MUSEUM OF NEW MEXICO

OFFICE OF ARCHAEOLOGICAL STUDIES

Prehistoric Burned Brush Structures and a Quarry Site along the Carlsbad Relief Route, Eddy County, New Mexico

by Dorothy A. Zamora

Contributions by Philip K. Alldritt David II. Hill Richard Holloway Susan M. Moga Palncla J. McBride Yvonne R. Oakcs Janw A. Quaranta Mollie S. Toll C. Dean Wilson

Submitted by Y vome R . Oakes Principal Investigator

ARCHAEOLOGY NOTES 203 .

SANTA FE 2000 NEW MEXICO ADMINISTRATIVE SUMMARY

Between April 14 and July25, 1997, at the request of were excavated and eachhad interior hearths andpits. the New Mexico State Highway and‘I’ranspol-tation The site seem to represent a winter occupation of Department (NMSHTI)), theOffice ol’Archaeologica1 sevcral families or groups of people at a seasonal Studies (OAS) cxcavated two prehistoricsites along the campsite. The ceramics and20 radiocarbon samples Carlsbad Relief Routein Eddy County, New Mexico. place the major occupation of the site within the ‘I’he sites consisted of a lithic procurement area (LA Fortnative periodat about AD. 770. However, site usc 29362) and an occupational area with several brush ranged between the Middle Archaicat ca. 2280 B.C. to structures (LA 29363). thc Late Globe phaseca. A.D. 980. The site is located Trojan Hill, I,A 29362, is a large quarry area on BLM land. where stone materials were beingtested. A variety of Funds providedby the NMSH‘I‘II and the Federal rnatcrials was present on the surface and in the Highway Administration were usedfor this project. subsurface deposits. Materials suchas chert, quartzite, silicified wood, chalcedony, rhyolite, andsiltstone were found. Most of thcsc malerials were waste pieces Submitted in fuliillment of Contract Agreement ranging from large to small in size, espccially the C03541I9X between the New Mexico State Highway quart.de. Evidence of attcmptsat tool manufacture and Transportation Department andthe Office of was notcd on some of the artifacts. The site is located Archaeological Studies, Museumof New Mexico. on BLM and B1.M-lcased land. Macho Dunes, LA 29363, is a large residential Bureau 01’ Land Managcment Permit21 -2920-96.r occupation containing several burned brushstructure Museum of Ncw Mexico ProjectNo. 4 1.641 1 remains with internal features and many outside New Mexico State Highway and Transportation associated thermal and storage pits. Five structures Ilepartrnent Project No. WTPP-76 1 S(8) CN 223I.

... 111 TABLE OF CONTENTS

ADMINISTRATIVESUMMAKY ...... ii

INTROI)UC?'I'IC)N ...... 1

ENVIRONMENTALSETTlNNG ...... 3 Physiography ...... 3 Flora ...... 3 Fauna ...... 5 Climate ...... 5

CULTUMLSIII"I'1NCr ...... 7 .. Palcomdlan ...... 7 Archaic ...... 7 Formativeperiod ...... 8

DATA KkCOVERY PLAN ...... 11 Theoretical Pcrspective ...... 11 Background Considerations ...... 11 IIunler-Gatherer Subsistence Systcms ...... 12 Data Recovery Qucstions ...... 14

SITE-SPt;.C:IFIC DATA REC,'OVEKY QUESTIONS ...... 19 I.A29362 ...... 19 LA29363 ...... 19

FIFl.DMETIIODS ...... 21

SITE SUMMARY-TROJAN HILL. LA 26362 ...... 23 Sctting ...... 23 Artifacts ...... 23 Site Interprutation ...... 23

SITE SUMMARY-MACHO DUNES. LA 2h3h3 ...... 29 Setling ...... 29 CulluralUnits ...... 25,)

CERAMIC ARTIFACT ASSFMHI.AGE by C . Dean Wilson ...... A3 Descriptive Attributes ...... 63 CeramicTypcs ...... 64 .. DrttingofSltes ...... 66 Rcginnal'l'rends ...... 70

PETROGRAPHIC ANALYSIS OF JORNADA BKOWN WARE CERAMICS FROM THE CAKI. SRAD RELIEF KC)IJ'I'E: LA 29363 by David V . Hill ...... 73 Analysis of the Ceramic Samples ...... 73 Discussion ...... 74

CHIPPt;,I) STONE ATTRIBU'I'ES by Philip R . Alldritt ...... 77 Methodology ...... 77 RawMaterials ...... 79

iv TABLE OF CON'I'b~N'TS DESCRIPTION OF CHIPPED STONE DATA-TROJAN HILL by Jatnes A. Quaranta ...... X1 Geologic Ilepositionof Materials ...... X1 Core Flake Materials-Knapping Properties...... 81 Material Type-Temporal Implications...... X2 Material Type-Functional Implications...... 82 Spatial Disaggregation of Sitc Components and Functional Implications ...... X3 FlakePorticms ...... X3 DorsalCortex ...... 83 Utilized Dcbris-Exotics ...... X6 Shatter-AngularDehris ...... X6 Platforms ...... X7 FlakcLength ...... XX Cores ...... X9 Technology-Behavioral Implications ...... 89 Utilization of the Lithicsand Functional Interpretations ...... 93 Tools ...... 94 Hammcrslones ...... 94 C:onclusions ...... 94

DESCRIPTION OF CHIPPED STONE IlAiA-TROJAN HILL by James A . Quaranta ...... 97 Geologic-Cultural Deposition ol‘ Materials ...... 97 Core Flake Material-Knapping Properties...... 97 Material Type-‘l‘etnporaland Mobility lrnplications ...... 97 Core Flakes-Behavioral Implications ...... 98 Jntrasite Assemblage Comparisons...... 100 CoreFlakes ...... 101 Shattcr-Angular Debris...... 109 Cores ...... 109 Tools ...... 109 Conclusions ...... 110

MAC110 DUNESGROUND STONE by Jatnes .A Quaranta ...... 113

MACIIO DUNES FAIJNA by Susan M . Moga ...... 119 Methodology ...... 119 Summary ...... 123

BOTANICAI. ANALYSES AT MACHO DUNESby Pamela McBrideand Mollie S . Toll ...... 125 Introductioll ...... 125 Methods ...... 125 Results ...... 126 Discussion ...... 136 Summary ...... 136

PAI.YNOLOGICAL ANAI.YSIS by Richard Holloway ...... 139 Methods and Materials...... 139 Discussion ...... 140

DATMCi TIIE CAKLSBAD SITES by Yvonnc Oakes ...... 141

SYN‘I’HESTS ...... 147 LA?9362, ‘frojanIIil1 ...... 147 LA29363, MachoDunes ...... 147 Conclusions ...... 150 ... .* ...... 153

APPENDIX 1 . ANALYSIS CODING FORMS...... 167 Ccramic AnalysisAttribute and Variable Codes ...... 167 Lithic Analysis Attributesand Variable Codes ...... 169 Ground Stone Analysis Attribulcs and Variable Codes ...... 173 Faunal Amlysis Attributes and VariableCodes ...... 176 APPFND1X 2 . LEGAL DESCKIP'I'IONS ...... 187 APPENDIX 3 . SITE LOCATION MAPS (removed from copies in general circulation) ...... 189

vi TABI. I. OF CONTENTS

TABLES

viii TABI,E OF CONTENTS

INTRODUCTION

The Carlsbad Relief Route is part of the Waste dates ranging between A.11. 1025 and A.D. 1195. Isolation Pilot Program (WIPP) road project (Fig. 1). These two hearths could have been used during the The New Mexico State Highway and Transportation period that people were occupying thesitc; however, Department (NMSHTD) proposesto improve County this is only spcculation. ‘l’he sitc straddles HLM and Koad 604 (NMSIITD No. WlPP-7615(8) in Eddy BLM-leased land. County. The excavation was requested by the Macho Dunesis located 011 the edge ofthe Alacran Environmental Section of the NMSIITD. A limited Hills sand dunes protectedby a high ridge covered with testing program was conducted by Office of dense vegetation.It is located on HI,M-owned land. It Archacological Studies (OAS), Museum of New covers 48,750 sq m within thc sand dunes. Several Mexico, in December 1996 and January 1997 (Zamora blowouts and dunes were extensively stripped and 1997). Funds provided by the NMSHTD and the consequently, six discrete areas of occupation were Federal Highway Administration were used forthis found. These areaswcrc identitied with theletters A pro.jcct. through G. Areas A and E contained thermal features Between April 14, 1997, and July 25, 1997, OAS that had been built into the sandstone bedrock. A personnel beganthe data recovery phase at‘I‘rojan Hill calibrated C-14 date of 380 B.C. to 5 B.C. was (LA 29362) and Macho Dunes(LA 29363). A total of produced for the hearth in AreaA and AD. X00 to 68 days per person were expendedin the field on the A.D. 1030 for the hearth in Area E. Both features project and the laboratory time wasa total of 136 days. exhibited oxidization and contained charcoal. Many ‘l’he principal investigator was YvonneR. Oakes, thc features werc found in AreasB through F, with Area B project supervisor wasDorothy A. Zamora, assisted by having tl~eeof the brush structure remains. AreaF was Philip R. Alldritt and Jamcs A. Quaranta. 1,aborers under a2.0 m high sand dune and contained12 pits and includcd Dixie Henry, Steve Kopccky,Jessie Murrcll, 1 roasting pit, suggesting an exterioractivity arca. The Bob Sparks, Chris Sparks, Jennifer Ware, and Alice other two structurcs were foundin Areas C and I), as Wydro. were several extratnuralpits. Area G did not contain ‘I’rojanHill is a large extensive quarrysite located any cultural remains. on a low rise overlooking the community of Ida Huerta. In each of the structures, interior hearths were It covcred an enormous area- 25,935 sq m. The found along with storagepits. One structure, Structure surface artifacts consisted of mostly primary core 2, had been heavily burned and contained sand flakes in large concentrations. Within these oxidation and charcoal staining. This was also the concentrations, there are many different types ol‘ decpcst of the structures, reaching 30 cm in depth. materials; however, the most common material type Each of the structures also produced a C-14 date was purple quartzite. No features and only four ranging from A.D. 720 to AD. 875, placing them ceramics were foundon the sitc withinthe right-of- within the Formative period. way, makingit difficult to determine theage or cultural An attempt was made to date the site by affiliation. One snlall broken, Archaic-like projectile archacomagnetic sampling, but it was unsuccesslirl point was recovcrcd. IIowever,it has been reworked because there wcrenot enough binding agents in thc suggesting that it was possibly curakd bylater peoplc. sand to hold the soil together. We had to rely on the Toward the cast, away from the sitc, two snlall carbon- 14 samples and ceramic cross-dating. hearths were found and excavated and producedC- 14 2 CARISBAD RELIEF ROUTE ENVIKONMENTAL SETTING

Physiography [lawley et al. (1974) bclicve theformation probably dates early lo middle Pleistocene, post-dating the The project area lies within thc pcninsular wedgeofthc deposition ofthe Ogallala Formation or Caprock (Katz C‘hihuahuan Desert biozone extending north Iiom the and Katz 1 W3). New Mexico-Texas border and is bounded on the east The Ciatuna Formation consists ol‘ reddish brown and west by the Idlano Estacado and the Guadalupe silt, sand, and clay (Hendrickson and Jones 1952%). Mountains (Medellin-Leal 1982). ‘I’he Pecos River is Jt contains sandstone, mudstone, conglomerates, the dominant physiographic feature in the area. It runs limestone and gypsum (Kelly 1971). Rlackdom, through a narrow channcl from the Sangc dcCristo Orchard Park, and IAewood are the three terraces Mountains and the gradientbecomes gentler and the present in the Pews Valley near Roswull (Fiedler and valley wider after it leaves L3e Haca County (Katzand Nye 1033) and have been traced south lo Carlsbad and Katz 1993). Today the Pecos River contains multiple north to I3 Sulnner (Collins 1985; Jelinek 1967; dams that control its flow makingit no1 as rough as in Hawley et al. 1976). ‘l‘he Rlackdom terrace is earlier days. Sheridan (1 975:3) states that the Spanish composed 01‘ sand, silt, and gravcls of the Pleistocene explorers mentioned it as “a rivcr ol‘violent extremes, (Hendrickson and Jones 1952:9). The Orchard Park excessive sins, and a habitual transgressor of natural terrace is made of silt, sand, andclay, and make up the law.” The prqject area is devoid of perennial water agricultu~xl surfaces of the Pecos Valley. The sources, except for thePecos River 6 km to the west. Lakewood terrace forms the bottomlandsof thePecos Any surface water is diverted into local sinks and River and major tributaries (Hawlcy et al.1076). It is basins or filtered quickly into the sands (Lovelace very alkaline and not good for agriculturalpulposes. 1972:30). ‘I’ypically, water has always been a problem Archaeological sites occur within and on top ol’ this to prehistoric and historic people in southeast Ncw surfdue, and some areincorporated into dune dcposits Mexico (Haskell 1977:13). (Collins 19x5). ‘I’he local gcology is defined by Permian Age San Between thePecos Vallcy and the 1,lano Estacado Andreas 1,imcslonc ofthe Itustler Formation (Dane and lie the Mescalero Plains (12eeves 1972). The eolian Rachman 1965) with extensive gypsum and anhydrite sand covcrs thc gravels thatcontain quartzite, chert, beds along the Pecos River. Associated derivative soils and olhcr siliceous materials Ibund in the Ogallala in the project area are Gypsiorthids-Torreothents- Formation. Dunes and hummocksare common in the Gypsurn Imdand Palcorthids-Haplargids (Makcret al. Mcscalero Plains. 1974) andarc manifested in thin eolian layers 10 to 80 cm deep over bedrock bases and broken caliche beds. Flora Additional eolian deposits offine-gainedI’ecos Kiver Alluvium are locally aggregated into active or dormant The majority of the locally sparse vegetation is associated with the Desert Grasslands (Figs. 2 and 3). dunes, in which depths abovc bedrockcan reach 5 m (Maker et al. 1974). Both Trqjan Hill and Macho Brown (1 982) rcfers toit as semidesert grasslands.He Dunes fall into the former of these depictions with also states lhat thisvegetation type is a “distinctive average depths under I m; howevor, Macho Dunes is separate biomc.” Shrub steppe combinedwith granla dominated by active dunes measuring 3 m in height. steppe form a piedmont desert grassland (Dick-Peddie Dissolution of thePermian bedrock that underlies 1hc 1903). Palatable grasses arereplaced by forb shrubsin Pecos Kiver is a process that may have begun in the intense gazing (Whitfield and Anderson 1938). Jurassic period (Hawlcy et al. 1976246). ThePecos Campbell ( 1929) also foundlhat overgrazing changed River and Ch*lsbad Caverns, along with nutncrotls grama grass range to mesquite dune type. The desert depressions, dcvcloped in the Permian (Katz and Katz grasslands border the Chihuahuan desert scrub (Dick- 1093). Katz and Katz (1903) also statc that three Pcddie 1993:107) and 1hc dominant grass is black massive collapses occurred under the PccosValley grama, although it is diflicult to separate thetwo. north of Carlsbad during the Quaternary period, Some of thegrasscs hat are present within the allowing the Pecos River tocut northward and capture Desert Grasslands arc alkaline sacaton, western any east-flowing streams. whcatgrass, Lndian riccgrass, galleta, blue grama and As the Pecos Vallcy widened and ducpcned, black grama, to name a few. Thcvegetation found in alluvial materials, such as the Ciatuna Formation, were deposited (Keeves 1972: I IO). Bachman (1 974) and the samplesfrom Macho Dunes correlate to the Desert Climate Grasslands as defined by Dick-Peddic (1993) and others. It includes little-leaf sumac, one-seedjuniper, Climatic changes occurredin New Mexico duringthe white thorn acacia, mesquite, fourwing saltbush, yucca, Tertiary period.Iksert vegetatioll appeared during the snakeweed, wait-a-bit, goosefoot,and purslane (Figs. time when uplifts of the Sierra Nevada, Cascades,and 2 and 3). Coast Range proceeded at a rapid geological pace Mesquite hasbeen utilized since prehistoric times. (Dick-l'eddic 1993). During the late Eocene to the There is extensive documentationofits use in southern Oligocene the climatein the Rocky Mountain West was New Mexico (O'Laughlin 1980; Oakes I98 I ; becoming cooler and drier (Dick-Peddie1993). With Carmichael 1981). Thereis a wide range of mesquite continued climatic changesand mountain uplifts during use, such as fiw food, salve, l'uel, fiber, and as a base the Miocene,new vegetation migrations wereinitiated for drink (Oakes 1981 ). Carmichael ( I98 159) states and grasslands wereformed (Dix 1962). Sage pollen that its popularity may stem l'roln widespread has been Iound from the dating to the late Miocene. availability, nutritional value, and seasonal Vegctation zones duringthe late Pleistocene were dependability. IJsually the pods are gatheredwhen ripc displaced to lower elevations of 3,000 to 4,000 ft in the fall; however, both the llower and pod can be (Dick-Peddie 1993). These zones are found today. eaten raw (Oakes 1982). Mesquite has an advantage DwarfJuniper was found at2,000 m (6,600 ft) in the over other vegetal resources, since it can be sorted Guadalupc Mountains. 'I'oday the closest is 400 km without special preparation (O'Laughlin1980). (248 milcs) north, In the Llano Estacado the spruce Other species suchas yucca, founving sallbush, forest changed to woodlands dominatedby piAon that goosefoot, and purslane havebeen utilized throughout stretched from southern Missourito north Texas (Dick- prehistoric timesin the Southwest(Toll 1997). Several Peddic 1993). Dick-Peddie also states that woodland pollen profilesat the Lubbock Lake site produced pine vegetation covered areas in southern New Mexico pollen dating ca. X600 to 8300 B.C. At Gardener 24,000 years ago. Today they are coveredby desert Springs cheno-ams decreasedand grass increased at grasslands or Chihuahuan desert scrub atan elevation about 2000 R.C. Pollen from oak, madron, hackbeny, of 1,750 171 (5,800 ft). juniper, and ash were foundat Pratt Cave dating 5S0 Major changes occurredbetween 15,050 B.C. and R.C. A.D. 1350. 'There was a warming and drying trend at ca. 10,000 H.C., with a drop in winter precipitation at Fauna approximately 6000 B.C. However,a drier shift took place between 3050 B.C. and 2050 B.C. Another Several faunal species are presentin thc area, suchas cooling trend occurred at 550 B.C. with a warming pronghorn, cottontail rabbit, and jackrabbit. Smaller trend at A.D. 850 (Dick-l'eddic 1993). flall (1977) rodent species are also found, such as gopher and believes that the climate stabilizedA.D. at I350 in New mouse. Antelope and mule deer have also been Mexico. occasionally noted. Species associated with riverine The local climate today is characterized by mild and wetland environments have beennoted on sites 2 winters and hot summers, with a mcan January to 3 milcs awayand include turtle, lish, and migratory temperature of 5. I "C (41 2°F) and a mcan of 26.3"C birds. (793°F) in July. The yearly mean temperature is Bison were present along the western boundaries approximately I5.9"C (60.6"F), with an average of200 near the PecosRiver until the late nineteenth century. frost-free days ('['uan et al. 1973). Summer-dominant Tro-jan Hill did not have any faunal remains; however, prccipitation patterns rcsult in a mcan annual rainfall Macho Dunes contained a number of taxa. They amount 01' 305 mm (12.0 inches), in which 203 nm include large mammal, medium mammal, small (7.9 inches) falls betweenApril and September (U.S. mammal, jackrabbit, cottontail rabbit, indeterminate Department of Commerce 3965; Tuan et al.1973). canis, mouse, gopher, rat, squirrel, porcupine, hawk, frog or toad, box turtle, and fresh water shell. One modern cow bone was recovcredjust below the surface (0-15 cm) in a small dune.

