partial right Cl, TMM 40664-58;left mandible fragment with roots of P2-3, TMM 40859-1; partial proximal end of right radius-ulna,TMM 40664-15; distal epiphysis of left radius-ulna, TMM 40856-32; left astragalus, TMM 40855-9. Description.—TheleftMl isworntothelevelof the cingulum anteriorly and almost so, posteriorly.The cusps are worn almost flat. The cingulum is continuous except where it merges with the lingualsides ofthe internal cusps. Greatest anteroposteriordiameter is 17.1 mm and the greatest transverse diameter is 15.3 mm. This tooth is exactly like the left Ml in TMM 31175-12 from the Blanco fauna except that the posterior portion of the cingulumin the latter has been destroyed by crowding with the M2. TMM 31175-12 was referred to Platygonusbicalcaratus by Meade (1945). Thecompleteupperrightcanine,TMM 40856-23,is 114 mm in length, 25.2 mm in greatest antero­posterior diameter, and 17.0 mm in greatest trans­ verse diameter. The anterior wear facet is about 40 mm long and there is a small posterior wear facet 8 mm long. Except for degree of wear, this tooth is identical to the right upper canine of TMM 31175-12. The partial upper right canine is so extensivelydamaged that no accurate measurements can be made, but it is slightly smaller. The left mandible fragment is broken at the an­terioralveolusoftheP 4 andabout 10mmanterior to the alveoli ofthe ?2-It compares favorably with several mandibles of Platygonus bicalcaratus from the Blanco fauna. The alveolar length of the ?2 is 8.6 mm.and oftheP3,9.7mm. The olecranon process is missing from the proxi­malendofthefusedrightradius-ulna. This specimenis slightly larger and the posterior side is more con­cave than in material from Laubach Cave, George­town, Texas, which has been referred to Platygonus compressus by Slaughter (1966). The proximal ar­ticulation of the radius is 33.3 mm wide and 16 mm deep. The distal epiphysis of the left radius-ulna re­sembles that of Platygonus compressus in shapeand size. It is 37 mm wide and 25 mm deep. The left astragalus is almost identical to a speci­men of Platygonus compressus, also from Laubach Cave. Length along the outer side is 39.2 mm and the width of the distal end is 24 mm. The postcranial material is referred to Platygonusbicalcaratus because of its resemblance to P. com­pressus and on the assumption that only one speciesof peccary is present in the Red Light local fauna. Discussion. —Hibbard and Riggs (1949) placed Platygonus texanus Gidley (1903) in synonymywith Platygonus bicalcaratus Cope (1893). The series of specimens that they described from the Keefe Canyon Rexroad fauna contained material which could be assigned to each species, but there were also specimens which were intermediate in character. The material in the TMM collections from the Blanco fauna was divided into the two speciesby Meade (1945). A re-examination of the Blanco material does not support the presence of two clearly differentiated species of Platygonus in the Blanco fauna. Probable habitat.—The Recent javalina is pri­marily herbivorous and inhabits brushy country.Platygonus bicalcaratus may have preferred a more open habitat, as shown by some of the fossil as­ semblages in which it has been found. Family CAMELIDAE Tanupolama sp. Figure 13, D, E Material—Partialleftscapula,TMM40855-7; par­tial left radius-ulna, TMM 40664-12; distal articula­tion of right radius-ulna, TMM 40855-11; rightunciform, TMM 40664-218; partial left tibia, TMM 40664-274; right calcaneum, TMM 40664-16. Description.— The partial scapula is extensivelydamaged; only about 10 cm of the distal portionremains. The acromion process, the medial cora­coid tuberosity, and both lateral sides of the glen­oid cavity are missing. The anteroposterior diameter of the glenoid cavity is 38.0 mm compared to 37.5 mm in a Recent specimen of Lama. The preservedportions of the dorsal and ventral margins of the scapula are almost straight and thick. In Lama,these margins are very concave and thin. The partial radius-ulna lacks the distal one-half and the olecranon process. It is slightly larger than the same element of a Recent Lama. The greatestwidth of the proximal articular surface of the ra­dius is 42.3 mm; in Lama it is 41.8 mm. The onlyqualitative difference between the fossil specimenand Lama is that the fossil specimen has a foramen just distal to the lateral tuberosity. The distal articulation of a right radius-ulna is very similar to that of Lama except that it is larger.The maximum width of the articulation is 52 mm compared to 43 mm for Lama and 63 mm for Tanupolama blancoensis (Hibbard and Riggs, 1949).This specimen appears to be too large to fit with the other material of Tanupolama sp. and too small to be assigned to Tanupolama blancoensis. It is included with the smaller Tanupolama specimensforlack offurtherevidenceregardingitstaxonomic position. The right unciform differs from that ofLama in having a longer posteroventral process and in beingslightly larger. About 5 cm is missing from the proximal end of Fig. 12.—A. Urocyon cf. progressus, partial left mandible with alveoli of P 4 through M3, TMM 40855-15,occlusal and external views (XI), B. Cams cf. lepophagus, partial right mandible with alveoli or roots of C through P4, TMM 40664-3,occlusal and external views(XI), C. Borophagus sp., proximal portion of left radius, TMM 40664-221, anterior and proximal views (XV2) D. Ischyrosmilus sp., partial right maxillary, TMM 40664-1, external and occlusal views (XV2). E. Ischyrosmilussp.,leftfibula,TMM40856-22,completeinternalviewandexternalviewsofdistal and proximal ends (XV2). F. Feliscf.rexroadensis,partialleftmandiblewithrootsofP3andP4,TMM40664-4,occlusaland external views (XI). G. Felis cf. rexroadensis, partial right squamosal with partial tympanic bulla, TMM 40664-6, posteriorviewshowing shapeofnotchbetweenglenoid fossaandtympanicbulla(X2). About 5 cm is missing from the proximal end of the tibia. The shaft is crushed anteroposteriorly to within 8 cm of the distal end. The distal articula­tion was extensively weathered before burial. The distal articulation has an anteroposterior diameter of25.3mmand agreatestwidthof37.7mm.The same measurements for a specimen of Lama are 28.5 mm and 41.3 mm, respectively. The measure­ments of the fossil specimen would have been slightly larger had it not been weathered. The esti­mated total length of the specimen is 370 mm, the same measurement inLama is 270 mm. The calcaneum is 92.7 mm long, the greatestwidth through the sustentaculum is 28.0 mm and the greatest anteroposterior diameter through the flbular surface is 39.1 mm. The same measurements foraspecimenofLama are84.6mm,24.7mm,and 37 mm, respectively. There are a few minor dif­ ferences between the articular surfaces of the fos­sil calcaneum and that of Lama. However, the most striking difference is that, in comparison to Lamathe body of the fossil calcaneum is tilted about, 20°posteriorrelative tothearticulararea. With the possible exception of TMM 40855-11,the material-described above is from a small Tanu­polama similar to the Recent Lama but with longerlimbs. Tanupolama mirifica Simpson (1929), T. longu­rio (Hay) (1921), T. blancoensis Meade (1945), and T. seymourensis Hibbard and Dalquest (1961) are much larger animals. The smaller, well described species of Tanupolama are T. macrocephala (Cope)(1893) and T. vera (Matthew) (1909) if one fol­lows Hibbard and Dalquest (1962) in synonymizing T. americanus (Wortman) (1898) and T. stevensi (Merriam and Stock) (1925) with T. macrocephala.Both T. macrocephala and T. vera are slightly largerthan Lama. The material from the Red Light local fauna may belong to either of these or to a new species. Probable habitat.—Skeletal material and denti­tions indicate that the smaller species of Tanupo­lama were probably browsers in relatively brushycountry. Tanupolama cf. blancoensis Meade, 1945 Figure 13,F Material— Three first phalanges, TMM 40856-14,40664-222, and 40664-226. Description.—The three first phalanges have lengthsof 117mm, 129mm,and 123mmrespect­ively. All are very slender. Hibbard and Riggs (1949)recorded lengths of 127 mm and 130 mm for an­terior first phalanges and 113 mm for a posteriorfirst phalange of a specimen of Tanupolama blanco­ensis from the Blanco fauna (AM 20085). First phalanges from the Rexroad fauna assigned to this species by Hibbard and Riggs (1949) were from 84 mm to 108 mm long. All of this material assigned to T. blancoensis is also very slender. The speci­ mens from the Red Light local fauna are similar to the material described by Hibbard and Riggs and tophalangesreferableto T. blancoensis intheTMM Blanco fauna collections. Probable habitat.—This long-limbed species was probably a browser with habitat requirements simi­lar to those of the Recent African giraffes. Undescribed Long-limbed Camel (=Leptotylopus percelsus Matthew, Meade 1945) Figure 14, A Material—Partial left radius-ulna, TMM 40664­ 25. Description. —The olecranon process and the dis­ tal portion of the specimen are missing. There is a prominent rugosity proximal and slightly anterior to the proximal interosseus foramen. This rugosityis elongated parallel to the long axis of the bone. Anterior to this rugosity there is a sharp groove.The bicipital rugosity is longer than wide and has a deep depression in the proximal portion. The lateral tuberosity is well developed. All of the radii-ulnae of Titanotylopus (=Gigan­tocamelus), Camelops, Tanupolama, and Camelus examined have no sign of a rugosity in the area of the proximal interosseus foramen and have a more rounded bicipital rugosity with little or no depres­sion. Lama has a small roughened area in a depres­sion proximal to the proximal interosseus foramen and a bicipital rugosity more similar to that of TMM 40664-25. However, the bicipital rugosity is not as oval and the depression is relatively shallower and smaller. The unusually long radius-ulna from the Blanco fauna, TMM 31179-28, referred to Leptotylopuspercelsus Matthewby Meade (1945) doeshave these distinctive characters and is of similar size. Un­fortunately, the total length of the specimen from the Red Light local fauna cannot be determined. Comparative measurements of these two specimens are givenintable 8. The two specimens are so much alike and so different from the other Pleistocene camels studied, that they must belong to at least the same genus. The same characters used to separate these long-limbed camels from the other Pleistocene camels are present in a specimen, TMM 31132-18, from Lapara Creek fauna (Early Clarendonian). This specimen is referred to Procamelus occidentalis byPatton (1969). There are also some other similari­ties between this specimen and the radius-ulna from the Blanco fauna. This may have a bearing on the ancestry ofthese long-limbed camels orindicate that these characters are primitive. — Discussion: The taxonomy of the specimens which Meade (1945) referred to Leptotylopus per­ celsus Matthew has had a history of confusion and nomenclatural error. TABLE 8 MEASUREMENTS IN MM ON RADII-ULNAE OF LONG-LIMBED CAMELS (= LEPTOTYLOPUS PERCELSUS) WIDTH, PROX. WIDTH AT ANT. OF DEPTH A T ANT. OF END OF PROX. INTER­ PROX. INTER­ RADIUS OSSEUS FORAMEN OSSEUS FORAMEN TMM 31179-28 “Leptotylopuspercelsus”Blanco fauna app.85 62 59 TMM 40664-25 Red Light l.f. 79 57 57 Matthew had referred, in an unpublished manu­script, to a long-legged camel from the Blanco Beds. He called it Leptotylopus and compared it to Alti­camelus. He neither described it nor gave it a spe­cific name. Meade (1945) was informed that a skeleton,AM 20085, of a long-legged camel from the Blanco fauna in the American Museum collections, was “ labeled Leptotylopus percelsus Type in the handwriting of Matthew. He then referred the material which he had to this unpublished name without examining the specimen to which this name had been applied. Hibbard and Riggs (1949) investigated AM 20085 and determinedthatitbelongedtothetaxon Tanu­polama blancoensis Meade 1945, which was de­scribed in the same publication that the name “Lep­ ” totylopus percelsus was used. Webb (1965), on the basis of Meade’s publishedmeasurements, considered this long limb material to be from an unusually long-legged Titanotylopus.There is no similarity, either quantitative or qualita­ “” tive, between Leptotylopus and Titanotylopus. This long-limbed camel now has no name since “Leptotylopus percelsus” isclearly anomen nudem. It would be very desirable to describe and name this unusual camel if only a dentition could be re­ferred to the limb material. Since this cannot be done and the possibility exists that a dentition be­ longing to this camel may already be described, I have decided not to add more possible confusion to the present situation by giving a name to this animal. Probable habitat.