ENVIRONMENTAL SETTING 5 6 CYAKLSBAD RICLIEF ROIJTE CULTURAL SETTING

The most recent dctailed overviews prepared for Climate conditions improve around 2000 B.C. southcastern New Mexico include Stuart andCiauthier Thwc is a decreasein Chenopodiaceac and increasein (19X3), Sebastian and Larralde (1989), and Kalz and grasses at Gardner Springs.Hall (1985:112) states the Katz (I 993). These overviews have outlined the major decline inhtriplex (saltbush) and an increase in grasses sequences occurring in southeastern New Mexico could possibly be a response to a slightly wetter Palcoindian, Archaic, and Formative periods. climatc. Pollen from Pratt Cave in the Ciuadalupe Mountains around McKittrick Canyon Creek,in Texas, Paleoindian (circa 10,500 R.C. to 6200 B.C.) contained oak, madrone, hackberry, juniper, andash. Above the canyon whcrc the environtncnt is drier, A total of 113I'aleoindian sites have been recorded for grasses, cornpositcs, and herbaceous plants arc southeastern New Mexico, of which 19 are in Eddy dominant. Van Devendcr(1980) suggests that piiion County. Katz and Katz(1993) observed that these were nuls were carried in by Archaic people. Hotter and located on hilltops and not in the sand dunes. Mostof drier conditions brought steppe grasslandsto areas that lhe raleoiridian remains are found along the Llano once supported savamahs and parklands, causing Estacado, the Mescalcro Pediment, or the eastern scvere erosionin alluvial valleys (Katz and Katz 1993). escarpment ol'the Guadalupe Mountains (Oakes1985). Desert grasslands were thcmajor vegetation lype. By Eleven radiocarbon dates have been obtained and the Idate Archaic, modern plant communities were range from 10,000 B.C. to 5500 H.C. for Clovis, established, containing cacti, yucca creosote bush, Folsom, and Planosites. Katz and Katz(1993) refcrto ocotillo, and agave(Katz and Katx 1993). Bison were these as Paleoindian I, Palcoindian 2, and Palcoindian present in the Pccos Valley and theon Llano Estacado. 3. 'l'hey all have similar subsistencc adaptations, which At Frcsnal Shelter, corn, mescal,pifion, and yucca were include large mammal hunting and wild food gathering present. (Kyte 1984h: 12). Paleoindian components arc Some fauna recovered from caves in the identified by the type ol' spear points, such as the Guadalupe Mountains containcd bison, Merriam elk, Clovis point (Clovis phase), whichis characterized as deer, squirrels, and mice (Sebastian and Larralde 1989). Wimbcrly and Eidenbach (198 1) recovered a large lanceolate with thin a concave baseand bifacial tluting. One site has becn found by Hester (1972)at dccr, antelope, bison and sheepfrom Fresnal Shelter. Blackwatcr Draw Locality I. The Folsom point At the Gamsey Springs <:ampsite, Parry and Speth (Folsom phase) is a medium-sized lanceolate with a (1 984) recorded rabbit hones. Freshwatermussel shells broad channel flake on both sides (Katz and Katz are frequently foundin the middle Pecos Valley during 1993). Sites associated with the Folsom phase include the Archaic period. Blackwater Draw (Scllards 1952; Hester 1972; Stevens Although Archaic sites arc found in all the major 1973) andElida (Warnica1961). ccozones, most arc Sound in low ecozones such as scrublands, grasslands,and desert-scrub (Katz and Katz Archaic (5200 B.C. to A.D. 500) 1903). Overhall' of the sites recorded in southeastern New Mexico are found in Eddy County, in blowouts, The Archaic period in the southcast region seemsto dunes, ridges, and hillslopes of the Pecos Valley. The Ivwest elevations wouldbe the blowouts and dunes, start during the Middle Archaic insteadof the Early Archaic period (Wiseman 1997). Thereseems to be a while ridges andhillslopcsare slightly higher (Katz and hiatus between the Palcoindian period and the Katz 1993). beginning ofthe Archaic period. Katz and Katz(1 903) IIearths, burnedrock rings, and mulliple features attribute this to warmer climatic changcs. The changes are characteristic of Archaic sites. Fire-cracked rockis seem tohave takcn atoll on populations, andby 2000 common, as are lithic artifact scatters. There arc also a B.C. a lnodcrn semi-arid climate and amixed modern large number of plant processing features recorded, flora and fauna had been established for southeastern suggesting that there is a shill in the use or the New Mexico (Katz and Katz 1993:1 14). The warmer landscape or subsistence activities to plant gathering climate is also attributed to the scarcity of bison and along with hunting (Katz and Katz 1993). Katz and other large gamc, cvident at Lubbock Lake and Yellow Katz (1 993) also point out that habitation features havc Canyon. Subsistence practices includedboth riverine been recorded at some Archaic sites situated in the and upland plant and animal species (Wiseman 1997). grassland zonein the way of'caves and brush structures

CUL7'IJRAL SETTING 7 suggesting that small groups ofpeoplewere stayingat projectile points lor the Formative I and the Late specilic localcs for a limited amountof time. Archaic are dart points with convexto straight edges, Classification of Archaic sites is frequently straight to convex bases, and diagonal notching (Katz determined byprojectile point styles found on the sites and Katz 1993). The most noticeable characteristic (Chkcs 1985). Those types foundin southeastern New change is the sizeof the Archaicdart points compared Mexico arc provided in Leslie’s (I 978) typology. The to the Formative 1 points. The differencesare that they Early Archaic projectiles arc characterized as large, are larger than an arrowpoint, but smallcr than dart. a straight-stemmed points with indented, concavc, or Katz and Katz(1993) rcfcr to them as ”darrow” points. straight bases (Hokanson 1996) such as Jay, Bajada, The ground stone artifacts are one-handflat manos and Pcdcrnales, Kinney, and Pandora, to riame a kw, but unshaped metates. Bedrock mortars are also these are not generally foundin southern New Mexico. characteristic. Brown ware ceramics dominate the The Middle Archaic or Avalon phase doesnot contain assemblage and thc majorityare Jornada Brown.Pccos any diagnosticpoints, but thcrc arc several radiocarbon Micaceous Brown, South Pecos Brown, and AlnlaPlain dates rangingli-om 2687 f 210 B.C.to 1170 * 70 B.C. are also present. (Katz and Katz19XS). Middle Globc Phase (A.D. 750 to A.1). 950). This Thc Tcrnminal Archaic or Brantlcy phase(AD. 1 to phasc correspondsto the Formative 2 (Katx and Katz AD. 750) continues the previous patterns and 1993), the early Querecho (Leslie1979), and late 18- evidenced a greateruse ofburned rock rings (Wisernan Mile (Jelinek 1967). During the Middle Globe, 1997). With (he continuation, there is a definite identifiable architecture, arrow points, and black-on- increase in upland subsistence resourccs such assotol white pottery are the main characteristics. Several and agave. Most burnedrock rings are associated with radiocarbon datedistributions suggest a more sedentary chipped stone scatters; however, diagnostic artifacts way of’life (Katz andKatz 1993). Leslie (I 979) reports may be absent (Katz and Katz 1993).Projectilc points the presence of clay floors in association with the are classified as tnedium-size dart points with stems shinnery oakbelt in the southeast. Jclinek(1 967) found that vary in shape. Thc appearance of diagonal pithouses and contiguoussurface roomsof this phasc notching onprtjectile pointsis also ovidcnt.Projectile in the Middlc Pccos Valley. At Brantlcy, stonecircles, points such as San Pedro,Pccos, and 1,eslie’s (1978) rings, and scatters were noted, suggesting an upland h;C, hD and 8A types are associated withthe Brantlcy subsistence and settlement orientation instead of a phase. riverine one (Katz and Kalz198%). The ground stone artifacts remain thc same, and Formative Period (AD.500 to A.D. 1540) there is no noticcable appearance change in the prqjectile points, cxcept in size. Types such as1,cslie’s Changes occur, marking the endol’thc Archaic period. 3A (Scallorn) and 3R (Scallorn variant) arc a Ceramics and arrow points arc found,as are domestic characteristic of the phasc.1,ocal brown ware ceramics features. However,Stuart and Gauthier (1981), Oakcs predominate and include South Pecos Brown and (1982), and Fificld (IgX4b) argue that Archaic Middlc Pecos Micaceous brown. Thefirst appearance adaptations continued into the Ceramic period, of painted pottery is found at this time. Red Mcsa incorporating ceramics and cultigens.Kats! and Katx Rlack-on-white is conlmonly found in the northern (1993) also find similarities between the Archaic and region and Cobolleta Hlack-on-whitcin the southern their Formative I phase. region. Late Globe Phase (AJX 950 to A.D. 1075). The Late Globe phase corresponds with the Formative 3 phase in the Brantley arca(Kats! and Katz 1993), Early Mesita Ncgrain the Middle Pecos area (Jclinek1967), Karly Globe Phase (kD. 500 to A.D. 750). According to Katz and Katz (1993) the Formative 1 Lehrner’s (1948) Early Capitan and Mesilla phases01’ period includesthe Late IIueco ([die1979), Early 1 8- the Jornada Branch, and Late Querecho(Corky 1965; Mile (Jelinek19A7), and Globe (Katz and Katz198Sa) Leslie 1979) in the Eastern extensions. phases. Because the Globe phasc is included in the The Late Globe phase is characterized by large Formative 1 period,thc dates fromthe Katz and Katz sites. Leslie ( 1979) notes largea concentration ol’small l!l93 overview are earlier and it is assumed that the pit structures locatedat permanent water sourcesand Globe phase began earlier than A.D.750. also oricntcd toward the exploitationof shinnery oak These earlier patternsofa riverine orientation and (Katz and Katz 1993). Larger sites wcrcalso observed short-term camps near water continuein thc Brantley by Jelinck (1 967) in the Middle Pecos rL&’’wn. arca of the Middlc Pecos (Katz and Katz1993). The

8 CARLSRAD KELlEF ROLJTE The projectile points remain tho same in size; Architecturally, the beginning of the Oriental however, thcrc arc more Scallorn and Scallorn variants phase contains subterraneanstructures. At about the (Type 3A to 3F) and 1,ivennorc types (Leslie1979). middle of the phase (A.D.1200) the structures become The ground stone exhibits more modificationseen as in rectangular slab-based surface rooms (Katz andKatx oval basin metatcs and convex manos (Leslie1979). 1993). Brown ware ceramicsstill dominate the assemblage; Other phase sequences for the same time period however, both local and nonlocal gray wares appear. are Jelinek’s (1967) Late Mesh Negra, Crosby phase, The dominant type of gray wareis Crosby Black-on- and Late McKenzic, Lcslie’s(1979) Maljamar and the gray. Local and nnnlocal brown wares continue and Early Ochoa, and the Katzcs’ (1993) Formative 4 Cebolleta Black-on-white replaces MimbresBlack-on- through 7 for the southeastern region. white.

Or.ientu1 Phrrse (A.D. 1125 to A.D. 137.5) Very little is known about the Phenix phase; however, While plain brown ware pottery continues throughout Katz and Katz (1993) state that there are dates and the region, thcrc are stylistic changes during the dated sites assigned to this phase. Intrusive pottery Oriental phase, the most noticeable being painted disappears from sites and local textured brown ware pottery. E1 Paso Polychrome and Three Rivers Kcd-on- such as Ochoa Indented is the remaining type. terrocotta are present in the early partof the phase (ca. Projectile points such asthe Washita and ‘Toyah are A.D. 1150), while Lincoln Hlack-on-red and present on the sites and are usually the only diagnostic Chupadero Black-on-white arcstill present aroundthe artifacts. Radiocarbon dah, along with local pottery 1400s. and projectile points show that several large Thc projectile points or arrow points arc small and architectural sites of the late Oriental phase continued triangular. ‘I‘hesetypcs occur throughout the Southwest to be occupied during the Phenix phase and in the Plains. They are similarto Leslie’s (1979) Tipi rings insteadof small stone circlesare now types 2A and 2B and are characterized as side-notched present. Gunncrson(1956) states that the Athabaskans forms with concave bases (Leslie1979). The ground apparently appeared in the Greater Southwest, stone remains the same inas the previous phascs, with including southeaslern New Mexico, sometime within oval basin metates; however, the manos (one-hand) this 125-year interval (Katx and Katz 1993).One othcr more frequently have opposing ground surfaces (Katz phase, thelate Ochoa (Leslie 1979), corresponds with and Katz 1993). the Phcnix phase. 10 CARLSBAD RELIEF ROLJTE DATA RECOVERY PLAN (Adapted from Wiseman 1996b)

'llcorctical Perspective response to the local environmenl.

Prehistoric occupationof southcaslern New Mexico has Background Considerations been documented from the Palcoindian periodthough the presence ol'the Mescalero Apache, a periodof over Katz and Katz( 1985a) providean excellent outlineof 13,000 years. Earlier research goalsin thc region were prehistoric cultural developments in the Guadalupe geared mainly to defining the culture history of the Mountains-Carlsbad region.But the Katxes would be region (Lehmer 1948) and its specific cultural attributes the first to admit that this sequencc, which covers (Mera 1943). Interest in refining cultural Palcoindian through early historic Nativc American manifestations, principally ceramics, has continuedup periods, requircs verificationand elaboration. 'The last to the presentwith the work of Greer(I 963, Runyon two Formative periods-Globe (AM. 750-1 150) and and Hedrick (1973), Brook(I 973, and Leslie (1979). Oriental (AD. 1 150-1450)"are not as well known as IIowever, today professional research goalshavc taken earlier ones, largely because aboriginal use of the on broader, more regional aspects with primary greater Carlsbad area had decreased markedlyin favor interests in differentiating sedentary phases (Whalen ofthc Guadalupe Mountainsand their foothillswest of 1977; Katz and Katz19XSa), dctcrmining site hnctions he Pecos Kivcr. The two project sites likely represent (O'Laughlin 1980; Wiseman 199h), assessing the Brantley (terminal Archaic) and Globe (Formative subsistence bases (Basehart 1974: Oakes1985), and period) phases and lie east of the foothills of the correlating site locations with environmental Guadalupe Mountains. In the nearby area, the parameters ((lakes1985; Katz and Katz 1993). While Querecho phase represents the same cultural adaptation explanation of the settlement-subsistence dynamicsof as the Globe phase. 'l'hey provide an excellent southeastern New Mexicois still tenuous at best, the opportunity to examine prehistoric adaptations during potential for systetnic cxplicationnow and in thc future these phases and to verify or modify the shift in is extremely promising. subsistence emphasis positedby the Katzes. Several general statements about the environment As discussed in more detail below, horticulture and the sitcsmay be made. The desertic conditionof evidcntly was not practiced prehistorically in the unproductive soils, frcqucnt lack of potable watcr, Guadalupe-Carlsbad region. This fact, plus other undcpcndable precipitation, and surficial nature of the characteristics, havc led Robert Malloul' (198s) to cultural remains would secm to preclude use of the suggest that the prehistoric remains of the southern sitcs as sedentary habitation units. We suggestthat the Guadalupe Mountains are more closely associatedwith sites were temporary campsites or specialized activity the 'l'rans-Pecos culture areaofwest Texas(the wcstern locales used by foragers and collectors ofdifl'ering "arm" of the state, exceptEl Paso County)than with the chronological periods. Jornada-Mogollon to the wcst and north. We concur We are basing our proposed research uponan with Mallouf. Drawingon the Katzes work at Brantley, environmental frame ofreferencethat states culture is wc suggest that the same appliesto the Carlsbad areaas adaptively organizedto solve specific problemsposed well, including the sites being considered for the by thc cnvironment. One primary problemlor regional present project. However,a formal line of demarcation groups is the acquisition of subsistence items. 'rhus, between thc Trans-Pecos (including the Ciuadalupc- variability in a culture's systemic organization is Brantley region) and the Jornada-Mogollon remainsto responsive to the variability in availability of food. be defincd. Some adaptive responses could include collecting, There are several implicationsof the assignmentof foraging, hunting, storage,trade, sedentism, mobility, the Guadalupe-Carlsbad region to the 'I'rans-Pecos. etc., or any combinationthereof'. We believe there were First, as far as can be ascertained at present,the peoples specific environmental variables the in Carlsbad region inhabiting the Trans-Pecoswith the exceptionof thosc that conditioned the selection of particular food at La Junta delos Rios on the Kio Grande (present-day procurement strategies such as (1) seasonality of Presidio, Texas)lived an Archaic-like, hunter-gatherer availability, (2) quantity ofbiomass,(3) accessibility of lifestyle throughout the prehistoric historicand periods. resources, and (4) density of the participating Many late prehistoric sitesin the 'Trans-Pecos produce population. We believe the generalized strategy ofwild small amounts of pottery,but all of it was probably food gathering would have been the best adaptive traded in from nearby regions. Most or all of the