—The long, slender limbs of this camel show that it probably was a giraffe-like animal. It probably browsed on high brush or small trees as does the Recent giraffe. Camelops sp. Figure 13, G Material—Leftmandiblefragmentwith P4,TMM 40664-14; partial tibia, TMM 40856-97; left cal­caneum, TMM 40856-109; right astragalus, TMM 40664-216; cervical vertebra, TMM 40664-293. Description.—The left mandible fragment is bro­ken just posterior to the P 4 and 120 mm anterior to the P4. Both breaks occurred prior to burial. The anterior break runs posteriorly from the dor­sal edge of the mandible and appears to have fol­lowed the posterior edge of the symphysis. The distance from the anterior edge of the P 4 to the posterior part of the symphysis is estimated as 75 mm. The mandible is 63 mm deep under the an­terior part of the P4. The diastemal crest is sharpand convex lingually. The P 4 has a crown heightof 33 mm, an occlusal length of 33.1 mm, and a maximum occlusal width of 14.4 mm. The lengthand width at the base of the tooth are 30.3 mm and 15.4 mm respectively. The anterior portionof a small enamel lake is present in the posteriorpart of the tooth. The rest of this enamel lake is broken out. A wear facet from the Mj is present on the posterior surface of the tooth. The mor­phology of the P 4 does not differ significantlyfrom that of most Camelops and is less complexthan that of Titanotylopus (= Gigantocamelus).Toothmorphology, large size,and absenceofaP3 are the bases for assignment of this specimen to Camelops. Comparative measurements for this speci­men and the two largest species of Camelops, C. huerfanensis (Cragin) (1892) and C. hesternus (Leidy) (1873), are given in table 9. TMM 40664­ 14 is considerably larger than either. It may repre­sent an unusually large specimen of either or an un­described species. The partial tibia is too damaged to provide ac­curate measurements or comparisons, but it is from a large camel. The left calcaneum is slightly damaged. The greatest length is 182 mm and the maximum an­teroposterior diameter is 86 mm. Corresponding measurements for thirteen calcania of Camelopshesternus reported by Webb (1965) had observed ranges of 157 to 170 mm and 65 to 76.7 mm,respectively. There are several differences between this specimen and calcanea of Titanotylopus spatu­las from the Blanco fauna. The fibular surface is much narrower, the posterior edge of the body of the calcaneum is less expanded, the curved distal astragalar facet has a larger posterior portion, and the fossa proximal to this facet is shallower in TMM40856-109 than in Titanotylopus. The morphologyof TMM 40856-109 is very similar to that of cal­canea from the Ingleside fauna, Texas, referred to Camelops by Lundelius (in manuscript). The right astragalus is slightly waterworn. The length along the external side is 94 mm and the width of the distal end is 61 mm. Eighteen astragali of Camelops hesternus reported by Webb (1965)had observed ranges of 80.5 to 91.9 mm and 55.3 to 64.6 mm for corresponding measurements. The major difference between TMM 40664-216 and as­tragali of Titanotylopus from the Blanco fauna isthat, in the former, there is no separation of the sustentacular and parasustentacular facets; in thelatter, these two facets are completely separatedby a low, well defined trough. Camelops astragalifrom the Ingleside fauna do not have these facets separated. In Webb’s (1965) figures of a Camelopshesternus astragalus, there appears to be a partial separation ofthese facets. The cervical vertebra lacks the central portionofthe neuralarch,andotherpartsofthe specimen are damaged. It is probably a third or fourth cervi­cal vertebra because of the relatively long hypapo­physes and moderately developed costellar processes(Webb, 1965). The length of the centrum is 230 mm. The maximum length listed by Webb for a cen­trum of a corresponding vertebra from Camelopshesternus is 218 mm. Probable habitat.—Camelops appears to have been an open-country form. Webb (1965) believed that it was adapted to grazing but occasionallybrowsed. I believe that the known species were primarily browsers because of the thin, procum­bent incisors which are typical of browsing artio­dactyls. Other Camel Material Vertebrae, some limb material, and tooth frag­ments could not be reliably assigned to any of the taxaofcamels describedinthispaper.Someofthis material seems to indicate the presence of addition­al taxa of camels in the Red Light local fauna. Family CERVIDAE Odocoileus? sp. Figure 13, H Material—Antlertine,TMM 40856-48. Description.—The tine is slightly sinuous and moderately compressed. The length is 51.7 mm, maximum diameter at base is 19.7 mm, and mini­mum diameter at base is 9.2 mm. The specimen is most likely from a deer, but it may be from some other cervid. Probable habitat.—This specimen probably rep­resents another browsing herbivore. Family ANTILOCAPRIDAE Capromeryx? sp. Figure 13, I Material—Distal end of left humerus, TMM 40855-21. Description.—Theposteriorportionofthe articu­lar area is damaged. The shape of the remainder of the articular area does not differ from that of a Recent Antilocapra but the specimen is onlyabout two-thirds as large. The greatest width across the articulation is 24.6 mm. The tentative assignment to Capromeryx is based on the small size r»f the specimen. Probable habitat.—Antilocapra, the living rep­resentative of the family, is an inhabitant of the open plains and is the fastest running North Ameri­can artiodactyl. The food consists mainly of browse and forbs with minor amounts of grass(Davis, 1960). The habitat of the smaller, fossil antilocaprids may have been somewhat different. Order PERISSODACTYLA Family EQUIDAE Equus aff. scotti Gidley, 1900 Figures 15, E, F, G; 16, A, B Material—Partial left maxillary with alveolus’forpi, DP2-4 erupting Ml, and unerupted M2, sec­tioned to ?show adult dentition, TMM 40664-23;right P3, TMM 40664-20; associated partial rightMl, M2, and fragments of m3, TMM 40664-255;right TMM 40664-195; left M 3 in maxillaryfragment, TMM 40664-21; right P 3 or P4, TMM 40856-39; partial left radius-ulna, TMM 40856-52;partial left astragalus, TMM 40664-52; right meta­carpal 111, TMM 40856-51; left metacarpal 111,TMM 40664-292; proximal portion of right meta­carpal 111, TMM 40891-2; distal portion of metapo­dial 111,TMM40856-95;threefirstphalanges,TMM 40664-37,40891-1, and 40856-1; three second pha­langes, TMM 40856-4 and -108, and 40664-3. Description.—Little remains of the maxillary, TMM 40664-23, except for the bone immediately surrounding the teeth. The bone and the teeth have been fractured and distorted by calichification. The deciduous teeth are virtually complete and moder­ ately worn. A piece of the anteroexternal portion of the DP2 is missing. The fossettes of the decidu­ ous teeth are large and complicated with well de­ veloped plications except for the pli hypostyle on the DP2. The pli caballin is well developed on all three. The protocone is small and oval on the DP2 Fig. 13.—A. Taxidea sp., partial left mandible with alveoli for M 2 and posterior alveolus for Mj, TMM 40855-14, occlusal and external views (XI), B. Platygonus bicalcaratus, left Ml, TMM 40664-17, occlusal and external views (XI). C. Platygonus bicalcaratus, right upper canine, TMM 40856-23, internal view (XVi). D. Tanupolama sp., partial left tibia, TMM 40664-274, anterior and distal views (XV2). E. Tanupolama sp., right calcaneum, TMM 40664-16, internal view (XV2). F. Tanupolama cf. blancoensis, first phalange, TMM 40664-226, anterior and internal views (XV2). G. Camelopssp.,leftmandiblefragmentwithP4,TMM40664-14,occlusalandexternalviews (XV2). H. Odocoileus? sp., antler tine, TMM 40856-48, side view (XV2). I. Capromeryx? sp., distal end of left humerus, TMM 40855-21, anterior view (XI). TABLE 9 MEASUREMENTS IN MM OF MANDIBLES OF CAMELOPS MANDIBLE DEPTH UNDER P4 LENGTHP4 WIDTHP4 CROWN HEIGHT P4 Camelops sp. TMM 40664-14 63 33.1 (occl.) 14.4 (occl.) 33 RedLight local fauna 30.3 (base) 15.4 (base) C. huerfanensis (type) USNM 7819, from Hay (1913) 27 15 - 30 Huerfano C. huerfanensis USNM 5315, from Hay (1913) 23 14 25 - Minidoka C. hesternus, 5 spec, from Webb (1965) 53-65 17.4-27.5 10.2-14.7 - Rancho la Brea TABLE ID MEASUREMENTS IN MM OF DECIDUOUS DENTITIONS OF VARIOUS SPECIES OF EQUUS DP2 DP3 DP4 TOTAL LENGTH WIDTH LENGTH LENGTH WIDTH LENGTH LENGTH WIDTH LENGTH PROTOCONE PROTOCONE PROTOCONE TMM 40664-23 E. aff. scotti Red Light l.f. 113 43.5 23.5 8.2 34.0 27.0 10.0 35.0 27.8 11.8 UMMP V46404 E. scotti Gilliland l.f. 43.3 23.0 7.8 32.0 23.7 8.9 UMMP V46820 E. scotti Gilliland l.f. 30.8 23.1 9.4 33.3 21.9 10.5 E. (P.) francescana from Schultz (1936)Coso Mountains 22.4 33 21.6 34 22.6 E. (P.) shoshonensis from Gazin(1936)Hagerman l.f. 118 35 25 and DP3 and slightly longer on the DP4. All three the Gilliland specimens have a greater lingual in­protocones have slight lingual indentations. All dentation of the protocones and the DP3 and DP4 enamel ridges have a slight degree of minor crenula-are narrower than in the Red Light specimen.tion superimposed on the major pattern. The Ml Since the bases of the permanent teeth were visi­and are newly erupted with only slight wear on ble on the dorsal side of the specimen, it was sec-the Ml and none on the M2. Table 10 gives the tioned to obtain the adult molar pattern. The sec-measurements of these deciduous teeth and a few tion is cut rather high and the tooth pattern is that from the Gilliland local fauna of Texas which are ofan early wear stage, particularly in the premolars.assignable to Equus scotti. The major differences It was impractical to section the specimen deeperbetween the material from these two faunas is that because of the fracturing by caliche and the incom­ plete development of dentine and cement. As it is,the occlusal surface of the M 2 was considerablydamaged and all teeth were slightly damaged bysectioning. The teeth are large and have long,lingually indented protocones extending well an­terior to the comissure in all but the P2. The fos­settes are large and complicated and usually have well developed plications. The right p3, TMM 40664-20, is only slightly worn and was also sec­tioned, about 15 mm below the crown. It is es­sentially the same as the teeth described above. Measurements ofthe above teeth and otheradult teeth from the Red Light local fauna assigned to Equus aff. scotti are given in table 11 along with teeth assigned to E. scotti by Hibbard and Dal­quest(1966) fromthe Gillilandlocal fauna ofTexas. The enamel configuration of both is very similar. The only apparent qualitative difference is that the Gilliland specimens have, on the average, slightly more complicated fossettes. The only significantdifference in measurements is that the Gilliland specimens have longer protocones except in theP 2 and P3. The difference in length of the P 2 is due to TMM 40664-23 being sectioned above the point where the anterior region of the P 2 has de­veloped its greatest length. The limb material is assigned to this species sole­ly on the basis that it is too large to be assigned to the other two species of Equus recognized in the fauna. The proximal articulation of the radius and partof the shaft are missing from the left radius-ulna. Gazin (1936) lists widths of 60 to 66 mm for the distal articulation of the radius-ulna in five speci­mensofEquus(Plesippus) shoshonensisand64mm in one specimen of E. (P.) simplicidens. The distal articulation of TMM 40856-52 is not complete but it would have been about 80 mm wide. The great­est width of the sigmoid articular surface of the ulna is 47 mm. The anaconaeus process is very ru­gose,'more so that in any otherPleistocene or Re­cent Equus studied. Only the inner portion of the left astragalus is preserved. It is 69 mm long. Measurements givenby Gazin (1936) for the lengths of the inner sides of five astragali from Equus (Plesippus) shosho­nensis are from 58 to 63 mm and for two astragalifrom E. (P.) simplicidens, 60 and 60.5 mm. Measurements of the two complete and one partial metacarpals 111 are given in table 12 with comparative measurements for other species of Equus. The distal portion of a metapodial 111, TABLE 11 MEASUREMENTS IN MM OF PERMANENT DENTITIONS OF EQUUS SCOTTI AND E. AFF? SCOTTI TMM 40664-23 TMM 40664-20 TMM 40664-195 TMM 40664-21 UMMP V46898 UMMP V42885 E. aff. scotti E. aff. scotti E. aff. scotti E. aff. scotti E. scotti E. scotti REDLIGHTL.F. REDLIGHTL.F. REDLIGHTL.F. REDLIGHTL.F. GILLILANDL.F. GILLILANDL.F. P 2 LengthWidth LengthProto­cone 38.8 26.4 11.0 42.1 27.2 10.6 43.0 28.4 12.8 P-* LengthWidth LengthProto­cone 33.2 29.5 16.2 34.0 32.8 13.5 32.8 29.8 13.8 31.2 30.4 17.7 P4 LengthWidth Length Proto­cone 31.9 28.8 15.7 31.8 30.2 16.1 31.0 29.5 18.9 M 1 LengthWidth Length Proto­cone 27.6 28.8 13.0 28.1 29.0 14.8 25.7 28.9 15.3 M2 LengthWidth Length Proto­cone 28.8 28.2 14.6 30.1 27.9 16.2 27.1 27.5 17.3 LengthWidth Length Proto­cone 31 25 14.2 30.0 23.3 13.4 30.3 23.6 16.4 13.5 24.4 18.3 Fig. 14.