DAT4 REL'OVLKY PLAN I I pottery on Guadalupe-Hrantley sites came from thc trash, and similar ranges of artifact types. They arc Sierra Blanca andE1 Paso regions to thenorthwest and occupied forrelatively short periods of time (daysor west. rcspectively. few weeks), relatively few items (manufacturingdebris, broken artifacts, etc.) should be left behind. IIuntcr-Gatherer Subsistence Systcms Collectors send out work parties to set up Past research in thc Guadalupe-Carlsbad region,as in temporary special-activity sites, collect the targel the Trans-Pecos in general, indicates that baked resource(s), and takethe food back to long-tenn succulents such as lechuguilla and sotol wcre a base camps. Thecharacteristics of both should bc fundamental aspectofpottery pcriod (IdatePrehistoric) as follows: subsistence (Grccr 19hS, 1967, 1968; Roney 1985; Katz and Katz 198Sa). Archaeological rcmains of Hnsc crrmps are generally quite visible baking ovens usually takethe form of midden ringsor archaeologically bccause theyare used fora wide circles of burned rock surroundingcentral pits, though range of daily activities, rcsulting in the burned-rock moundsof other shapesare also known (S. accumulation of a wide range of artifact typos, Katz, pers. comm. 1996; R. Phippen, pers. comm. activity areas, and refusedeposits. Some form of 19%). Midden circlcs date as early as the Middle structurc, whether ephemeralor more substantial Archaic pcriodin Texas but aremore common in later in construction, is usually present, asare pits for titne periods. Most datcd ovens in the eastern Trans- the storage of food and other items. Base camps Pecos, including theCruadalupe Mountains, belongto are generally used over long pcriods of time the post-A.D. SO0 pottery pcriod (Koney 1985: 144). (several months) each ycar for several years, Since these succulents providea reliable, year-round sometimes in scquential years and sometimesin source of carbohydrates, they were understandably staggered years or sets of years. A logistically important to prehistoric andhistoric diets and probably organized group generally has only oneor two obviated thc value of, or need for, many othcr basc campsthat it uses during a givenycar. carbohydrate sources including corn (Sebastian and Larralde 1989; Roney 1985). Spwiul nc*tivity.sitcs,on the other hand,are created W. H. Wills (1988:54-55) points out that during collccting expeditions, mightbe used only succulents are usually scattered acrossthe landscape oncu, and are almost invisible archaeologically rather lhan clumped, which probably afltcted humans because they are usedfor only short periods, have in yet another way.He posits that the scattcrcd nature little or no accumulation01' nonpcrishablc dcbris and year-round availability of thcse resources in the and broken artifacts, and have limited artifact 'l'rans-Pecos led to the retention of a more nomadic, inventories that reflect comparatively few "forager" pattern, rather than a less nomadic, activities. logistically organized pattern (Binford 1980). Tn simplest terms, foragers move to the food, and While we generally agree withWills' proposition, we, collectors move the food to the people. Colleclors do like Scbastian and 1,arralde (1089) and Collins this by meansoftask groups that are sent out to obtain (1991:8), emphasize the viewthat these strategics-- specific resources and return thcm to the group, a foragers andcollectors--arc two endsof a continuum, behavior warrantedby resources that occur in clumped not a dichotomy.In a given year or over a seriesof or patchlikc distributions. The primary differences years, some groupsmay actually employboth strategies bctween collector andI'oragcr lifcstyles are the degrees because offactors relating to season, climatic regimen, and waysin which peopleplan, organize, and conduct economic success, demography, competition, and othcr their food-quest in response to resource distributions factors (see Boyd ct al. 1993 for a recent discussion). and seasonsol'availability. Sebastian and Larralde present an example of a In theory, forager and collectorsites should have "mixcd" foragericollector strategyin the concept of fairly distinctivc attributes. 'These might be summarizcd "scrial foraging." Using the Archaic peoples of as follows: southeastern New Mexicoas an cxamplc, they define serial foraging as follows (Sebastian and Larralde Forager sites should be similar, and their 1989:55-56): archaeological visibility should be subtle, perhaps even inconspicuous, becausupeople are tnovedto A strategy of serial foraging involves a small the resourccs, the sites are inhabited for shortcr residential group that moves into the general periods of time, have smaller accumulations of vicinity of an abundant resource and campsthere,

12 CARISBAD RELIEF ROIJTE uses the target resource and other hunted and on the other hand, would rely mostly on those plant gathered resources cncounteredin the gcneral area species that me avajlable throughout the year, until the target resourcc is gone, or until another precluding the need forstorage but usually requiring desired rcsourceis known to be available, and then greater mobility becausetheir distribution across the moves on to the next scheduled procurementarea. landscape is general, not patchy. Collins suggeststhat Such a strategy couldbe expected to crcate a great animal species mightalso be conducive to this type 01’ deal of redundancy in the archaeological record, analysis, but becausc animals aremclhile, they are not an endless seriesof small, residcntial camps from particularly useful in this regard. which daily hunting-and-gathering parties move In regards to subsistence strategies, it is out over the surrounding terrain, returning to appropriate to touch on the subjects of gardening- process and consume the acquired foods each farming and foodstorage. The evidence forprehistoric evening. If the resources were randomly horticulture in the Guadalupc-Carlsbad region is distributed, allthe sites would look generallythe minimal at present. Roney (1985:44) statesthat corn same. But since many of the resources appearin was recovered from only thrcesites, all of them caves the same place year after year orin some other in the Guadalupe Mountains, butin each case, few cyclical pattern, some sites tendto be reoccupied. remains were found. ThePratt Cave example (now published as Schrocdcr1983:67) involves oneor more Reoccupied sites, then, would bea clustering of small, corn kernels recovered from thevicinity of a hearth. single-event, serial-foraging sites. But, Sebastian and Since two chile seeds were recovered from a lower Larralde (1 (dX9:Sh)envisage a complicating factor: level in thc same test, it seems likely that thc corn was introduced during the historic period by Apaches, The only exception to the rule of basically rather than during Archaic times as suggested by redundant but sometimes overlapping small Roney. According lo Roney, the proveniences and campsites would be the winter camps. Giventhe temporal associations of the other two reports of’ corn relatively brief winters ol‘ the Koswell District, are also uncertain. This leads us to conclude that many of the sites would, on the surface, be no horticulture either was not practiced by many of the different in appearance from reoccupied short-term prehistoric inhabitants of the Guadalupes or was cainps. Excavation of such sites might recover practiced on only a vcry limited scale. Degree of resources indicating a winter seasonal occupation dependency on corn in thc Carlsbad area is an or features indicative ofstorage, however. If we unresolved issue and will be addressed with the were able to differentiate single, large-group subsistence data obtained from thc two sites to be occupations from multiple, small-group excavatcd. occupations, we might findthat winter sites diffcr Storage, usuallyin the form ofpits, is believed to from warm season camps in that they were be a keysignal as to the existence andidentification of occupied by larger groups. base camps and habitation sites. The storage of quantities offoodstuffs is a characteristic of logistically In thc above scenario, the settlement types ofserial organized subsistence systems. Generally speaking, foragers should thenstart taking on the appearanceof storage implies a location that is easily protected or collcctors’ sites. Whilethis introduces some difficulty otherwise secure from theft. Scbastian and Larralde in archaeological studies, it probably approximates (1 989:RA) advance the interesting hypothesis that, reality to a greater degree and certainly scemsto make because some resource patchcs are spread overthe better sense with respectto the archaeological recordof’ landscape and create a logistical problem for southeastcm New Mexico as we become increasingly exploitation, some people may actually havc cached familiar with it. foods in the collection areas and then movedtheir In addition to feature and artifact content ofsites, families from cacheto cache as needed throughoutthe Collins (1991:7-X) suggests biological correlates of winter season.‘I‘his constitutes yct another variation on forager and collcctor sites,particularly those involving the forager theme. But whileit may actuallyreflect the burned rock middens.He suggeststhat the difference situation in southeastern New Mexico,it also has the between thetwo might be signaled by whether the plant strong potential for confusing the interpretation of species processed are r-selected or not. ‘l’hat is, archaeological remains. collectors would focuson r-selected species that are So how does one cometo grips withthis problem‘? available in large numbersiamounts during short Collins ( 199 1 :7-X),in discussing researchon burned- periods of time, requiring some form of preparation rock middens in ‘l’exas, provides us with a test for and storage for long-term benefitto humans. Foragers, determining whether a forager systemor a collector system prevailed during the occupation of a spccific will investigate these features. Partof the effort will site or set of sites. He posits: also focus onfinding and excavating anypits or othcr features currcntly masked by the dunes and artifact Therefore, complex components associated concentrations. Every effort will be made to rccover with burned rock middens which evidence nnd record information pertinent to the research quantities of remains of any one or more r- outlined below and thespecific questions that follow. selected resources to the near exclusion 01' other kinds of resources imply, at least to (I) Evaluate (verify or modify) our perceptionofthe some degree, the adaptive characteristics cultural content of thc Brantley and Globe phases, and listed above and would favoran interpretation whcrc possible, augment the criteria by which the that burned rock tniddens were specialized phases can be distinguished. These phases span the food preparation leaturcs. Mesquite beans, terminal Archaic through the early Iate Prehistoric prickly pcar tunas, all deciduous nuts suchas periods in the regional sequence (Katz and Katz pecans and acorns, and psoraleaaxe examples 1985a). of r-selected plant foods. The geographic distribution of burned rock tniddens [in (2) Evaluate (substantiate, refute, or modify) the Texas] docs not encompass the range of any subsistence trend outlinedby the Katzes (I 985a) for the notable r-selected animal species, however, C'arlsbad area. The Katzes believe that a major seasonal availability of sotne animals, such as subsistence shift took place during the prehistoric bison or migratory waterfowl could sequence. Riverine resources suchas mussels were sometimes trigger bchavior similarto that of important foods during the Avalon, McMillan, and r-sclcctcd resource exploitation, but thc Hrantley phases (Middle Archaic through terminal availability of such resources is not Archaic), and nonriverine resources were largely sufficiently reliable to result in the supplemental. Rut starting in the Brantlcy phase, and establishment of the same adaptive pattern. continuing throughout the Globe, Oriental, and Phenix phases (the entire Late Prehistoric period), upland In contrast, plant and animal foodsthat resources became more important and riverine are edible and available forall or much ofthe resources lcss important. Whilethis is better conccivcd year (sotol, prickly pcar pads, lechuguilla, as a change in emphasis, rather than a sharp changc antelope, rabbits, deer, bison in some areas, from one set of resources to another, it Icd to a fish, mussels, hlrkey, and othcrs) can be markedly reduced human prcscnce along the Pecos exploited in the morc generalized foraging Rivcr. strategy and have different behavioral Although the Katzcs' reconstruction of the correlates. Evidence that foods of this kind settlement and subsistence patterns appearsjustificdby providod the principal staples of groups their data, we believe that the number of sites and responsible forbumedrockmiddcns wouldbe components invcstigatcdby them are relatively few in evidence that these were not specialized food number and, being concentratedmar thc Pecos Kiver processing facilities, and that those channel, do not fully rcprescnt the river valley responsible may havebccn foragers. occupation. Our projectsites, being closerto the river, should permit us to he-tune our perceptions of the entire riverine settlement.

The investigations proposedfor the projectsites will be (3) Determine whether the inhabitants of the directed towards answcring basic questions about Guadalupc-Carlsbad region farmed andifso, determine settlement and subsistcncc behavior the in north endof' how promincntly cultigens figured inthe diet rclativc the Trans-Pccos culture area. The main thrust will to wild foods. Given their proximity tohorticultural focus on documenting and validating the culture peoples of the Southwest, it is surprising that sequence recently formulated and outlinedby Paul and prehistoric peoples in the Guadalupe-Brantley region Susana Katz(19859, expressed as follows. did notfarm. Hut, assuming that they did not farm, we Judging by surface manifestations, LA 29362 is then need to determine whethcrthe rcasonsare cultural, Archaic and LA29363 is early in the Late Prehistoric demographic, climatic, or some combination of these. or Formative period. k'eature types identified include Could it be that the increased utilization of succulent possible hearths, structures, burned-rock scattcrs, and baking precluded the need for, or uscfulncss of, the artifact scatters. The proposed data rccovcry project adoption of farming, as hasbeen suggested'?

14 CARLSBAD RT:I,IEF KOU'I'E 1. Are the prehistoric components of the projrct also present on this and LA 19362'? It is virtually sites foraging or collecting, base camps/habitation guaranteed that the two sites were occupied more than sites, special activitysites, or some combination? once during the prehistoric period. Assuming so, we need to discover not only what kinds of featuresarc If' LA 29362 lunctioned as a foraging locus, the present, but also which ones were contemporaneous following characteristics are expected: and which were not. Were activitiestho or site function during each component the same ordifferent'? (1) Evidence of repeated short-term occupations At this stage in the investigations we have few (numcrous redundant features scattered over the observational data and facts by which to judge the landscape). Attributcs that should not occur include answers to these questions. More intensive workwill long-term storage features, residential structures, probably greatly modify our perceptions and formal midden deposits. Attributesthat may be present interpretations of the prehistoric components at the may include ephemeralstructures, sheet trash deposits, project sites. The minimal data available suggestthat and a wide variety of manufacturing maintenance and two or more components arc presentLA at 29363 and food procurement activities (Moore 1996). probably rcpresent twoor more phases in the Katzes' sequence. 'I'he validity of' this expectation requires (2) Formal interior heating should be absent contimation. To do this, we will need to discover, isolate, and study fcatures andartifacts belonging to (3) Evidcnce for a wide range of floral and faunal separate occupations (componcnts). resources in the diet. Cultigens wouldlikely be rare. Once individual componentsare defined, wccan Only local food remains should be found. thcn proceedto document theactivitics that took place at each. Thecultural fcaturcs (storagepits, other types If I ,A 29363 was used by ceramic periodlogistical ofpits, hcarths, bakingpits, etc,), associated artifactual task groups, the following characteristics could be materials, and the patterning of these remains are expected: critical in defining site types throughan analysis ofthe activities represented. lrnportant subsidiary studieswill (1 ) Evidence of relatively longer periodof occupation assist in determining site type, as wcll as overall than at LA 29362. Storage facilities may bc present and subsistence patterns, and include floral, faunal, and there maybe specilictrash disposal andactivity areas. artifactual data, as discussed below.