—A. Unnamed long-limbed camel, proximal portion ofleft radius-ulna, TMM 40664-25,anterior and internal views (X 1/3). B. Unnamed long-limbed camel (=Leptotylopus percelsus Matthew, Meade, 1945) from the Blanco fauna, proximal portion of left radius-ulna, TMM 31179-28,anterior and internal views (X 1/3). TMM 40856-95, has a width across the distal ar-natural groupings. Since three groups were found ticulation of 53.3 mm and a depth across the keel in both the teeth and the skeletal material, these ofthe distalarticulation of37.0mm. were correlated on the basis of size. Then, com-Measurements of the three first phalanges are parisons were made with described material. Be-given in table 14, and of the three second pha-cause there was no associated material, assignmentlanges, in table 13. of skeletal elements may be partially in error. Comparison of the foot elements from the Red Discussion.— This is the first report of the pres-Light local fauna with those of the Gilliland local ence of an Equus scotti-like horse in a Blancan fauna shows that the primary difference between deposit as far as I am aware. However, there is no them is that the metacarpals 111 from the former reason why this type of horse should not have lived are significantly longer. in North America prior to the Irvingtonian. Simi- Assignment of the Equus material to species was lar E. scotti-like teeth may also be present in other accomplished in the following manner. Teeth and Blancan faunas from the southwest (C. W. Hibbard,skeletal material were each divided into apparently personal communication, 1969); It is interesting that in the nearby Hudspeth local fauna (Strain, 1966) which correlates very closely with the Red Light local fauna, there is a horse of similar size but represented only by a pair of lower jaws. The cheek teeth of this specimen are generally plesip­pine in character but somewhat atypical and per­haps advanced, especially in that the external re­entrant valley in both Mjs and the right M 2 does not pass between the flexids. Strain assigned this specimen to E. (Plesippus) idahoensis. There are two possible explanations for this situa­tion. These two faunas of equivalent age and sepa­rated only by a minor mountain range may have had a different species of large horse in each, or both may have had a single species of large horse in which the upper cheek teeth were similar to E. scotti and the lowers similar to E. (P.) idaho­ensis. I favor the latter explanation. Only the dis­covery of associated upper and lower dentitions will clarify this point. Along this line, it should be mentioned that in some of the lower dentitions of E. scotti from the Gilliland local fauna, the external re-entrant valley doespenetrate between the flexids in a few molars and the metaconid-metastylid val­ley is more V-shaped than in more typical E. scotti material, thus showing some plesippine characteris­tics. The lower premolar, TMM 40856-39, is onlypartially preserved but resembles the premolars of the Hudspeth specimen. Probable habitat.—All of the large Pleistocene horses were grazers living in more or less open country. Equus (Plesippus) cf. simplicidens (Cope), 1893 Figure 16, C, D Material.—Partial right mandible withDP2.4 and alveoli for incisors, TMM 40856-46; partial rightmandible with DP2-4 and alveolus for Pj, TMM 40664-24; two left DP3s?, TMM 40664-190 and 40856-40; left radius-ulna, TMM 40664-268; rightcalcaneum, TMM 40664-248; two partial calcanea,TMM 40664-55 and -208; five first phalanges, TMM 40664-39, -41, and -42, 40856-2, and 40855-8; two second phalanges, TMM 40855-6 and -35; five un­gual phalanges, TMM 40855-10, 40856-15, 40664­30, -31, and -34. Description.—The identification of the medium sized horse in this fauna is primarily based on a lower cheek tooth found in an abandoned sheep­herder’s cabin in the outcrop area of the Love For­mation. This tooth can be definitely assigned to Equus (Plesippus) simplicidens, but since its originallocality is unknown, it is not catalogued, figured, or described. The deciduous material described be­low is considerably smaller than material from the Gilliland local fauna assigned to E. scotti by Hib­bard and Dalquest (1966). It is too large to belong to E. (asinus) cumminsi and is similar in size and morphology to deciduous teeth from faunas con­ taining E. (P.) simplicidens in southwestern Kansas. The partial right mandible, TMM 40856-46, has the alveoli for the incisors so damaged that it is difficultto identify the individual alveoli. The man­dibular foramen is present but the area where the Pj would be is missing. All of the mandible pos­terior to about the middle of the alveolus for an erupting M j is missing. The teeth are only slightly worn. Since wear is so slight, a description of the enamel pattern would be oflittle use and the read­er is referred to the figure of the specimen (fig. 16, C). Measurements of this dentition and other deciduous teeth are given in table 15. The other mandible with a deciduous dentition,TMM 40664-24, is broken at the base of the alveoli for the incisors and only fragments of the mandible are preserved posterior to the DP2. The mandibular foramenand asmall,shallowalveolusfortheP\ are preserved. The teeth are unworn, only the cement covering the highest cusps showing any wear. The two left DP3S? are moderately worn. The parastylid is broken off of TMM 40664-190. Both teeth are laterally compressed, very narrow in com­parison to length. The metastylid and metaconid are long and narrow with a wide, V-shaped groovebetween them. The pli caballinid is well developed.The parastylid and hypostylid are very poorly de­veloped but the configuration of the lingual side of the teeth indicates that both may develop with greater wear. These specimens are quite similar to deciduous material referred to E. (Asinus) pons byQuinn (1958). However, the validity of this taxon is questionable. The following skeletal material is referred on the basis ofrelative size. The left radius-ulna is damaged but most meas­ urements can be made or estimated fairly accurate­ ly. Table 16 gives the measurements for it com­ pared with several other species of Equus. The length of the complete calcaneum is 122 mm. Gazin (1936) lists lengths of 104to 119mm for five calcanea ofEquus (Plesippus) shoshonensis and 114 and 120 mm for two calcanea of E. (P.)simplicidens. The two partial calcanea are similar in size to the complete one. Measurements of the five first phalanges are given in table 14; of the two second phalanges, in table 13; of the five ungual phalanges, in table 17. Probable habitat.—This form is another open country grazer. Equus (Asinus) cumminsi (Cope), 1893 Figure 16, E, F, G Material—Right P2, TMM 40856-104; right low­er molariform tooth, TMM 40964-1; left lower molariform tooth, TMM 40664-2; left upper mo­lariform tooth, TMM 40664-227; left calcaneum,TMM 40856-9; two left astragali, TMM 40856-7 and 40891-13; three first phalanges, TMM 40856-3 Fig. IS.—A. Mastodont, gen. et sp. indet.; cone from cheek tooth, TMM 40664-26, occlusal view (XI). B. Leporidae, type A gen. et sp. indet.; cross-section of upper incisor, TMM 40857-72 (X10). C. Leporidae, type B gen. et sp. indet.; cross-section of upper incisor, TMM 40857-73 (X10). D. Leporidae, gen. et sp. indet.; partial right mandible with P4, TMM 40856-45, occlusal and ex­ternal views(X2). E.Equusaff.scotti;partialleftmaxillarywithDP2-4,Ml,and TMM40664-23,externaland occlusal views, occlusal view of adult teeth after sectioning (XV2). F.Equusaff.scotti;right P3?,TMM40664-20,external,anterior,internal, occlusal,andsectioned views(Xy2). G.Equusaff.scotti;right P4,TMM40856-39,occlusalview(XV2). H. Equus sp., aberrant P2, TMM 40664-22, occlusal and external views (XVi). Fig. 16.—A.Equusaff.scotti,associatedpartialrightMland TMM40664-255,occlusalandinternalviews(XVi). B.Equusaff.scotti,leftm 3inmaxillaryfragment,TMM40664-21,externalandocclusalviews(XV2). C.Equus(Plesippus) cf.simplicidens, rightmandiblewithDP2.4,TMM40664-24,occlusalandexternalviews(XH), D. Equus (Plesippus) cf. simplicidens, left DP3?, TMM 40856-40, occlusal and external views (XV2). E. Equus (Asinus) cumminsi,rightlowermolariformtooth,TMM40964-1,occlusalandexternalviews(XVi). F. Equus (Asinus) cumminsi, right P2, TMM 40856-104, occlusal and external views (XVi). G. Equus (Asinus) cumminsi, left lower molariform tooth, TMM 40664-2, occlusal and external views (XV2). H. Nannippus phlegon, second phalange, TMM 40855-22, anterior view (XI). I. Nannippus cf. minor, associated left and TMM 40866-1, external and occlusal views (XI). and-8,40664-38; ungual phalange, TMM 40664-32. Description.—The lower molariform teeth are re­ferred to the subgenus Asinus for the following reasons. They are smaller than in any described species of the subgenus Plesippus. The median val­ley does not extend between the metaconid and metastylid asit doesinthemolarsofPlesippus. The metastylid does not have the “pig-ear” shapecharacteristic of Plesippus and the internal valleybetween the metaconid and metastylid is U-shapedrather than V-shaped as in most specimens ofPle­sippus. The metastylid is much larger than the metaconid. Measurements of these teeth and of a dentition from the Hudspeth local fauna, referred to Equus(Asinus) cf. cumminsi by Strain (1966), are givenin table 18. The measurements of the material from the Red Light local fauna are very close to those of the specimen described by Strain. These measure­ments are much larger than those of the type of E. (A.) francisci as given by Hay (1915). It is my opin­ion that E. (A.) pons Quinn (1958) cannot be dis­tinguished from other species of Asinus on the ba­sis ofthe type material. The upper molariform tooth is smaller than anyreferred to Equus (Plesippus) but this may be due in part to the extreme state of wear. The lengthalong the ectoloph is 23.6 mm, the greatest width is 25.7 mm, and the crown height is about 17 mm. The cement is very thick. The fossettes are almost completely worn away. The protocone is 11.7 mm long and extends about 3 mm anterior to the rela­tively wide commissure. The lingual side of the protocone has a shallow groove just anterior to the midpoint. These characters of the protocone are very similar to those of Equus (Hemionus) fran­ cisci (Hay, 1915). The left astragalus, TMM 40856-7, is 49 mm long medially and 48 mm wide across the distal articulation. Measurements ofthe other left astraga­lus are 48 mm and 49 mm, respectively. The left calcaneum is 96 mm long. The measurements of the three first phalanges are given in table 14 and, of the ungual phalange,in table 17. Probable habitat.—This form is another open country grazer. Other Equus Material Numerous vertebrae, partial teeth, and partial limb material could not be reliably assigned to anyofthespeciesofEquus describedinthispaper.One tooth, an aberrant ?2, is figured as a matter of in­terest (fig. 15,H). Nannippus phlegon (Hay), 1899 Figure 16, H Material— Second phalange, TMM 40855-22; dis­tal and proximal ends of right humerus, TMM 40857-2. Description.—There is no significant difference between this specimen and a second phalange asso­ciated with teeth of Nannippus phlegon, UMMP 33352, from the Deer Park fauna (Hibbard, 1956).Measurements of both specimens are given in table 19. Both are larger than the material of Nannippuscf. minor described from the Yepomera fauna byLance (1950). The distal and proximal ends of the right humer­us fall within the size range ofNannippus phleton.Theproximal end istoo damagedfor measurement. The distal articulation is 42.2 mm wide and 20.9 mm deep. Probable habitat.—Nannippus phlegon was one of the small grazing horses. Nannippus cf. minor (Sellards), 1916 Figure 16, I Material—Associated left M2? and M3, TMM 40866-1. Description.