(2) Structures should be shallow andreflcct wam- 2. What artifact assemblages are present at the season use, although interior hearths could bepresent. project sites'? What typcsof tools and manufacturr debris are present'? What are the relative (3) Evidence for a wide range offloral and faunal abundanccs of the various types?On the basisof the resourccs in thc diet. Cultigens may occur. Foods from artifacts, what typesof activities were performedat nonlocal sources may be found. the sitcs?How do these assemblagus comparewith those from other sitcsin the region? (4) Structures andor thermal features should he present and may evidence signs of reuse. Or there may be Both sites have producedlithic artifacts. LA 29363 evidence of redundant or related features representing has also produced pottery and ground stone. Intensive repeated use over time. surface investigation and excavation may produce other artifact typcs (projectile points, bifaces, ornaments, (5) A variety 01' food containers (ceramics) should he etc.) as well, present, although thcy may he limited in number. The typesof artifacts ata site hclp deline the kinds of activities that took place at each specific location It may be difficult to distinguish between these patterns (component). Manos and metatcs imply grinding plant of use in some cases, particularly if curation of tools foods, projectile points imply hunting, and scrapers occurred at either site. However, the compilation01- imply hide dressing. Multipurpose tools such as data should allow us to ~SSCSSthe assemblages and hammerstones, awls, anddrills, and manufacture debris determine he patterns of use. such as chipped lithic debris, shell fragments, and some LA 29363 (Macho Dunes) containsa possible pit types of fragmentary artifitcts, imply a host of structurc and a baking pit, and several burned-rock generalized activities involving the manufachlre or areas. But arc storage pits, other kinds of pits (for maintenance ofitems associated withday-to-dayliving. processing foods), and other typcsof thermal features A wide rangeof artifact and debris types implya base campihabitation situation, and fewer artifactand debris document, it is itnperative thatwe cstablish whetheror typcs imply special activity sites.‘I’hc percentage of not domestic plants were grown in the Guadalupe- cach categorywill providc a wry rough indcx to thc Cxlsbad region. Leslie’s (1979) assessment of the relative frequencyof occurrence of each activity at thc structural sites in the vicinity of Hobbs in far site. southeastern New Mexico, though without beneiitof Caution is required in intcrprcting thc datain this ilotation and pollen recovery techniques, suggeststhat tnanner becauseofthe cffccts oftool use-life on artifact corn was not being grown east of the Pccos River assetnblage composition (Schlanger 1990). Thisline of within New Mexico. The WIPP Project (Lord and interpretationrnakes several assumptions aboutthe data Keynolds 19R5), locatcd between Leslie’s sites andthe and thc activities they represent, and the technique Pecos River, excavated three r~onstructural sitesbut greatly simplifies a numberof complex variables and failed to find evidence of‘ cultigens in llotation and conditions. pollen samples. On the other hand, corn was clearly We will compare the project siteswith other sites being grown within the Pecos Valley at Koswell in the Guadalupe-Carlsbad region. Sites tobe usedin (Kelley 1984, Appcndix 6; Rocck and Spcth 1984; this comparison includc cave, shelter, andopen sites Wiseman 1984) and probably near Fort Sumneras well investigated by the Katzes (1985a) and Southern (Jelinek 1967). ‘Thus, if cultigens arc documentedfor Methodist University (Henderson1976; Gallagher and the project sites, thenthe relative quantities may help Hearden 1980; Koney 1985; Applegarth 1976). us dctermine ifthc sitc occupants were fanncrs or full- time hunter-gatherers. Relatively large numbers of 3. What plants and animals were being processedor domestic remains would indicate that the people were consnmed at the project sites? Whal biotic farmers. Small amounts of cultigens would be less communitirs wcrr being rxpluitcd? Wrrc thr clear, for huntcr-gatherers could have obtaincd themin inhabitants ofthe sites exploitingall available biotic trade from farmers. communities or only selected ones? Were cultigens being grown and consumed? During which season 4. What exotic materials or items at the sites or seasons were thcsitus occupied? indicate exchangeor mobility?

The project sites havethe potential 01’ producing Intensive surface investigation and excavationat burncd plant retnains and possiblysome animal bone. thc two sites may produce cxamplcs of imported Choking activities probably took place at both sites,as materials. At the prescnt time, some scholars also attested by the probable hearths, baking pit, burned- believe that all pottery is intrusive to the SevenKivcrs rock concentrations,and quantitics of burned rocks on region inthat it was produced in the Sierra Blanca and LA 29363 and by thc extensive lithic scatter at LA traded into Seven Rivers. Since exotic or trade 29362. materials are by their very nature generally few in Plant and animal remains rccovcrcd at number in any site, concertedeflbrl will be made to archaeological sites provide first line evidence for recover them. reconstructing various aspectsofthe human food qucst. Materials and artifacts not naturally availablein a Animal bones and the pollen and charred remnantsof region are indicativeof either exchange relationships plants will be studied to identify thc species present and with other pcople or a mobility pattern that permitsa the biotic zones exploited, characterize the diet and group to acquire these items during their yearly round. food preparation techniques, and provide insights into Judging which situation is applicable to the project sites the el‘fects of taphonomic proccsses on the is diflkult and will require careful comparison with archaeological record.tilord and faunal data also have data from the Roswcll region. If we can dcterminc the potential to provide informationon season of the whether the site occupants acquired the goods through year that they were collectedor hunted. Although only trade or by direct access, wewill gain perspectiveon certain plant and animal rcmains provide seasonal data, the territory they used and therefore on the identity of they are very usefulin helping define the time of the the people themselves. year the sites were occupied. Sinccit is unlikely that The abscnce of’ cxotic materials isanother matter the data frotn the project sites constihlte a totalof view entirely. In small sites and sitesof short occupation, thc the diet throughoutthe year or through time, itwill be absence of exotics can be tnisleading simply because necessary to compare these resultswith those of other such items may not have had time to find their way into projects in the region to gain a better understandingof the archaeological record. Or, perhaps thc occupants the total subsistence system. simply did not acquire exotic materials. Rut this is As mentioned in an earlier section of this prcciscly where comparisonswith other assemblages in

16 CARLSBAI) RE1,IEF ROLJTE the region and the long-tenn accumulation of chronology, phase sequence, and postulated cultural excavation data from numeroussites, both large and changes for the Guadalupe-Carlsbad region. small and of all types, is necessary for acquiring perspective and, eventually, resolving the problcm. The dating situation is critical in southeastern New Mexico where dendrochronology, themost accurate 5. What are the dates of occupation at thcvarious and preferred dating technique,works poorly ornot at project sites? all (W. Robinson, pcrs. comm. 1975). Few absolute dates derived by other techniqucs are currently Sincc it is likely that the sites were occupied on available (Sebastian and Larralde 1989). Rccent one or more occasions during theprehistoric period, advances in radiocarbon dating makeit the most viable dating individual features and componentsis crucial. At technique for southeasternNew Mexico atthe present thc individual fcature level, we need to determine time. Obsidian hydration and thermolumincsccnce have which are contemporaneous (or approximatelyso) and bccn tried in the region, but bccause these techniques which are not. Thiswill enable usto define thc datesof have many problems and are not gencrallyreliable, each component and thc activities performed atthe they will not be used in this study. diffcrent time periods at the sites. 'Ihis in turn will During excavation, charcoal will be recovered permit documentation of site and region use through from as many features and cultural situations as time, whetheror not these uses changed through time, possible. Because ol' the importancc of dating the and if they did change, thcdircctions, intensity, and, project sites, we will submit both very small samples hopefully, the reasons for those changes. Dating (for accelerator mass spectrometry analysis) and bulk information will also permit us to assess the Katzes' samples (carbon-stained sands) for dating if necessary. CARISBAD RELIEF KOUTE SITE-SPECIFIC DATA RECOVERY QUESTIONS (Adapted fromOakes 1997)

LA 29362 activity at the site, such as rock quarrying, gathering of specific wild food, or hunting for game. The nutnerous Trojan Hill consists of an extensive lithic artifact lithic artifacts and lack of subsurface tnaterials spread concentration focused in three areas on thc south side of along the ridge would suggest repcated encampment of the proposed highway construction. It is believed to bc a very short term by the samc or similar cultural group. late Archaic. However, conteniporaneity of thc several Broad areas ofthe site will be surfice stripped to thc concentrations needs to bc established through absolute former prehistoric surface in order to recover any hearth dating tncthods or by thepresence of diagnostic artifacts. or pit areas. Any soil found in these features will be Excavation procedures will concentrate on the denscr rctricvcd for nlacrobotanical and palynological analyses. artifact clusters in search of hearths or pits Illat may Dating of features is a priority. In the absence of datable contain charcoal for dating. Flotation and pollen samplcs matcrials, the lithicartifacts will be exarnincd forknown will aid in dctcrmining the subsistencc items utilized on Archaic attributes, e.g, high frequencies of bifacc the site and the season of sitc use. thinning flakes. Eighth-inchscreen will be used in order The presence of a large number of' lithic artifacts to retrievc any small lithic items. will help in detcnnining the activities performed at the site and provide a functional sitc classification. lfthc site LA 29363 is Archaic, a large proportion of the lithic assemblage may be biface flakes. Efforts to carefully retrieve such Given that pits and thermal features are present at LA flakes will bc madc through the use of smaller-sized 29363, we propose that the area served as a locale for the screens, if necessary. No sherds should be associated temporary collecting of subsistence items. Thc presence with any Archaic features. of ground stone indicates that processing of materials We also needto determine ifthe sitc is characterized occurred on thc site. by a single use episodc or is a result of multiple Macho Dunes has many discrete blow-outs occupations. Ifthc site represents apalimpscst ofArchaic containing culturalmaterial within thc dunal topography. activities over time, any rctricved dates may exhibit a Whilc ceramics post-dating A.D. X00 werc recovered relatively wide rather tlm narrow time frame. Our from several of them, not all locales contained shcrds. understanding of Archaic mobility patterns arc almost This could indicate an earlier and repcatcd use over a unknown iix the region. If Archaic pcoples are fully long period oftime for thissite. netcrmirling temporality mobile, we would expect expedient investment of labor for the many cultural proveniences within the proposed in hewths, storage facilities, and dwellings. The presence right-of-way is important for undcrsknding periods of or absencc of exotic materials will tell us considerably sitc use and various site functions. The patterning of about Archaic mobility and cxchange networks. cultural features within thc dunes is also critical to Llomestication of cultigens would not be likely. If site understanding cultural associations of the various blow- occupants maintained a seasonal round between thc out locales. riverine cnvirmmcnt of tlle I'ecos Rivcr and the The presence of possible residential pitstruchlres is Guadalupe Mountains, only seasonal subsistence very important because such features arc rare in the resources should be present in thc archaeological record. Carlsbad area and would provide a major stup toward In addition, a highly mobile group wouldnotbe expectcd understanding residential mobility patterns during the to produce long-term storage facilities (as stated above); Formative period. The roasting pit in another howcvcr, tenlporary facilities for the galhering and provenience could providemuch nccded subsistence data holding of specific food items while awaiting processing and information on seasonality of use or the area. might be present. Surfaces around these and all cxcavated featurcs will be I ,A 28362 tnay represent a foraging locale probably carefully traced in order to recover all associated used repeatedly over time by Archaic populations. components. Hcarths should be extramural and exhibit expedicnt Whilc it is assumed that all ceramics are intrusive to preparation, although rnmy may be scattered over the the Carlsbad region, petrographic analysis of all, or a site. 'J'hese could also be evidence of some specialized sample of, the recovered sherds should determine the presence of local or exotic tempers and help to resolve If LA 29363 is a ceramic period collecting area, the issue of whether all ceramics itre intrusivc. Because used possiblyas a tcrnporary campsite, then pits, hearths, this issue has not been definitively decided, an OM grinding stones, and food containers should be prcscnt. ccratnic expert will visit thc site region in search of’ ‘fhc presence of possible temporary rcsidential pit potential clay and tempering resources. Kecovered structurcs would probably indicate a base camp area. ceramics will also be used to cvaluatc the validity of Interior hearths within structures would suggcst a winter placing the site within the Globe phase (or Formative occupation period. period; Katz and Katz 1993). Every attempt will be made to locate all cultural Ground stone in several discrete locations on the sitc feahlres and retrievc macrobotanical and palynological indicates the processing of subsistcnce items, possibly samplcs from these. Datable materials will also be mesquite products, or grasses. It is also possible that cum rccnvered, ifpossible. Sparial plottingof artifacts around was grown nearby and processed on the sile. Pollen features will assistin determining function ofthe feature washes from ground stone and soil samples may inform and in assigning a time frame to the feature. on the potential use of domesticates.

20 CARISBAD IiELIEF ROUTE FIELD METHODS

Excavation methods were lbllowed as outlinedin the and C- 14 sarnplcs were collectedfrom each kature on Data RecoveryPlan (Zamora 1997). Datumsset during the site. the testing phase were reestablished and north-south Scvcral mechanically dug trenches were placedat and east-west baselines were placed over thc sites. Macho Dunes in areas of low artifact densities to These provided the coordinates for a 1-by-1-m grid ensure that no cultural features were present. Several system to be imposed over eachsite. Elcvalions were high dunes were leveledto the sterile red sandin 20 cm taken from the northwest corner of each plotted grid levels in ordcr to expose buried features. using a transit and stadia rod in order to maintain Individual grid excavation forms were filled out horizontal control. All surface collections and for every Icvcl excavated as was a fcature form for cxcavation units were within the grid systcm. IIand each fcaturc, detailing depth, typeofarlifact recovered, tools such as shovcls, trowels, picks, and brushcs were and soil type and color basedon the Munscll scale. All used in the excavations. At LA 29363, mechanical stratigraphic levels and cross sections were drawn cquipmenl wasused to remove overburden fromlarge along with aplan view for each fcature. Photographs dunes to assist in uncovering buried culturalfeatures. were taken both bcforc and after excavation. Whenall The dunes were removedin 20 cm incrcmcnts in order the cxcavation work was complcled, the sites werc to minimize damage lo the underlying cultural mapped using atotal station. resources. Thermal features arc featuresthat could not be We did not employ any mechanical workon LA absolutely identified as either a hearth or roasting pit, 29362 because the site was very shallow with no more but exhibit characteristics such as oxidation, burned than 5 cm of depth in some arcasand 15 cm inothers. rock in and around the fcature, and charcoalin thc li11. Most of the depth came fromaround bushes, wherethe The lill containedan abundant amountof charcoal with soil had accunlulated. thc sides of the pits exhibiting oxidationfrom burning. Excavation units were placed in areas of known Storage pits wcre identified by the absenceofburning features or surface artifact concentrations and then or charcoal witha mottlcd fill and ranged from smallto extended as features were encountered.All were dugin large in sizc. Potrests were identified as shallow pits 10 cm lcvcls except where stratigraphic breaks wcrc present on thc floorofthe structures, usually containing evident. '[he features wcrc dividedinto halves, a soil ceramic fragments on the bottom of pit. the Postholes profilc was drawn, and then the rest of thc fill was containcd wood fragmentsin the fill and were smallin removed in stratigraphic levels. Areas around the size. features were stripped down toutilized a soil surface All artifacts and sarnplcs collectedwill be curated where possible, enabling us to find othcr cultural at thc Archaeological Research Collections(AKC) at feahlres. the Laboratory of Anthropology, Museum of New All soil from the excavations was passed through Mcxico. Fieldmaps and notes, analytical data shccts, an %inch wirc mcsh, and all of the artifacts wcrc and photographswill be deposited at the Archeological bagged by level. Floor artifacts limnd at LA 29363 Records Managemcnt Section (ARMS) of the Slate were mapped and bagged separately.Flotation, pollen, Historic Preservation Division. SITE SUMMARY-TROJAN HILL, LA 29362

Trojan Hill covers 25,935 sqm (399-by-65 m). It is an discoloration, and a trowel revealed thcrc was very extensive lithic procurement area with three discrete little depth (4 cm). No charcoal was noted. artifact clusters (Fig. 4). Thc North Loop (County Road 604) road has previously removeda portion of the site Artifacts to thc north. Three dense artifact scatters (Areas A through C) wcrc excavated for atotal of 630 sq m. The artifacts from ‘l’rojanHill consist oflithic artifacts Area A was the largest of the scatters and 456 sq m except for four ceranlic fragments. ‘l‘hey total 12,169 were dug. In Areas I3 and C 100 sy m were excavated artifacts. (her half of the artifacts are core llakes in each unit.Thc artifacts are scattered throughout the (‘lable 1). Thereis an abundant arnount ofnatural chcrt sitc; however, the densest concentration wasin Area A and quartzite found in the gravels above bedrock. (Figs. 5-6). There wereno features present: however, These materials werevisible in the road cut. two hcarths were excavated67 m beyond thcsite to the east. A C:- 14 sample was collected from both hearths Site Interprctation and produced calibrated datesof A.1). 1195 for Hearth I and A.D. 1025 for Hearth2. ‘I’he wood used in thc It is clear that the site was a lihic procurcment areaby hearths was identified as juniper. Hecause of the the amount of large cores and debitage found onthe distance separating the two, it is not known if these surface. Many concentrations found onthe site contain hearths are rclated to the site occupation. debitagc that can berefitted onto the core. The two most One possible deflated hearth was noted, butwell common matcrial types werechert and quartzite. ‘J’hc away from the scopeof work to thc south on the hill larger cores werethe quartzite probably becausethe raw slope, outside of the right-of-way.It consists of fire- material is more abundant andlarger in six. cracked rock (FCR) with few lithic artifacts ‘Thereare no datesto place the site within a spccilic surrounding an area of approximately -m1 sq. a cultural period. One biface foundon the site is broken A total of797 grid units were hand excavated with and not identifiable. The presence of four ceramics an average depth 01‘3.75 cm. There was29.8 cubic m would suggest that itmay have been occupied duringthe of soil removed and screened. There was no Formative periodbut this is only speculation sincethe mechanical means of soil removal (Figs.7-8). ceramics consisted of four Playas Incised sherds (ca. A.D. 1200)from the same vessel. Setting The two hcarths found67 n1 east of the site date to the Formativeperiod. Hearth 1 dates to the Late Globe or ‘l’hc site is located on top ofa small ridgethat contains Early Phenix phases, andHcarth 2 to the Late Globe mesquitc, creosote, and variousgrasses. They had been phase. Withso much space betweenthe hearths andthe mechanically retnovcd partially to the north by the main portion of the site, it is possible that these hearths original construction of thc North Looproad. A light are from small campsites notrelated tothe site. However, scatter of artifacts is present on the other sidcof lhe it could be that the inhabitants using these hearths also road; however, thereis absolutely no depth. Surface uscd the site to procure someof their lithic material. artifacts are all that is present on the north edge of the While the raw nlaterials were probably cxploited site. duoughout R long time span, there is no evidence to To the south, outside the right-of-way, thescatter confirm use by a particular group. The absenceof tools is not as heavy as on top of the ridge; howevcr, a or tool-shaping flakes would suggcst that theprimary use dcllated hearthis present. A scatter of FCR measuring of the site was for rnatcrial selection and procurement. 54-by-60 cm makes up this feature. Therc is no soil I 1

!4 CARLSBAD KlSl,IEF ROlJTE

I

c

'

+ 28 SITE SUMMARY-MACHO DUNES, LA 29363

Macho Dunes is probably a seasonal base camp Cultural Units because ofthe several brushstructural remains present. It covers almost 48,750sq m (524,757sq ft). The site ArrLr A contains the remains of 5 brush structures and 35 thermal and storage features (Fig. 9), dating primarily Area A was thefirst unit to be excavated.It is located to the Globe phase(AD. 752 to AD. 800), although on top of thc knoll where bedrock outcrops near the fcatures dale from the Middle Archaicto the Late surfacc. One hearth was foundon the edgeof the road Globe phasc at ca. AD. 980. Preservation of the site cut and wasbuilt into the bedrock. It measured53 cm was very good becausethe features had been sealedby north-south by 35 cm cast-west and was15 cm in depth the sand leaving themin fair to good condition. Most of (Fig. 13).A total of 65 sq m (700sq ft) were excavated the disturbance was caused by root and rodentactivity. around the hcarth in orderto locate other features; However, mechanical disturbance was notable around none, however, was present. Structures 2 and 4. A radiocarbon sample was collected from thc There is evidencc of spalling on the ceramics, feature and produced a calibrated and corrected of date indicating cooking was being pcrformed on thesite. 380 to 5 B.C. with an intercept date of 185 B.C., The extramural thcrmal featuresand the interior hearths suggesting the hearth was used duringthe Late Archaic would probably confirm cookingactivities. The pollen pcriod. sarnplcs were negative; however, the macrohotanical Artifacts. Area A contained 78artifacts of which samples includedgoosefoot, purslane, and carpetweed all were lithic artifacts (Table 2). The most common as subsistence item and mesquitc, one-seed juniper, material type was chert (73.1 percent) and the core Mexican crucillo, and four-wing salt hush for fuel flakcs (71.X percent) dorninatcd the assemblage(‘Table wood. 3).