—The enamel is rather thin on both teeth. On the M2?5 the posterior border of the pre­fossette and the anterior border of the postfossette are moderately complicated. The medial external rib projects strongly anteroexternally. The proto­cone is small and oval. The unworn M 3 has been sectioned 16 mm from the occlusal end. It ap­parently is aberrant in that the prefossette is openposteriorly. The prefossette is complicated and the postfossette is simple. The protocone is a laterallycompressed oval with pointed anterior and pos­terior ends. The measurements of these specimens and others referred to Nannippus minor or N. cf. minor are given in table 20. All of them are smaller and less hypsodont than Nannippus phlegon. The specimens from the Red Light local fauna fall within the range of variation of the material from the Yepomera (Rincon) fauna ofChihuahua, Mexico referred to Nannippus cf. minor by Lance (1950). Another (hitherto unreported) very similar specimen is UMMP 31350, one complete and several incompleteuppermolariform teethfrom the Caliche Horizon of the Missler Member of the Ballard For­mation, Meade County, Kansas (see Hibbard, 1956).TheholotypeofNannippusminor,fromthe Ala­chua Formation of Florida, is an apparently de­ciduous upper premolar and a permanent uppermolariform tooth (Sellards, 1916). The permanenttooth is slightly narrower, has a slightly smaller protocone, and has more complicated fossettes than the other small Nannippus teeth discussed. No othermaterialofN. minorhasbeendescribedfrom the Alachua Formation. The type material is not sufficient to definitely determine whether or not it is similar enough to the other small species of Nannippus discussed above so that they may be TABLE 12 MEASUREMENTS IN MM OF METACARPALS 111 OF VARIOUS SPECIES OF EQUUS TOTAL WIDTH OF DEPTH OF WIDTH OF DEPTH ACROSS LENGTH PROXIMAL PROXIMAL DISTAL KEEL OF ARTICULATION ARTICULATION ARTICULATION DISTAL ARTICULATION TMM 40664-292 E. aff. Scotti Red Light l.f. 282 54 6 38 48 2 49 TMM 40856-51 E. aff. scotti Red Light l.f. 270 . 47.6 43.2 TMM 40891-2 E. aff. scotti Red Light l.f. 59.9 36.0 _ _ E. (P.) shoshonensis from Gazin(1936)6 specimens Hagerman fauna 240-254 46.5-53 30-33.5 43.3-47.8 32.3-36.9 E. (P.) simplicidensfrom Gazin (1936)Blanco fauna 252 51.2 34 46 34 UMMP V39375 E. scotti Gilliland l.f. 233 54.5 35.7 53.4 38.8 UMMP V39376 E. scotti Gilliland Lf. 255 52.8 33.5 49.4 41.4 considered as conspecific.Plaster casts of several additional specimens of N. minor from Florida were supplied by Dr. D. S. Webb of the Florida State Museum. They com­pare fairly well in size and pattern to the teeth from the Red Light local fauna. Discussion. —These specimens represent a lineageof small nannippines from the Hemphillian to the Aftonian. The type is apparently from the Hemp­hillian (D. S. Webb, oral communication, 1966) as is the material from the Yepomera fauna (Lance,1950). The material from the Red Light local fauna is assigned to the Nebraskan (this paper) and the material from the Missler Member is assigned to the Aftonian (Hibbard, 1956). Probable habitat.—This group of small nannip­pines appears to have primarily a southern distribu­tion, A number of specimens are known from the Yepomera fauna and from the Alachua Formation (D. S. Webb, oral communication, 1966).They are veryrare northofMexico and Florida.The smallest species of Pliohippus, P. hondurensis Olson and McGrew and P. castilli (Cope), also have a southern distribution. Olson and McGrew (1941) believed thatthesmallsizeofthese speciesofPliohippuswas a result of living in the more forested subtropicalenvironments which would be rather marginal en­ vironment for horses. I believe that the southern distribution, relativelybrachydont teeth with thin enamel, and size of the small Nannippus specimens are reflections of their adaptation to a browsing habitat in subtropicalforested or brushy areas. AGE AND CORRELATION There are two methods of subdividing the Pleistocene other than isotopic or geomagneticdating. The first is based on faunal assemblages and divides the Pleistocene into the Blancan, Irving­tonian, and Rancholabrean Provincial Ages (Savage, 1951; Hibbard, etal, 1965). The Blancan also in­cludes the latest Pliocene. The second method is to correlate with the several glacial and interglacialevents of the Pleistocene. It should be emphasizedthat these two age scales are independent, althoughoften correlatable. The former is based on faunal criteria; the latter, on climatological criteria. The Blancan Provincial Age is defined on the ba­sis of the association of Nannippus and Plesippuswithin a fauna (Hibbard, 1958). Glyptotherium,Equus (Plesippus) simplicidens, E. (Asinus) cum­minsi, Geomys (Nerterogeomys) paenebursarius,Sigmodon hudspethensis, Platygonus bicalcaratus, TABLE 13 MEASUREMENTS IN MM OF SECOND PHALANGES OFVARIOUS SPECIES OFEQUUS GREATEST WIDTH OF DEPTH OF WIDTH OF LENGTH PROXIMAL END PROXIMAL END DISTAL END TMM 40856-4 E. aff. scotti Red Light l.f. 49.4 52.0 32.7 44.6 TMM 40856-108 E. aff. scotti Red Light l.f. 52.2 51.0 33.3 48.2 TMM 40664-36 E. aff. scotti Red Light l.f. 48.0 49.8 31.6 43.0 TMM 40855-35 E. (P.) cf. simplicidensRed Light l.f. 45.7 41.2 28.7 37.4 TMM 40855-6 E. (P.) cf. simplicidensRed Light l.f. 41.9 39.6 26.7 37.6 E. (P.) shoshonensis from Gazin (1936)6 ant. phalangesHagerman fauna 43.6-49.9 47-51 30-32.1 41.5-46.3 E. (P.) shoshonensis from Gazin (1936)4 posterior phalangesHagerman fauna 47.2-49.5 45.8-49.5 31-31.4 38.8-42.5 E. (P.) simplicidensfrom Gazin (1936) ant. phalangeBlanco fauna 48 50.5 32.2 47.4 E. (P.) simplicidensfrom Gazin (1936)2 post, phalangesBlanco fauna 49-50.2 48-48.7 31.5-32 40-41 UMMP V46511 E. scotti Gilliland l.f. 53.6 56.2 36.8 app. 46 UMMP V46514 E. scotti Gilliland l.f. 50.4 53.7 34.1 48.0 UMMP V35669 E. scotti Gilliland l.f. 55.1 55.5 36.1 53.0 and the undescribed long-limbed camel are seem-from the Rexroad local fauna which is assigned to ingly restricted to the Blancan. Only Equus aff. the upper Pliocene part of the Blancan (Hibbard,scotti of the taxa described from the Red Light 1954). Using these two criteria, I assign the Red local fauna is unexpected in a Blancan fauna. Thus, Light local fauna to the Pleistocene portion of the the Red Light local fauna can be assigned to the Blancan, either Nebraskan or Aftonian. BlancanProvincialAge. If,asdiscussedinthesectionofstratigraphy,the