Setting Arm B

Macho Dunes is located east of Carlsbad along the Area B contained several features (Fig. 14). The North I mop Koad, in Eddy County.It is on the edgeof remains ofthree brush structures were excavated along the Alacran Hills sand dunes. It sits directly south of‘ with nine extramural features.‘l’hc area covers 17 nl (and protected by)a high ridge of sand covered with north-south by I4 mcast-west for atotal of 238 sq m. dense vegetation that prehistorically provided shelter It is located southof Area A whereis it protected fro~n from the prevailing north andnortheas1 blowing winds the wind (Figs. 15-14), Theartifact dcnsity was among (Figs. 10-12).Ofthe almost 48,750 sq m (524,757 sq the highest(n 483) on thesite. This includes surface ft) onthe site, 924sq m were excavated byhand and and subsurface artifacts (Table 4). Descriptions of 1,637 sq m (1 7,h2 1 sq ft) were mechanically removed. artifacts from each structure areas follows:

Table 2. Lithic Artifactsfrom Levels in AreaA Figure 9. Muchu Dunes, LA 29363, site map.

30 CARLSRAD RELIEF ROUTE Table 3. Lithic Artifact MaterialTypes from Area A

Table 4. Artifacts from Arca1%

Ths structure is the largcstofthose round and is located at thc bottom olhe knoll ina flat area. It is oval in shape and has been dug into the sterile redsand andcontained floor l'eatures. Stratigraphy. Three strata were identiliedin the fill of the structure (Fig. 17). Stratum 1, Munsell color 5YR 514 reddish brown, was the modem top soil, a very dry, fine eoliansand. Thls level was8 cm thick.Stratum 2,32 cm thick, was an eolian sand with cultural fill that was mottled, charcoal stained, and contained charcoal and Figure IO. Plun qf hund excuvution und artifacts. The Munsell color was SYR 6/X redd& yellow. mechanical scraping, Arecr A. Heurth I is located in Stratum 3 was the sterile redsand hat underlies the site. north westerncorner. Its thichless ranged from IO to 30 cm and is SYK 6iX reddish yellowon thc Munscll colorchart.

Storuge Pi/s. Five storage pits were excavatcd in Structurc 1 (Figs. 21-22), All the pits were burned and contained charcoal.The iill for each wasa black charcoal- stained sand with oxidization. These pits are listed in Table 5.

‘Tablu 5. Storage PitDimunsions and Dates, Structure I

I I I I Pit 8 41 38 h A.D. 665- OhO

Extramural Featurrs. An cxterior surface was found in Area B that contained several features, of which three were associated with Structure1 (l’ablc 6).Two are storagc pit.. and the other is possiblya posthole. Pit A is 0 20 crn located next to the structure. Although thc lill was - charcoal-stained, it lacked the oxidation that was present L Figure 13. Area A, Hearth I, built into the bedrock. in the other hearths, suggesting thatthe feature mayhave Projilc is north-south. been a storage pit insteadofa hearth. In addition, this area contained so much charcoal stainingin the till of all the features and on hefloors and surtices, that ifa feature did Architectural Details. The remains of the structure not exhibit some type of oxidation we considered it a measured 7.4-by-4.6m by 1 X cm in depth (Figs. I X- 19). storage pit. Pit 7 is identical to Pit 6 (Fig. 23). It is Wnlls. The walls of the feature are curvedand slope approximately 3 m southwcst of Structure 1. ‘l’hcfill again inward. ‘l’heyarc constructed of the natural stcrilcsmd, was charcoal-stained, but oxidation was lacking. which is semicompact. In areas it is very hard. When the sand is wet there is no differencein compactness. ‘I‘ahle 6. Extramural Storage Pits and Postholes Fhr. The Boor surface of thc structure is heavily charcoal-stained and oxidized in some arcas. A scmicompact surface makes upthe floor; however, when exposed to thc sun md wind for a pcriod of time it becomes verysandy. I’hc areas hat are very compact and contain a lot of calcium carbonate are fairly stable and hold together very well. One small posthole was found 60 cm north of the pit There was an areato the east where therc may have structure. No other postholes werefound in this area; bccn an entrance. ‘[he floor kpered out in Grid howcvcr, there was an abundant amount of roo1 and 171N/22SE and the surface disappeared. rodent activity so that they could have beenobliterated. Floor Features. ‘l’here wcrc live floor features Artifacts. A few artifacls were recovcred from present on thcfloor, including a hearth, possible posthole, Structure I, They include9 ceramics, 76lithic artifacts, and three storagepits. and 9 pieces of ground stonefor a total of93 artifacts. Hc~urd1.The hearth measured 48-by-50 ctn with it Most of the artifacts werc from the fill of the structure, depth of24 cm (Fig. 20). The fill was a dark gray, orange, except for one South Pecos Brown sherd foundin Pit 9. Table 7 represents thc ccramics recovered from Structure and black mottledsand with a large amountof charcoal. A calibratcd and corrected C-14 datcof AD. 865 to A.D. I. 985, with an intercept datcof AD. 895, was produced. 185NI 223E 185NI211E 185N1213E 185N/215E 185NI217E 1$5N/ HZ1 E 185N/ 225E r I 1 I

excavation limits - - 183Nl225E

\:\ mechanically 1\>. disturbed area 1r y’ - 181NI 225E

N - 179NI 225E

Structure 2 177NI 225E disturbed road cut

A

‘rs 175NI 225E

-! 0= storage pit = posthole I 0 = ash pit Q = hearth 0 = pot rest ’1 0 3 meters

Figure 14. Area B, sfructural remuins and cxtrmnurcrl features.

34 CARLSBAD KELIEF ROUTE Figwe 1.5. Area R,befiwe cxcavutiorr.

Figure 16. Area B., excavated, Structure 1 is in the lejifiwegrrmnd. 0 Imntnr L .-

Figure 18. Structure 1, Area H, projile.

0 1 mcter SF= storage ptt I.I Fi,gurc 19. Structures 2 and 3, Areu B, projile,

36 C'ARLSKAD WLlEF ROUTE ructure 1 for Stnlchlre 1 are in Table X. Corc flakcs (60.5 percent)

Row 'Iotnl represent more than half of the assemblage. The most common materialtype was chert (73.7 pcrccnt). Just two cores were recovered andboth had flakcs rclnoved from multiple directions. Therc is also a fairalnount ofangular debris, suggesting that tool manufacturing was pcrl'onned.

Table 8. Lithic Artifacts from Structure1

C:eramic cross-dating of the site placcs it into the Globe phwe or the Formative period. The later McKcnLic shcrd wouldsuggest that the sitc was occupied Ten ground stone items wcrc collected from during the later portion of thc phase (A.D. I200 to A.D. Stmcturc 1. Eight of the ground stone pieces fromthe 1300). Howevcr, it is not impossiblethat this ccralnic was structurc were retrieved from the fill. One indeterminate transportcd to the site by someone passing through. The lnctate fragment was rccovered fromPit 9. The ground C:- I4 date fromHeat-thI and this ceramic suggestthat the stone from Structure 1 is listed in Table 9. site was occupicd duringtl1c latter portion of the phase. Sandstone is the most co11m:only used material type 'I'hc lithic artifact assemblage is rcpresented by for the ground stone and is available throughout the area. angular debris, core flakes, and cores. The material typcs The quartzite can be found along thc river beds andin the alluvial dcposits.

Table 9. Ground Stone Artifactsfrom Structure 1

Fine Sandstone A A' I f ' ...... ", ...... unexcavated::::::::: U 0 10cm LJ

Figure 21. Structure 1, Pit 2; cast-west prufilc.

oxidized sand, S cm to X cm. The structure is saucer- shapcd and thewalls flare out in profile insteadofhaving a basin shape. Floor. The floor surfrlcc is a charcoal-stained Stmcttue 2 is the deepcst oral1 the structures on the site. oxidized redcompact sand containing several features. It It is locatedat the basc of the knoll near the road cut. It is was more stablc and not sandy when exposed to the circular in shape and wasburned, with oxidatjon present elements, possibly becausc or its greater depth. throughout (Fig. 24). Several samples for Floor F~utums.Eight lloor features were cxcavated archaeomagnetic dating were collected from the hearth, Structure 2 and are listcd below. but did not produce results. in Hearth. The hearth consistcd of two adjoinillg pits, Stratigraphy. There were only two stratigraphic with one serving as an ash pit. It is located oKccnter of lcvcls in this structure (Fig. 25). Thc Grst level was the the structure and is 30 in diarnctcr and I0 dccp modern top soil that cxhibitcdstaining and the beginning cm cm (Fig. 24). The ash pit is immediately next to the hearth of the oxidation, Munscll lOYR 313 dark brown. The giving an appearance of'a Figure 8. The ash pit is 20 c~n secolld level was thefiJ1 of thc structure and it consisted diameter and I2 cln dccp. The fill consisted ofa black ofred oxidized sand, charcoal staining and an abundant in charcoal-stained line sand. amount of charcoal. Thc Murlscll color was 5YR 514 Skmzge Pitx. Two storage pits werc cxcavated in reddish brown. Two calibrated and corrected dates of Structure 2 (Figs. 27-28). The fill consisted of a black, A.D. 665 to A.D. 960 with an interceptdate 0fA.D. 785 charcoal-stained sand with flecks of charcoal. t iowcvcr, and AD. 665 to A.D. X85 with an intercept of A.D. 770 there was not enough charcoal liom the features Ibr wcrc obtained for the stnlcture. radiocarbon samplcs. All the shallow pits wcrc built into Architectural Details, The structure is2.75 mnorth- the red sterile clay and eachexhibited little oxidation. The south by 3.25 m cast-west and 35 cm deep (Fig. 26). measurc1ncnts for the features arc in Table 10. Wulls. The walls are fairly shallow, semicompact A II'

0 lOcm

I C

Figure 22. Structure 1,Pits 8, 8, and IO; cast-west projiles

Table 0. Storage Pit Dimensions from Structure 2 J. stained sand and it measured X cm north-south by 12 cm cast-west with :i depth of' 6 cm. This was the only 11 (cm) (ern) I L)rpth posthole found in the structureand may have served as a II 3 20 24 0 II support post. l'hc outside surface was also excavated around the f'cature but only onc fcaturc was found. Po/ RE.s/.s.The pot rests wcrc identified as shallow Extramural Feature. One extramural pit was saucer-shaped pits on the floor and base ofwalls (Figs. excavated immediately nortll ofthe structurc (Fig. 33). lt 29-3 I). They arc 2 to 3 cm deep with the cxccption of a was circular in shape and lncasured 25-by-32 cm and 4 Feahlre X, which is large and sits against the wall ofthe cm dccp. The fill was a black, charcoal-stained sand with structllre. l'hc iill fiiom the pot restswds a black, charcoal- oxidization. It is possiblc that this feature may havebeen stained sand withminute charcoal fleclting.No charcoal used for storage; howcver, the oxidization in the pit sarnples were recovered fromthc features. Measurements would suggest some type of thennal activity. lor the features are in 'I'ablc 1 1.

Table II. Pot licst Dimensions for Structure 2 Structure 2 contained 21 lithic artifacts, I4 ccmmics, 23 pieces ofground stone, and one turtle carapace fragment for a total of 59 artifacts. Two corrected calibrated C-14 datus wcrc obtained for Struct-ure 2. They arc from n storage pit (AD. 665 to A.D. 960) and a p~stholc(AD. 665 to A.D. 88s). These dates are associated with the Globe phase and wouldcorrespond with the ceramics. Postldt,. One posthole was excavatcd 40 cm north of the hearth (Fig. 32). The fill was a black, charcoal-

Table 14. Ground Stonc Artifactsfrom Structure 2

Table 15. Ground Stone Foundon the Floorof Structure 2

Row

ltldeterminstu

IUO.O%

100.0%

Stratigraphy. The fill of the structure was Fhw: The floor is also scmicompact and shallow, consisting of cz blackcncd sand with a littlc charcoal-stained. Near the hearth the floor is Inore oxidization especially in the center around thehearth. compact and exhibits oxidation.One floor feature was The feahwe was not founduntil the overburden,30 cm excavated. thick, was removed. Thcrc were stratigraphic no breaks Floor Feutures. A central hearth was excavated in the fill ofthe structure. that measured 36 cm north-south by 32 cm east-wcst Architectural Details. The structurc is 1.65 m and 10 cm deep.The fill consisted of charcoal-staincd north-south by I .h7 M east-west. Thestructure is very sand and charcoal. Theintcrior and rim of the hearth shallow witha depth 01‘8 cn1 to 10 cm. was reddened. An archaeomagnetic sample was taken Wulls. The saucer-shaped structure has gcntly from the rim ofthe hearth, but it did not producea date. sloping sides that are scmicompact and charcoal- A radiocarbon sample produced a calibrated and stained. Where there is heavy charcoal staining, the corrected dates of’ A.D. 865 to A.D. 985 with an walls are compact. intercept at AD. 895. The date placcs the featurc

SITE SUMMARIES”IMACH0 DUMKS 43

ground stone artifacts made OS red sandstone wcrc consist of scmicompactrcddish sand. recovered from the sudace. Seven artifacts were recovered from insidc thc structure. They consistedof B Jornada Brown sherd, El Table 20. Lithic Artifactsfrom Area C Paso Brown shcrd, three pieces of angulardebris, a core flake, and a biface llake. The lithic artifacts were madeof chert and siltstone. No ground stone was presentin the structure.

Two otherthermal pits were foundin the extension. Pit 40 was found28 cm underthe modern sand. It is 34 em north-south by 44 cm east-west and8 em deep (Fig. 40).

'he fill consisted or mottled blackened sand with I'ested C'obble charcoal. Pit 40 myhave served asa hearth. Pit 4 I is a hearth, which was found under a small 40- cm-high sand dune,It is 43 cm north-southby 38 em east-west and 25 cm deep (Fig. 41). The fill was black, charcoal- filled sand. It contained a single flake, which had been burned.A calibrated and corrected c" 14 datc OS AD. 665 to A.D. 88.5 was obtained with a intercept datc of A.I)* 775, placingit during the Middle to I.ate Globe Table 21. Ceramics fromArea C phase.

Area D is located on the east side of the NorthLoop Road (Fig. 42), It covered472 sy 1n (1,548 sq ft), ofwhich 299 sq m was hand excavated. Mechanically rcmovcdfill included dunes over2.0 m high. Below these dunes were cultural features including Structure 5, several storage pits, postholes, hearths, and a rarnada-like unit.

Structure 5 was found undera 2.6-rn-high sand dune. The ovcrburden was removed mechanically.It is a saucer- shaped, irregular oval that is very shallow. Withinthe structure wcrc a rectangular pit and two smallpostholcs. There was no hearthpresent; however, the exterior fire- cracked areaand a small thermal feature would suggest that most of thc activities were performedoutside. There Fhr.Thc floor is a fairly even, reddish brown sand with little staining and no oxidization. is not a C:- 14 date for the structure itself, but anexternal pit, Pit h I, did produce a calibratcd and corrected dateof Floor Foaturtx Thee features were found on the floor of the structurc (Fig. 44). They consist of a A.1). 655 to A.D. 890 with an intercept date of A.D. 770. Stratigraphy. The fill of the ktture consisted ofa rectangular-shapedpit and two small postholes. Thc Pit charcoal-stained, dark brown sand, IOYK 313 on the is 24 cm long, 8 cm wide, and10 cm dccp. The fill was Munscll colorchart. There wereno stratigraphic breaks mottled with black, charcoal-stained sand and minutc within the fill. charcoal flecks. The interior of the pit was very compact Architectural Iktails. Thestructure is I .30 m north- sand, suggesting it had been smoothed. south by 1.29 m east-west and 2 cm deep (Fig. 43). 'I'he two postholes are small, 12 cm in diameter and Wnl1.s. The sides of the structure are 2 cm high and 5 CM deep. The interiors ofthesc postholcs were also

48 CAKLSBAD REI,IEF ROU'L'E

210N1246E-

200N1246E-

190N/24bE -

18ONIZ46E -

\ \ \ \ \ \ \ \ \ \ \ \ ! \ \ \ \ \ \ 175N1246E - \ \ \ \ \ \ \ \ \ \ \ H= hearth \ \ S= storage plt \ \ P= posthole \ \ \ \ \ \ 170N/246E \ \ - \ \ \ \ 0 10 meters ,avatlon ension

Figure 42. Area D, showinghand excavations nnd mechunicul scraping limits.

CARISBAD RELIEF ROU'I'E

A I

I A' 0 50 cm L..L a

A A' I I1

50 cm h ~. I Figure 48. Area D, Hearth 51, plan and east-west Figure 49. Aren E, Hearth, plan and north-south profile. projile.

Table 24. Ceramic Artifacts from Area D bedrock (Fig. 49). In most cases, bedrock was reached throughcxcavation exceptin areas aroundthe vcgclation. Cells: Count Level Row Tutu1 Row Onc grid unit was taken down to bedrock (23 cm) and reached sterile red sand. Although there was some depth

Fill in this area, approximately 20 cm, the hearth wasvisiblc thc surface. Fire-cracked rock was scatccred in a 1-m Jonnda Blown 4 4 on 1OO.OY" IOO.O'X, area andthe soil was charcoal-stained. 13.3% 12.9% The hearth measured44 cln north-south by60 cm El Paso Brown 22 1 23 95.7"h 4.3% 100.0% cast-west and12 cm decp (Fig. 49). The fill ofthe feature 73.3% 100.0% 14.2% was heavily stained sandwith bits of oxidized sand.The South Pecos 2 2 bedrock undcr the fill was also burned, exhibiting Hmwn 100.0% 1 OO.OO/u oxidizationand crazing. Enough charcoal wasprcscntfor a C- 14 sample. A calibrated and comcctcd date 0fA.D. X00 to AD. 1300 with an intercept at A.D. 975 was produced for the hearth. This placesthc feature into the later portion ofthe Middle Globe phaseand the bcginning of the Late Globc phase. There were no artifacts associated with this feature.

Areu E Artqucts

Area E is located to the east across the North LoopRoad Although noartifacts werc found in the hearth, the area on the top ofthe knoll. A total of 66 sq m (7 10 sq ft) were surface-strippcd around a hearth built on top of the excavated around thc feature produced129 lithic artilhcts, six ceramics, and four ground stoneartifacts. The chipped stone artifacts consisted of mostly corcflakes (Table 25).

Table 25. Lithic Artifactsfrom Area E

mechnnlcnlly excavated area - ...... -. - ...... ,...... - ...... - .. . .

Figure 50. Areu F,plan view of excavations. The ceramics rccovcrcd consistedof two typcs, El Paso Brown (n= I) and South PccosBrown (n = 5). The ground stonc fragments consistcdof two basin metates, Table 26. Thermal Feature Measurements one slab mctate, and an indeterminant fragment. Their functions were determined bythc shape of the ground surticc. The basin metateshad concave grinding surfaces with circular striations and thc slab lnctate had a flat grinding surface.

Area F was undcr a 2.0-m-high sand dunc,ofwhich half was removed byhand. It covers an area of 577 sq m (6,2 I 1 sq ft), of which 233 sq m (2,508 sq ft) wcrc hand excavated and 354 sq m (3,X I0 sq ft) werc rcmovcd A roasting pit (31) was also cxcavatcd and was mechanically (Fig.50). It contained hearths, storagc pits, located east from the mahconccntration ofpits (Fig. 58). and a roasting pit. Undcr thc sand dunes,an area with 15 It was under a mesquite bush causing thc firc-crackcd featlrrcs was excavated (Figs. 51-52). These features rock to he scattered. The pit was a vcry shallowirrcgular included hcarths, postholes, storagcpits, and a roasting oval with charcoal-stained sand. Limestone lined the pit. The pits had been duginto thc rcddish sterile compact bottom of thc pit, which was86 cm long byX 1 cm wide sand. and 9 cm dccp (Figs. 59-60). The limestone rocks Thrnnal Features. The thermal features were exhibited blackcning andcharacteristic burning suchas identificd by the amount of charcoal foundin the fill of brealung andcrazing. A light scatzcrof fire-cracked rock the pit (Table 26). Oxidization was another factor that was also present outsidcthc fcaturc. Charcoal was found was considered. Thefill ofthc thennal features consistcd throughout the fcaturc and produceda date of 2460 R.C. of charcoal-stained sand with charcoaland charcoal to 21 65 B.C. with an intcrccpt datc or 2280 B.C., flecking (Figs. 53-57). Burned rockswere usuallyfound suggesting use during the Middle Archaicpcriod. in the soil and cxhibited reddening.C-14 samples were StomgrPits. Several pits that didnot exhibit burning collected from cachof the thermal fcatures along with were idcntificdas storage pits (Figs. 6 1-66). The fill from flotation samples. Pit 29 wasrock-lincd with limestone at thcsc pits was a reddish brown, mottled sandthat did not the bottom of thc pit. The charcoalflccks present in the contain any charcoal (Tablc27). GI1 were too small andtoo few tocollect.

54 CARLSBAD RELIEF ROUTE

A I

I A'

A A' I fire-cracked rock ,

I 20 cm

Figure 53. Area F, Thermal Feature20, plun and projile. IF ,.,

Thermal Feature 26

...... unexcavatedi::::::::::::::::::::::::jj:::::~~:~:~:~:~ 0 30 cn 1 u Figure 55. Areu F, Thermal Feature26,plun and profile*

lable 27. Area F Storage Pit Measurements

Postholrs. Two possible postholes werc cxcavatcd. Thcy were small, shallow, round fcalures (Figs. 67-69). One (28) was 20 cln in diameter and5 cm dccpand thc other (30) was 1.0 cm in diamcter and 5 cm deep. Thefill was a charcoal-stained sand withcharcoal flecking. No artifacts were found in the fill. Thc interiors of each were 0 cm 30 cornpactcd sand.No rodent activity was noted in either of 1 thc fcalures. Figure- 54. Area F, Thernznl Feature 23, plun and north-south profile.

56 CARLSBAD RELIEF ROUTE D I I E utilized E' Dl L I Thermal surface I

10 cm 0 20 cm 0 u Figure 57. AreaF, Tliernzal Feature 29, plun view Figure 56. Area F, Thermal Feature 27, north- and east-west profile. south profile. Several features did contain artifacts includingchipped Art4facts stone and ceramics.The lithic artifacts totalcd 1X items (Table 29). A total of 298 artifacts were recovcrcd from Area F. Thc ccramics from the features place them during This includes the surface, subsurfacc, and fcahlre the Formative pcriod (Table30). The C-14 datcs rangc artifacts (Tablc 28). fiom AD. 655 to A.D. 1035, wllich suggests that the fcaturcs wcrc used duringthe Globe phase. Table 28. Artifacts from Area F

Figure 58. I

SITE SUMMARIES-MACHO DUNES 57

Table 30. Ceramics from Features in Area F

The threepicccs of*ground stonewerc round in the associated with Thcnnal Fcature34 and thc indeterminate general fill. Two ofthe three were associatcd withPits 32 ground fragmcnt was recovered from aroundStoragc Pit and 34. They consist of an indctcrminate fragment,onc- 32. hand mano, and metate fragment. The metate fragmcntis 62 CARLSBAD RELIEF ROUTE CERAMIC ARTIFACT ASSEMBLAGE

C. Dean Wilson

A total of 291 sherds were recovered duringthe data 01' milky white to clear gray grains probably rccovery phase of the Carlsbad Rclicf Project. This representing feldspar along with some quartz. Dark includes 287 sherds froin LA 29363 and 4 sherds from fragments rcpresenting hornblende are often absent, LA 29362 (Table 31). The analysis of these sherds and when present they consistof vcry small and sparse involved thc recording of both descriptive attributes fragments associated with these other mineralparticles. and typological categories. In order to compare trends Fragments noted in sherds from thissite assigned to noted at these sites to those in surrounding areas, this temper appear tobe rclatively small, and the large analysis strategies and categories previously discussed grains notedin sherds assigncdto this typc fromthe El in other studieswere used here (Hill 1996a; Jelinck Paso area tendto be absent. 1967; Kelley 1984; Mera 1943; Runyon and Hedrick Another temper group was distinguished by 1987; Wiseman1996; 1997). numerous vcry small and profuse clear to dark fragments, andis refcrred to hereas crystalline igneous Descriptive Attributes rock. Larger grains are somctimes present andare usually roundish and crystalline in structure. These A rangc of ceramic descriptiveattributes of potential fragmcnts appear to be crystalline or sugary in use in the dating sites of and components,as well as the appearance. This group may represent the use of Capitan aplites. Pctrographic analysis indicates a dctcrnlination of the area of origin and useOS ceramic vesscls reprccscnted, were recorded. Ceramicattributes granite aplite (Hill, this volume). recorded during thisanalysis include temper, surface Thc other temper group is rcprcsented by the manipulation, sherd thickness, and vesselIbrm. dominancc of gray tkldspr lkagments presumably from syenites from somewhere in the Sierra Blanca Tcmper Cutepries (Wiseman 1998).Feldspar fragrncnts tendto be sirnilar in appcarance and are angular and sparsely scattered. Tempcr was assigned to basic categories by examining Smaller grains of other mineralsare rare if present at freshly brokcn sherd surfaces through a binocular all. Feldspar fragments are large compared to other microscope. Igneous tempcrs were dividedinto several temper fragments and are oftcn readily visible even without the aid of a binocular microscope, and protnlde groups basedon color, size, and crystalline structure. More detailed characterizationsoftemper areprovided into the shcrd surfaces. Thesc fragments tendto be through petrographic analysisof seven sherds assigned opaque and grayto off-white in color. Smaller grainsof lo various temper categories (see Hill, this volume). other mineralsarc rareif prcsent. Almost all of the tempersidentified represent crushed Sand refers to the presence of rounded or igneous rock. subroundcd, white to translucent well-sorted medium The most common group is represented by a to coarse quartz grains. Small angular fragments leucocmtic igneous rock containingboth feldspar and somctimes occur along with thcsc grains indicating the quartz that may reprcscnt crushed granites or use of crushed sandstone or sand weathered from monzonites. This group is identilicd by the dominancc nearby sandstone.

Table 31. Ccrarnic l'ypes from Carlsbad Relief Sites

Pottery Types LA 29363 LA 29362

Count I'emcnl Count I'crccnt

Plain Krown Body 202 91.3

Plain Hmwn Rim 14 4.9

'I'hin Drown Warc 11 3.8

Playas Inciscd 4 100.0 Total 287 4

CERAMIC ARTIFACTASSEMBLACE 63 Surfice Manipulation could have originated fromthc source (Shcpard1965). The colorof each sample was recorded usinga Munsell Surface manipulations reflecting the presence and types Soil Chart. of surface textures and polishing were recorded for both interior and cxterior sherd surfaces. Plain Ceramic Types unpolished refers to surfaces wherecoil junctures have been completely smoothed, but surfaces are Ceramic types reprcsent groupings that relay unpolished. Polished surfacesare those that have been information about combinationsoftraits with tcmporal, intentionally polished after smoothing. Polishing spatial, and runctional significance. All of the 287 implies intentional smoothmg witha polishing stoncto shcrds recovered fromLA 29363 reprcsent plain brown produce a compact and lustrous surface. Slightly warc utility warc types, while the four sherds fromLA polished rcfcrs to a plain, slightly compacted or 29362 belongto a single rcd ware vessel. polished surfacc. Heavily polishedrefers to a lustrous compact surface that has obviously been heavily polishcd. Unpolished striatcdrefers to the presenceof a series oflong shallow parallel groovcs resulting from The prcscnce of brown ware shcrds, derivcd froma brushing with a fibroustool on an unpolished surl'we. relatively large number of vcsscls at LA 29363, Surface worn or missing rcfer to cases where a provided a rare opportunity to examine ceramic distinctive surface cannot be recognized due to the patterns associated withan occupation dating before absence or disturbance ofthat surface. the introduction of textured and painted ceramic types into the Pccos Valley. Plainbrown wares appear to have been produced in some areas of the Jornada Mogollon in southeasternNcw Mexico as earlyas A.D. Sherd-based vessel form categories reflect shape or 200 and representthe dominant ceramicsuntil at least portion ofthe vessel from whicha sherd was derived. the middle thirtccnth century (Hill 1996a, 1996b). Categories identitkd were based on rim shapeor thc Unfortunately, it is difficult if not impossible to presencc and location of polish and painted distinguish Jornada region brown wares produced decorations. Whileit is often easyto identify the basic throughout this long span. Thus, the placementof the form (bowl versus jar) of body sherds from many ceramic asscmblage atLA 29363 earlyin the Jornada Southwcstcrn regionsby the prcscnce and location of Mogollon ceramic sequence is based on the lack of polishing or painted decorations, suchdistinctions are associated textured or painted types rathcr than the not as easy for Jornada Brown Ware typcs. For characteristics of plain brown ware shcrds recovered examplc, Jornada Brown Ware bowl jar and shcrds can from this site. be polished or smoothed on either side. Therefore, Brown wares from various areasof the Jornada body sherds exhibiting equal amountsof polishing on region including the Carlsbad area have long been both sides were simply assignedto an Indeterminate divided into a series oftypes based on combinations of body, polished on both sides category. Shurds with attributes thought to be spatially sensitive. Distinctions heavier polish or painted decoration on interior of various typcs are based on differencesin surface surfaces wereclassificd as bowlbody. Thoseexhibiting color, polish, and temper postulated for plain brown a higher polish or painted decoration on the exterior warc ceramics from different areas of the Jornada surface were assigned ato jar body category. Cooking- Mogollon region (Jelinek 1967; Jermings 1940; Lehmer storagc jar neck sherds wercidentified by thc presence 1948; Whalcn 1994; Wiseman 199(ia, 1998). of distinct saddle-shaped curvescharacteristic of necks Unfortunately, recent studies indicate a great dealof belonging to this form. Cooking-storage jar rim sherds overlap in the attributes of brown utility ware also exhibit thedistinct shape ol'a necked jar. Row1 rim tempering materials fromdiffereut arcas of the region rcfers to sherds exhibiting inward rim curvature (Hill 1996a, 1996b; Whalcn 1994). These examinations indicativc of bowls. Seed jar rim refers to sherds indicate strong sirnilaritics in both pastes and derived from spherical vessels that do not exhibit manipulations of brown ware pottery found in both thc distinct necks, but have rounded openings nearthe top. riverine and mountainous areas of the Jornada Clips from shcrds were also fired to controlled Mogollon. Even petrographic analysis involving oxidation conditionsat a temperature of 950 degrees C detailed characterizations of pastes and temper from in order to standardizc ceramicpastes. This provides distinct arcas of the region could oftcn not distinguish for a comnlon cornparison of pastes based011 the pottery originating in distinct areas of the Jornada influcncc of mineral impurities (particularly iron) on Mogollon country (IIi111996a). Therefore, many rcccnt paste color, and maybe used to identify shcrds that

64 CARLSRAD KLiLlEF ROUTE studies have simply lumped plain brown ware shcrds Still, the use of different Jornada Brown Warc type prcviously assigned to types such as El Paso Brown, categories rnay provide an opportunity to monitor Jornada Brown, or South Pecos Browninto a single variability in assemblages rrom sitesin diffcrent areas plain brown waretype, and have attemptcdto exalninc that may bc of spatial or temporal significance, and variation through the independent cxamination of may be cumbersome to monitorby distributions of various pastc and tcchnologicalattributes (Hill 1996a, attributes alone. 1996b; Whalcn 1994). During the present study, different classification Others have suggested that subdivisions within strategies were employed during various phasesof the Jornada Brown Warc pottery are in fact useful, analysis. All but the four Playas Red Incised sherds although factors othcr thanspatial variation may also from LA 29362 were assignedto Jornada Plain Brown bc reflected. In reccnt studies, Wiseman(1996a, 1998) Warc type categories group(Tablc 3 1). All four shcrds continues to usc modified versionsofbrown warc types from LA 29362 were classilied as Playas Incised and described by Jelinek (1967). In this typology Jornada were fromthc same vesscl. These shcrds exhibitedfinc Brown, also recently referred to as Sierra Blanca inciscd decoration 011 a very polished surfacc. While variety (Wiseman 1998), is described as gcnerally Playas Inciscd was described for sherds fromCasas having well-polished surfaces that obscure temper Grandcs in northern Mcxico (DiPcsoct al. 1974), a grains. Temper fragments are described as small, varicty ofthis typc also appears to have been produced consisting ofa profusion of small equally sizedgrains. in the Sierra Blanca region (Wiseman1981). Jornada Brown Ware vcsscls are usually thick (6 to 8 All of' the 287 shcrds from LA 29363 represent mm) and vcssel shapes tendto be regular. variations of plain brown ware from the Jornada Another category somctimes recognizcdis South region. Three categoricsofJornada PlainBrownWarcs Pecos Brown. This typc isgenerallywell smoothed, were recognized based on the prescncc ofa rim and and polishing ]nay bc strong to absent. Ternpcr is thcir thickness (Table 31). Shcrds exhibiting plain represented by sparse large gray feldspar fragments, polished surfaccs were assignedto either a plain brown which Frequcntly show through the surface, and body or plain brown rim catcgory. A few very thin appears to indicate syenite from theSierra Blancas. sherds, lcss than 4 mm in thickness and similar to those This tcmpcr results in blocky to tabular pastc cross derived from ElPaso Polychrome, were placed intoa sections. The protruding tcmper cracks arc often thin brown ware category. surrounded by very small radial cracks and surface Plain brown ware shcrds assigncd to these clays that have contractcd or shrunk toward thc body categories wcre thenplaced into a previously defined clay. Vessel shapesalso tcnd to be moreirrcgular than regionally specific grouping. A sample ofthe sherds Jornada Brown. A few sherds exhibitingthe overall classified as plain brown ware wasfirst separated by characteristics intcrnlcdiate between South Pecos Keggc Wiseman. Bascdon information regardingsuch Brown and Jornada Brown were assigned to a divisions andmy examination ofshcrds that Wiscman JornadaISouth Pccos Brown catcgory. These appcarto assigned to various types, I assigned the rclnaining be similar to sherds described as South Pecos-like brown warc sherds to various categories (Table32). Brown by Wiseman (1996a). Variation in paste and temper rcsulted in thc El Paso Brown is mainly distinguished from recognition of a number of diffcrcnt brown ware Jornada Brown by the profusion of large fragmnts groups. Definite ovcrlapis reprcscnted; in some cases including rounded quartz fragments, representing shcrds assigned to different vessels could havc been granite temper, which may protrude through the produced in thc same regions and evenby thc same surface. ElPaso Brown shcrds also tendto be soft and potters. Recording thesc categories provides havc less evidence of polish or lustcr and more information relating to the nature of brown warc scraping markson the interior surfacc. Pastes tendto bc variation at sites excavated duringthc Carlsbad Rclicf dark or brown with a dark core. Project. Additional informationrelating to plain brown McKcnzie Brown is charactcrized by abundant warcs fromthis project are indicatcd by distributionsof crushed quartz temper. Pastcsare tan to black witha attributes recorded (Tables33-37). These distributions black core. Walls tcnd to be thin, and surfaces are are consistcnt withattributus described for the various srnoothcd withsome polishing. categories, as the dominance of darkhigh iron paste, Most problcms in the use of these brown warc small igneous temper, well-polished surfaces, and widc types stem fiom various mixes in attributes used to vesscl walls are all typical of Jornada Brown. define diflbrcnt plain brown warctypes. For example, Distributions ofvarious attributcs noted in plain brown the presence of a temper category uscdto define one warcs are discusscd in more detailin later. typc may occur along with surface manipulation commonly uscd to define another (WisemanI996a).