The Blancan includes latest Pliocene and early Love Formation was formed during a pluvial stage,Pleistocene through the early Kansan (Hibbard, et the age of the fauna is Nebraskan. None of the al, 1965). Glyptotherium, Camelops, and Paramy-fauna is adapted to a cool climate but it is not lodon are unknown in the Pliocene portion of the necessary to postulate a cool climate during the Blancan(Hibbard, etal., 1965).Sigmodonhudspeth-Nebraskan this far south of the glacial advance ensis is more advanced than Sigmodon intermedins (Hibbard, 1960). After studying molluscan faunas, TABLE 14 MEASUREMENTS IN MM OF FIRST PHALANGES OF VARIOUS SPECIES OF EQUUS GREATEST WIDTH OF DEPTH OF WIDTH OF LENGTH PROX. END PROX. END DISTAL END TMM 40664-37 E. aff. scotti Red Light l.f. 87 56.8 37.1 43.9 TMM 40856-1 E. aff. scotti Red Light l.f. 88 57.9 40.9 43.2 TMM 40891-1 E. aff. scotti Red Light l.f. 90 app. 58 app. 42 TMM 40664-39 E. (P.) cf. simplicidensRed Light l.f. 85 53.7 35.6 40.2 TMM 40664-41 E. (P.) cf. simplicidensRed Light l.f. 80 app. 46 41.3 TMM 40664-42 E. (P.) cf. simplicidensRed Light l.f. 82 45.6 33.2 42.1 TMM 40856-2 E. (P.) cf. simplicidensRed Light l.f. app. 78 app. 50 41.0 TMM 40855-8 E. (P.) cf. simplicidensRed Light l.f. 78 48.0 37.2 39.0 TMM 40856-3 E. (A.) cumminsi Red Light l.f. 75 44.2 app. 32 34.0 TMM 40856-8 E. (A.) cumminsi Red Light l.f. app. 72 app. 37 app. 27 app. 33 TMM 40664-38 E. (A.) cumminsi Red Light l.f. app. 79 41.5 app. 31 E. (P.) shoshonensis from Gazin(1936)6 anterior phalanges Hagerman fauna 82.5-87.3 46-51 34.2-38.5 39.5-43 E. (P.) shoshonensis from Gazin(1936)5 posterior phalangesHagerman fauna 78.3-83 47-51.6 35.4-38.7 38-40.4 E. (P.J simplicidensfrom Gazin (1936)anterior phalangeBlanco fauna 86 53.2 37 43.5 E. (P.) simplicidensfrom Gazin (1936)1 posterior phalangesBlanco fauna 78 52.4-52.8 37.2-38 40.7-42 UMMP V46523 E. scotti Gilliland l.f. 93.0 65.6 44.0 45.8 (CONTINUED) TABLE 14 (CONTINUED) GREATEST LENGTH UMMP V46501 E. scotti Gilliland l.f. 89.7 UMMP V46518 E. scotti Gilliland l.f. 86.8 UMMP V46524 E. scotti Gilliland l.f. 91.2 UMMP V46500 E. scotti Gilliland l.f. 89.6 Taylor (1966) asserted that the effect of the Ne­braskan was less than that of subsequent glaciationsand suggested that the Nebraskan and Aftonian faunas of the Great Plains are actually older than the classic Nebraskan. Hibbard, et al, (1965) re­marked that the late Kansan Cudahy fauna is the oldest fauna that shows a major southward ecologi­cal shift of the mammals in the Great Plains area. Correlation of the Red Light local fauna with other Blancan faunas except the Hudspeth local fauna is meaningless. Few of the taxa of largermammals can be identifiedto species. Those which can be identified seem to range through much of the Blancan. The numerous publications of Hibbard on the fossil faunas of the Great Plains indicate that only the smaller mammals are useful in detailed correlations. The two taxa of rodents,identifiable to species, from the Red Light local fauna are not known from any other fauna except the Hudspethlocal fauna. Even when the faunas from the bolson area of the Southwest are better known, correlation with the well known Plio-Pleistocene sequence of the Great Plains will be difficult. The bolson area was probably relatively drier at all times and less af­fected by the Pleistocene glaciations. The moun­tains in this region probably provided different ecologic niches even if the climate resembled that of the Great Plains. Therefore, the makeup of the faunas, particularly the microvertebrates, would have been different in the two areas at any giventime. It may become possible to tie these areas together when the Blancan faunas from the Texas Panhandle are more completely studied. The micro­vertebratesfrom thesefaunas arevirtually unknown. Correlation of the Red Light local fauna with the Hudspeth local fauna (Strain, 1966) from the Hueco Bolson, west of the Red Light Bolson, is very good. A similar stratigraphic succession occurs in both bolsons and the faunas are similar. A com­parison of faunal lists (table 21) shows that of 17 WIDTH OF DEPTH OF WIDTH OF PROX. END PROX. END DISTAL END 61.6 40.5 49.6 61.0 39.3 50.2 60.2 38.4 57.2 60.6 38.9 45.6 taxa of vertebrates reported from the Hudspethlocal fauna, only Scalopus, Citellus mcgheei, C. finlayensis, Equus (Plesippus) idahoensis, Tapiruscf. copei, and Gigantocamelus (=Titanotylopus) are lacking in the Red Light local fauna. Giganto­camelus can probably be removed from this list because the metapodial fragments assigned to this genus by Strain may belong to the large Camelopsfrom the Red Light local fauna or some of the ma­terial assigned to the large Camelops from the Red Light local fauna may actually belong to Giganto­camelus. Equus (Plesippus) idahoensis might also be removed from the list as explained in the system­atic description of Equus aff. scotti. The absence of the other taxa from the Red Light local fauna may be ascribed to accidents ofcollection or preser­vation. Many taxa in the Red Light local fauna are absent in the Hudspeth local fauna, probablybecause more than ten times as many specimens are known from the Red Light localities. Both Sig­modon and Geomys evolved rapidly during the Pleistocene. Because these genera are representedby identical species in both faunas, I believe that the faunas are ofequivalent age. Correlation of these two faunas would indicate an age of Nebraskan for the Hudspeth local fauna rather than Aftonian as suggested by Strain (1966). In summary, the Red Light local fauna is Blan­can, probably Nebraskan, equivalent in age to the Hudspeth local fauna, and not faunally correlatable with other Blancan faunas. PALEOECOLOGY Most of the Bramblett Formation was depositedduring extremely arid conditions. This conclusion is inferred from the type of deposition and the ap­parent lack of fossils in all but the uppermost partof the formation compared to their abundance higher in the section. I interpret the Aguila local fauna, from the upper TABLE 15 MEASUREMENTS IN MM OF DECIDUOUS TEETH OF EQUUS (PLESIPPUS) CF. SIMPLICIDENS FROM THE RED LIGHT LOCAL FAUNA TMM40856-46 DP?. 4 dp2 dp3 dp4 TMM 40664-24* DP 2 . 4 dp2 dp3 dp4 TMM 40856-40 DP 3? TMM 40664-190 DP 3