CERAMIC ARTIFACT ASSEMBLAGE 65 Dating of Sites not be associatcd with quarry activity. The presence of this vessel, howcver, indicates somc presence of The four sherds from the single Playas Incised vessel Jornada Mogollon groups during the thirtccnth and at LA 29362 indicates a pot drop, which may or may fburteenth centuries.

Table 32. LA 29363 Temper by Brown Ware Group

Cclls: Count I Paste Row Total Kow Column Dark gray to Dark gray to blaPk Brown and rcddish Oxidized, red-br'uwn outside, Ked and black uxidizcd cross scctiun thrnugllout dark gray-black fnsidc gray streaks Jomada Drown 10 17 2 2 20 SI 19.6% 13.3% 3.9% 3.9% 3Y.2'% I00.0% 30.3% 9.0% 17.8%) El Pnso Brown 12 70 104 I I .5Y" 61.3% 1 OO.O%, 36.4% 37.0% 36.2Yt SOUlll I'ocos 9 80 I04 Drown 8.7Y" 76.9% 100.U% 27.3% 42.3% 36.2% I 6 II 9.1% 54.5% 100.0% 3.0% 3.2% 3.8% h b I0U.U'YU 100.0% 2.1 Y, 11 IOO.OX, 3.071 5.3K 3.8% Column 'L'otal 33 1 89 287 I 1.8% 65.9% 100.0% 100.0'26 \

Table 33. LA 29363, Paste Cross Sectionby Brown Ware Group

I'ecos Brown

66 CARLSBAD RELIEF ROUTE % I 'Ob Table 36. LA 29863 Exterior Manipulation by Brown Warc Group

Data from LA 29363 provide important IJnforhtnately, all of these carly ceramic phases are inibnnation relating to thc characteristics of pottery poorly defined. Thisrcsults from thc gencral absence of froin the early Jornada Mogollon occupations, dating shldics reporting dated assemblages with relatively prior to the introductionof textured, paintcd, or slipped largc numbers of pottery without later textured or typcs. An important issuc is whcthcr the sample of decorated types. Regional sequences dehcd for sherds fromLA 29363 is large enough to dctcrmincif southeast New Mexicothat may include occupations the abscncc of later decorated types is not simply thc with early brown warc componcnts includethe Sierra rcsult 01' chance occurrence from thc brcaking of a Rlanca (Kelley 1984), the Middle Pecos (Jelenck small number of vessels in an area whcre decorated 1967), the eastern extension (Corlcy 1965; Lcslic vessels wcrc always rare. Despite the srnall sample 1979), thc castcm Guadalupes (Applegarth1976; Katz sizc, shcrds represented are derived from a largc and Katz I979), andthe Carlsbad Basin (Katz and Katz number ofvcsscls as indicated by their very scattered 19%). contexts of rccovcry (Zamora,this volume) and wide Other sites in areas of Texas tothc south andcast rangc of surface and pastecharacteristics. Thus, the of southeast New Mexico containing similar plain absence of' dccorated sherds at this site appears to brown wares have also been documented. Along the rcflcct an occupation during the time prior to thc Pecos Riverjust to the south ofthc Ncw Mcxico bordcr is thc northcrn scctor of. thc Eastern Trans-Pccos as appearance of'dccorated and textured wares sitesat in defined by Mallouf (1985). These include sites with southcastem New Mexico. Such occurrences of brown Jornada Mogollon ccramics that appcar to rcflcct ware shcrds from this large a number of vesscls occupations closely related to those along the Pecos without any associated texturedor decorated seemsto River in southeastern New Mexico. Brown ware sherds be unique in this rcgion,as most sites assigned to early have also been noted atsitcs along the Pccos River in ceramic phases yielded smaller ceramic samples or Pecos and Crockett counties thein northeastern Trans- contained later contaminants and Katz 1993). (Katz Pecos countryof Texas (Rogers 1972). Someof these Excavated features at LA 29363 included five sherds were assigned to Caddoan types, others to structural remains anda number of hearthsand smallcr Jonlada Mogollontypes, and a fcw wcrc intcrprctcdor surlhce features (Zarnora, this volurnc). described as local manufacture of either Caddoan or Several phase scqucnccs havebeen defined for the Jornada ini-lucnce (Rogers 1972). A fcw sitcs to the prehistory of southeastern Mexicothat include phascs cast of the Ncw Mcxico bordcr assignedto the nco- or occupations containing plain brown waresas thc American phase as defined for the Llano Estacada sole ceramic type for various rcgional sequences region of Texas have also yielded low amounts of (Scbastian and Larralde 1989; Katz andKatz 1993). Jornada Brown Ware pottery and evidenceof semi- hX CARLSBAD RELIEF ROUTE pennancnt stnlcturcs (Collins 1968). While such until after A.D. 900. More recent investigations have occupations in Texas have bccn assigned to other rcsultcd in considerable cvidcnce indicating pottely regional schemes, they appcar to bc most closely was much earlierin this rcgion (Anyon 1985; Hatchoct related to occupations in New Mexico assignedto the al. 1985; O'Laughlin 1980; Hard 19X3b; Miller and eastern cxtcnsion ofthe Jornada Mogollon. Stuart 199 1; Moore 1996; Stuart 1990; Whalen 1980, Katz and Katz (1993) proposca broad chronology 1981, 1994a). It appears thcn that pottery may havc for southeastern New Mcxico that crosscuts various been introducedinto this arca as earlyas AD. 200, and rcgional sequences and includes scvcn pottery-yielding was well established shortlyaftcr. periods. Pottcry-yiclding occupations begin withthe Similar carly brown ware occupations often Formativc 1, which is tentatively postulated as dating consisting of shallow pithouse villagcs appcar to between A.D. 500 and 750, and is solely separated underlie moretypical ccramic componentsin almost all from Late Archaic occupations by the additional Southwestern rcgions including the Mogollon, Anasazi, occurrence of rclativcly low frcquenciesof brown warc and Hohokam spanningfrom at least A.D. 200 to about pottery. Types postulated to occur at components 550 (Reed et al. 1998; Stark 1995; Whittlsey et al. dating to this phase include Middlc Pccos Brown, 1994; Wilson andRlin~nan 1991; Wilson et al.1996), South Pecos Brown, Jornada Brown, andAlma Plain. and probably rcprcsents a pan-regional adaptive Local phases includedlatc Ihcco for the southeastern strategy and material complex. The extremely extension (Lcslic 1979), early 18-Milefor thc Middle conservativc and long-lived naturcof locally produced Pecos area north of Roswcll (Jclinck 1967), undefincd El Paso Brown and pithouse architecturc in thc early ccramic pcriod of theSierra Hlanca (Scbastian southern Jornada Mogollon region has rcsultedin the and Larraldc 1989), andthe Early Globe definedat the assignment of' an extremely long span to the Mesilla Brantley Reservoir near Carlsbad (Katz and Katz phase. This includesa bcginning date at A.D. 200 and 198Sb). Formativc 2is datcd between A.D. 700 and an ending datc somc time duringthe twclfth ccntury, 900 (Katz and Katz 1993) for which a reduction in after which intrusive, dccorated pottery types andEl activity during the phase or a more centralized Paso Polychrome arc common. Some studius havc settlement pattcrn wasidentified based onthe existing suggested changesin the range ~Tcharacteristicsof El suite of radiocarbon dates (Katz and Katz 1993). The Paso Brown produced duringthis long span (Whalen first sedentary villages of pithousc and surface rooms 1994b). appcar in thc Middle Pecos atthis time (Jclinck 1967). Another ceramic marker forthc later part of the Ceramic assemblages were dominatedby Middlc Pecos Mesilla phasc is the occurrencc of intrusive Mimbres Micaccous Brown but werealso described as having whitc warc types (Aten 1972; Mcra 1943; Lehmer Red Mesa Black-on-white inthe north and Ccbollcta 1948; Whalcn 1980, 198 1, 1994a). Mcsilla phase Rlack-on-whitc in the south (Katz and Katz 1992). occupations datingto the tenth ccnlury oftcn have low These white wares types are not prcsent at sites in frequencies of MimbrcsBoldface Hlack-on-whitc while southeastern New Mexico until after A.D. 900 those datingto lhc clcventh and twelfth centuricsmay (Pcckham 1990). Other infonnation rclating to the havc some late Milnbrcs Classic Black-on-whitc. A timing of thc introductionof white warcs is discussed similar markeris represented at sites in morenorthern later in this chapter. regions of thc Jornada Mogollon whcre Red Mesa Given the lack of wcll-datcd sites in southeastern Black-on-whitc occurs alongside plain brown wares at Ncw Mexico yielding distinct brown ware ceramic sitcs dating to the tenth ccntury (Jelinek 1967; Lcvine asscnlblages, much of the datingof carly brown ware 1997; Wiseman 1998). occupations in ths area is based on comparisons with Somcwhat conflicting dating evidencerelating to better-dated sitcsof the southern Jornada Mogollonto the beginning dates for early Jornada brown wares the west where similarceramics havc been recovered, rcsults from radiocarbon dates from various arcas. as well as dated sites in other regions yiclding these Thcsc include a date in thc A.D. 200s at contexts ceramic types. Thc great majority of the wcll-dcfincd yielding radiocarbon datesat Dcadmans Shelter in the and dated early brown ware occupations of the Jornada Texas Panhandlc (Tlughcs and Willey 1978). Thcse Mogollon arc associated withthe Mesilla phasc dcfined contcxts yiclded29 sherds classiticd as JornadaBrown for the southern Jornada area along the Rio Grande Ware by Helene Warren that were interpretcd as trade Vallcy in south-central New Mexico andthc wcstern warcs from the Jomada Mogollon, and appear to Trans-Pecos of Texas (Lchmcr 1948; Whalen 1994). indicate widesprcad tradeas well as rcprcscnting oneof The carly beginning date for these occupationswas not the carlicst dates for Jornada Brown Ware. Thc recognizcd by Lehmer(I 948) in his original dcfinition possibility of a somewhat latcr bcginning date for of Jornada Mogollon during whichhe dcscribcd brown Jornada Brown Wares at lcast in some areas of thc warc pottcry associated with pithouse sites as not Jornada may bc reflectedby the absence of ceramics at having appeared inthe southern Jornada Mogollon area contexts dating betwccn AD. 450 and 500 at the Sunsct Archaic site located west of Roswell in the Paso Polychrome. It is likely that thcsc should not be eastern foothills of the Sicrra Blancas (Wiseman considered as El Paso Polychrome, but they belong to 1996a). At the Sunsct sitc ceramics are absent jn the range of sherd thickness for early ElPaso Brown. conlcxts yielding large numbers of flaked lithic This indicatcs some weaknesses in assigning thin artifacts and evidence ofagriculture, thus indicatinga unpainted brown waresto El Paso Polychrome. post AD. 500 datc forthe introduction of ceramics in at least some areasof the northern Mogollon. Regional Trends Radiocarbon datcs from morctcntative contexts in the Carlsbad arca also suggest occupations dating Thc fact that pottery associated with the Jornada during tho span of early brown warcs as dciined for tradition appcars in the Carlsbad area indicatesthis area other areas ofthe Jornada (Katz and Katz I985b; Staley was not isolated from developments by Fonnativc 1996a, 1996b). Katz and Katz (1985b) note groups in both thc mountains and lowlandsto the west. radiocarbon dates of AD, 776 and A.D. 906 from The lack of evidence ofagriculturc the in Carlsbad and ceramic-bearing contcxts. Atotal ofelevencomponents surrounding areas indicatcs that thcsc groups may from Brantley Reservoir were assignedto the Globc simply represent an extension of the Jornada Mogollon phase (A.D. 750 to1 lSO), which rcpresentsthe time of horticulture-based adaptation (Katz and Katz 1993). heaviest cxploitation of the project area. The Globe The timc ol'occupationof this site falls about halfway phasc is initiated by the appearanceofbrown warcs and through the Mesilla phase defined for areas of the ends with the appearance of decoratedtypcs. This is southern Jornada to the east, and basic characteristics defined as a time of shifkfrom a rivcrine to an upland of the pottery are similar to those common in Mesilla orientation. A wider rangc of subsistence activities is phase occupations. Other early occupations in most indicatcd, although therc is no evidence ofhorticulture. areas of the .lorn& Mogollon in southeastern Ncw A number of dates falling within the span assigned Mexico areso poorly dated and definedthat is difficult to the Globe phase werealso recovered from a number to compare the materialculture of these early ceramic of' features excavated during the Mescalero Plains complcxcs to that represented at LA 29363. The early Pro-jcct associated with ceramics dating betweenA.D. 18 Mile phasc dcfincd for contcrnporaneous 200 and 1100 east of Carlsbad (Hill 1996a. 1996b). occupations along the Middle Pecos Valley northof This appcarsto fall in the early part ofthc Globc phasc Roswcll appears to differ from thc occupation atLA (A.D. 750 to 1200) as dcfincd fbr thc occupational 29363 in both architecture andby thc presence of some sequence at Brantley Lake near Carlsbad (Katz and gray warc pottery (Jclcnik 1967). Katz 19XSb). Also, east of Carlsbadis the Laguna Plata While basic forms,color, and surface treatments site, which yielded several structures butno evidence noted at this site are similar to those noted at Mesilla of agriculturc, and appearsto datc bctwccn A.D. 1250 phasc sites along the Rio Grande Valley,a wider range and 1350 (Runyon 1972). ofpastcs and tcmpcrare represented in the potteryfrom The recovery of charcoal from almost all the LA 29363, resulting in the recognition of' a larger features cxcavatcd at LA 29363 resulted in a large number of attributes types. Thus, while most early number of samples submitted for radiocarbon dating Mesilla phase sites are dominated by very similar (Zarnora, this volumc). Withthe cxccption oftwo early pottery, almost all of which is classified as El Paso dates from a roasting pit and a hearth rcflccting an Brown, a range of characteristics was noted in the Archaic occupation, almostall the samples dated to :I surface and pastcs of brown ware pottery fromLA similar rangc,with median datesin the eighth andninth 29363. This variation resulted in assignment of pottery centuries. These dates indicate anoccupation sometime from this site to a number of type catcgorics including betwcen AD. 750 and 800 during thc early Globc El Paso Brown (36.2 of all pottery), South Pecos phasc, orat the transition of Formative period I and 2 Brown (36.2 percent), Jornada Brown (17.8 percent), as previously defined. Therefore, ceramicdistributions El Paso ThinPolychrome? (3.8 percent), JornadaiSouth and associated radiocarbon datcs thusfar offer the Pccos Brown(3.8 percent), and McKenzie Brown(2.1 strongest evidence1 know of fora ceramic occupation percent). Tables 33-37 also indicate a closc but not at such an early datc anywherein the Jowada Mogollon perfect relationship between the defined typcs and other than Mcsilla occupationsin arcas of thc southcm various attributes. Although relationship is circular, Jornada area alongor near the Rio Grande. particularly Ibr attributes such as temper, the Most of the types occurring atthis site are those independent recording of attribute traits provided for previously noted to occur at the earliest ceramic the documentation of expected differcnccs. These occupations in this area with the exception ofthe few included the much higher occurrenceof unpolished very thin shcrds resembling unpainted portionsof El

70 CARLSRAD RELIEF ROUTE Table 37. Mean Vessel Thickness for Sherds from LA 29863

interior and exterior surfaces for sherds assigncdto El Sierra Rlancas. One sample reflects granites from the Paso Brown, and a higher mean thicknessfor sherds Franklin Mountains in the El Paso area. Thus, it is assigned to Jornada Brown. Rcfiring analysisindicates likcly that most, if not all, of the pottcry fromLA similar ranges of paste colors of shcrds assigned to 29363 was producedin arcas to the west and northwest. various regional groups. This is consistcnt with observations andinterprctations This widc range of brown ware pottery typcsat ofccramics from othcrsites in the Carlsbad area, wherc this site may eithcr rcflect pottery produced at a the grcat majority, if not all of the pottery examined, number of dispersed areas in the Jornada Mogollon, has also been interpretedas having derived from other variability within locally produccd pottery, or a arcas, particularly thc Sierra Blanca locality (IIill combination of both of these. It still has not been 1996a, 1996b; Katz and Katz1985b). The variability established whethcr pottery was produced in thc noted indicates vessels rcpresented atLA 29363 may Carlsbad Basin and other areas of thc easternmost have been produced ovcr a very widcarea. Exchange cxtcnsion of the Jornada Mogollon, although the of ceramic vessels with groupsto the wcst couldeasily possibility of pottery inthis area has bccn suggested meet the dcmands for the small numbcr of vesselsin (Wiscman 1998). Leslic (1979) notes several possible use duringa given time. local varieties of Jornada Brown in thc eastern Many ofthe scattered early ceramicsites probably extension. Thcsc varieties are widcspreadbut morc reprcscnt seasonal occupationsby distinct groups who comnlon in spccific areas, and a corrclation within cach eithcr originated in or came in contact with areasof thc varicty with the availablelocal materials, espccially wcst where this pottery was produced. These typcs of clays and sandstone available for use as tcmper,is ties and seasonal movementof groups may ultimately postulatcd (Leslie 1979).A southern area brownwas have facilitated the movement of Jornada Brown Ware also dcfinedas a local variant of ElPaso Brown found pottery over very large arcas covering much of primarily in the Maroon Cliffs/Nash drawma (Leslie southeastern Ncw Mexico and west Texas. One 1979). This varietyis charactcrized by local rcd clay possibility is that these occupations do not represent and sandstonc tcmper from thc Permian Formation, typical Jornada Mogollonpopulatians, but more mobilc which results in a friable, easily wcathered surface hunting and foraging groups scattered ovcr muchthe of (Lcslic 1979). High Plains, somc of which wcrc closely intcracting All of thc pottery identified during the prcscnt with the Jornada Mogollon. If this is the case, thcn study exhibited tempers,pastcs, and surfaccs similarto interaction and relationships between Plain hunters and types known to have commonly been produccd in Southwest agriculturists postulated for later either the El Paso area or in Sierra Blanca County occupations (Spielman11 1991) may havc begun earlicr (Wiseman 1998). Tcmper from thesc sherds appearsto than oftcn assumed. he derived from granitc, aplite, and monzonite available from sources in these areas but not availablc Pottery Function in thc Carlsbad area(Hill, this volume). Collcctionsof clays and tempers fromthe Carlsbad area conducted as Data concerning both thecharacteristics and forms of part of the present studyfailcd to locate local cxamples pottery fromLA 29363 providc information aboutthe of the tcmpcr sources. Petrographic analysis of 10 use of ceramic vessels at seasonal sites in southwcst sherds resulted in the idcntification of five distinct Ncw Mexico. Prcscntly, it is usually not possiblc to temper groupsthat may seemto indicate several arcas detenninc the general forms of vesscls from brown of origin (.Hill, this volurnc). Themost conmon group ware body shcrds basedon location ofpolishing. While contained materials wcathered from thc Capitan in some arcas of the Southwcst, combinationsof sherd Mountains. Onesample rcflccts syenite availablein the shape and location of polishingor painted decorations

CERAMfC ARTIFACT ASSEMBLAGG 71 provide clues concerning associated vcsscl forms,this primarily on the examination of Mcsilla phase sites, does not apply for Jornada Mogollon Brown Ware Whalen (1 994b) argues for more continuity outof thc sherds. Exanlination of brown warerim sherds from Archaic through adaptive patterns supplemented by many Jornada Mogollon assemblages indicates that horticulture and pottery but with strong pcrsistenceof there was often no association betweenvcsscl shape aspects of earlier Archaic adaptations, with some and the presenceor location of polishing. Attemptsto differences in overall seasonal strategies. Such recognize spccific vessel forms were limitedto rim and continuity may largelyreflect I imitations imposed by jar neck shcrds exhibiting shapescharacteristic of' a desert environmentsof this area onagriculture. Similar particular form, and some sherds exhibiting distinct patterns appear to be represented until the appearance surfaces. Thcsc examinations indicate the presence of' of a short-livcd agricultural adaptation after AD. 1000. a variety of form including bowls, seed jars, and Likc El Paso Brown Wares, ccramios associated ncckcd .jars. This indicates that undccoratcd brown with the earliest Southwest occupations tend to be warc vessels at thissite appear to havebccn used for a relatively rare, and the earliest pottery from thcsc wide rangeofactivities including cooking,storage, and regions representsan undecorated polished brown ware serving. produccd with high-iron alluvial clays and often While somecxterior sooting andfire clouding was exhibiting a dark paste (Wilsonet al. 1996; Whittlsey noted, the lack of complete or partial vessels made it et al. 1994). A wide range of forms is associated with impossible to identify sooting patterns indicative of these early brown wares including bowls, jars, seed and cooking. Still thc presence of distinct pitting on interior necked jars. Such ceramics arc oftcn associatedwith surfaces indicates usein cooking (Skibo and Blinman groups that practice agricullurc, but still are fairly 1999). mobile and dcpcndcnt on wild food sources. While mostattention about the conservativc nature Characteristics of' ceramics produced in many regions of brown ware tcchnology in areas of the Jornada of the Southwcst changed significantly by the region have usually focusedon chronological concerns, bcginning of the seventh century, when both painted the apparent lack of anysignificant change in surfacc and textured decorations become more prevalent. Such forms over many centuries inthis arca may also have changes seemto correlate with increasingly sedentary important functional implications. As previously lifestyles that may have resulted in increased indicated, ceramic assemblages andarchitectural fonns specialized use, differentiation, and decoration of from Jomada Mogollonsites are very similar to thosc pottery vessels. For example,the incrcased distinction noted for the first pottery-bearing occupations inthe of decoration, paste, and form between utility and Mogollon Highlands, Hohokam, and Anasaziregions. dccoratcd warcs may reflect increased reliance on The most obvious difforence in occupations in thc specialized activities associated with sedentary Jornada Mogollon dating from aboutA.D. 500 to AD. agriculture, including the boiling and servingof corn, 1100 and contemporary occupationsin other regionsin rather than the very gcncralizcd vessel assemblages thc Southwest is the lack OF regionally specific associated with earlier mobile or seasonal strategies. developments in technology, painted decorations, and The lack of such a shft in ceramics in muchof the textured trcattnents. In contrast, pottcry producedin the Jornada Mogollon region mayreflect the continuation Jornada Mogollon arca continucdto be dominated by of mobile or seasonal patterns of plant and game plain brown ware pottcrysimilar to that associated with exploitation similar to those associated with earlier the earliest ceramic horizons. occupations elsewhere in the Southwest (Whalcn This long consistency in ceramics produccd by 1994a). Such a scenariois also in agreement withthe groups in Jortlada Mogollon may reflect a more low frequency of ceramics as compared with other conservative adaptation by groups in this area. Based artifacts, particularly flaked lithics.

72 CARLSBAD KELLEF ROUTE PETROGRAPHIC ANALYSIS OF JORNADA BROWN WARE CERAMICS FROM THE CARLSBAD RELIEF ROUTE: LA 29363

David V. Hill

Ten Jornada Brown Ware sherds fromLA 29363 were The texture of the monzonite is anhedral granular.Also submitted forpetrographic analysis. Theceramics were present in fhc paste are a fewcoarse-sized inclusions of analyzed using a Nikon Optiphot-2 petrographic altered biotitc. Some of the larger monzonite grains are microscope. Each ofthe sherds was t'trstexanlined and a stained with hematite. description was madeofthe paste. Asecond examination of the sherd sample was conducted comparing thc Sample IO similarities and differencesin thc paste to one another. The sizes ofnahlralinclusion? and tempering agentswcrc The paste of this sherd is a light brown color andis described in termsof the Wcntworth Scale, a standard slightly birefringcnt. Sparse finc sands and brownbiotite, method forcharacterizing particle sizes in sedimentology. much of which has altered to hcmatite, is nahlrally These sizcs were derived from measuring a series of prcscnt in the ceramic paste. The vessel was tempered grains using a graduated reticlc built into onc of the using a monzonite or quartz monzonite porphyry. Quartz microscopes optics. The ages of inclusions in the is present in trace amounts as isolated mineral grains and ceramics wcre estirnatcd using comparative charts in one rock fragment. The texture of the monzoniteis (Matthew et al. 199 I; Terry and Chilingar 1955). subhedral granular. Finer grained monzonite fragments contain sparsc magnetite with hematitic rims. A fcw of Analysis of the Ceramic Samples the largermonzonite fragments contain green hornblendc. The feldspars display alteration to clay minerals and Sumple 8 sericite. The monzonite fragments and isolated grains of orthoclase, plagioclasc, and quartz account for about I5 The paste ofthis sherd is brown andslightly bircfnngent. percent of the pastc and range in size from mediumto The paste contains about 5 percent fine- sized quartz and very coarse in six. feldspar grainsthat are likcly natural inclusions in thc clay. The paste was telnpcrcd using a crushed microclinc Sample I I granite. The granite consists of quartz, plagioclasc and orthoclase with abundant microcline occurring in both The paste is a reddish brown color and is slightly rock fragmcnts and as isolated mineral grains. The birefringent. The paste contains about 30 percent fragments of granite and the isolated mineral grains inclusions. The majority of these inclusions are veryfine derived from them range in size from mediumto very to medium-sized. About 10 percent of the inclusions coarse and constitute 15 percent of the paste. While a few range between medium and coarse, with a fcw very of the larger feldspar grainsdisplay some alteration to coarse rock fragments present, These larger fragments sericitc and clayminerals, most of the fcldspars have an consist of isolated mineral grains, primarily Weathered unwealhcred appearance. orthoclase, and rock fragments. The rock fragments consist primarily of orthoclase with granophyric or Samplr Y micrographic texture. Some of these rock fragments also contain brown biotite that has altered to hcmatite and clay The paste of this sherd is a reddish brown and slightly minerals. Onc rock fragmcnt contains orthoclase with birefringent. The paste contains about 15 percent sparse magnetite cubes, Weathered biotite, and epidote. inclusions. These inclusions, which consist primarily of Isolated mineral grains consist mostly of altered quartz, orthoclase, and some plagioclase, range in size orthoclase, although a few grainsof quartz, epidote, from fu~eto coarse. However, only a fewgrains can be plagioclase and black opaque inclusions, probably altered classified as mcdium to coarsc. It is likely that the hematite, are also present in thc paste. The majority ofthe inclusions represent natural materials in the ceramic body. feldspars are highly weathered, often to thc point of' The few largerrock fragments indicate that the source of opacity. It is likely that all of thc inclusions in the ceramic the scdiments in the clay was a porphyritic quartz body represent rnatcrials naturally present in the source monzonite, where orthoclase wasthe porphyritic mineral. clay.

PETROGRAPHIC ANALYSIS 73 Sample 12 Sumple I6

The pasteof this shcrd is a light yellowish browncolor. The pasrc of this sherd is a yellowish browncolor. The The paste contains isolated mineral grains that are pastc contains about20 percent silt-sized to fine feldspar bimodally distributed in terms of size. The paste contains and quartz sands. There are also opaque black inclusions about 25 percent silt-sized to fine isolated mineral grains present in trace amounts. Also present in the clay bodyis consisting of weathered feldspars and quartz. Medium to about 10 perccnt mediumto very coarseisolated mineral very coarse inclusions are also present. These larger grains dominated by orthoclase. A few grains of inclusions consist OF weathered orthoclase, microcline, mcsoperthite are also present. The orthoclase displays plagioclase or some sparse quartz and makeup about 10 slight weathering. percent of the ceramic body. Magnetite, brownbiotite, usually altered to hcmatitc, and clay mineral and green Sample 17 hornblende are occasionally presentin rock fragments that consist primarily of subhedral granular orthoclase Thc paste of this sherd is a yellowish browncolor. The laths. paste contains about 15 percent silt-sized to fine inclusions. Coarse to very coarsc-sized mineralgrains and B few rock fragments are present. Thc mineral grainsare predominantly orthoclase, although some microclineand The paste of this shcrd ranges from a dark brown to plagioclase arc also present. The rock fragments consist opaquc black color. The body contains about35 percent of orthoclase with someperthritic texture. Also present silt-sized to fine inclusions that consist of weathered are occasional magnetite cubes and brown biotite, usually feldspars and quartz with the feldspars predominating. altcrcd to hematite. Sparse brownbiotite is also dispersed Medium to very coarse inclusions makesan up additional throughout the paste. A single coarse-sized trachyite 7 percentofthe ceramic body. These inclusions comist of inclusion was also prcsent. isolated mineral grains and rock fragments. The few rock fragments present consist of orthoclase with sparse Discussion amounts of quartz and plagioclase. The amount of orthoclase is also greater than that of the quartz and Samples 10, 12, 16, and 17 bear a strong resemblanceto plagioclase. Many of the feldspars display granophyric each other in terms of pastc color andthe presence of texture. Sparse brownbiotite, usually altered tohematite coarse-sized weatheredfeldspars. It is likely that these and clay minerals,is also present in the paste. ceramics werc derived fromthe same productive source. It is also likely that these shcrds werc produccd using Sumple 14 materials dcrivcd froin deposits weathered from the Capitan pluton, based on their porphyritic texture The pasteof this sherd is a medium brown colorand is observed in the samples as a bimodalsize distribution of slightly birefringent. The paste contains 35 percent fme and coarseinclusions. While the Capitan pluton has isolated orthoclase grains with sparse microcline, been described as a granite, the eastern margin of the plagioclase, and quartz. These isolated mineral grains pluton has less overall quartz present than the western range in size from fine to medium-sized. The fewrock edge (Allenand MacLemore 1991). fragments presentconsist of anhedral granular massesof Sample 9was made usinga clay bodythat contained orthoclase or orthoclase and quartzor plagioclase. Some monzonite. It is possible that this sherd also contained brown biotite is also present in the clay body and is materials dcrivcdfrom the Capitan pluton, but froma usually altered to hematite and clay minerals. difycrcnt resource area thanthat used for Samples 10,12, 16, and 17. Surnplr IS Samples 1 1 and 13 also contain monzonite.Both of these sherds display granophyrictexture. Granophyric The paste ofthis shcrd is an orange-browncolor. The texture is common on the western margin of the Capitan pastc was tempered using a intrusive igncous rock, pluton (Allen and MacLemore1991). classifiable as a syenite. This rockis charactcrizcdby the Sample 8 contains a microcline granite. Microcline prcscnce of plagioclaselaths that enclosc brownbiotite granite has becn reported in ceramics produccd in thc and magnetite.Fragments ofhtrusive igneous rock make southern Hueco Bolson.The sourceof the granite lies in up about 30 percent of thc ccramic body. the Franklin Mountains (Hill 1991). However, little compositional work has been donewith ceramics from

74 CARLSBAD RELIEF ROUTE further north in the Tularosa Basin wheregranites are also Sample 15 was tempered using a syenite. Sycnites available. It is possiblc that more than onearca could are available in the Sierra Blanca area (Allen and Foord have produced ceramicsthat contain rnicroclinc granite. 1991).

PETROGRAPHIC ANALYSIS 75 76 CARLSDAD RELIEF ROUTE CHIPPED STONE ATTRIBUTES

Philip R. Alldritt

Lithic artifacts foundon LA 29362 and LA 29363 were Attributes used for analysis are dcscribcdin the indicative of activities undcrtaken by the fonncr following discussion. occupants. By looking at the attributesof cach artifact (described in the methodology section), reduction Material Type strategies and techniques were rccorded to b'we an overall view of the functional diversity of each site. A three-character code was givento classify most lithic From the survey and testing it bccame apparentthat LA material. The codes wercset up so that major material 29362, because of its location and quantity of lithic groups fellinto specific sequencesof numbers and so material, would most likely be determined to be a that thc sequences progressed from gcneral material quarry site and was approached as such during groups to specific named materials with known excavation. Bedrock in thc area was 3-5 centirnetcrs sourccs. Named rnatcrials with known sources wcre below surface and many artifacts exhibited noncultural givcn individual codes.Gaps werc left in the variable wear patterns from cattle or vehicles. During testing on list to insert codes for local materialsor to make other LA 29363, thermal features indicated that at lcast a distinctions that might be dcsired. temporary camp existed, and the presence of ceramics suggested that some type of food storagc was taking Materiul Qualiv place. Excavation ofLA 29363 was carriedout with the hopes of at lcast somc shallow storage pits or Texture andrnatcrial quality are subjective measurcsof possible cooking pits which would suggest a seasonal grain size within material typcs. Within material types, occupational pcriod for the sitc. Lithic artifacts the textureof individual artifacts was scaled fromfinc recovered from both sites during testing warranted to coarse. Therc may bc great variability in texture further investigation, within matcrial typcs withthe exccption of the glassy category, which in the Southwest applies only to Methodology obsidian and glass, that exhibit no range in grain size. Inclusions that do not affcct flaking quality, like specks of differcnt colored materials in Pedcrnal cherts, or Lithic artifacts were systcmaticallycollectcd by 1-by- 1~ m grid unitsfrom LA 29362 and LA 29363. Thcy werc dendritcs in chalcedonies, wcre not considercdflaws then cleaned and analyzed formatcrial type, material and werc not coded as such. All rnatcrials have a quality, artifact morphology, artifact function,dorsal tcxture and a quality, and thereforea "not applicable" cortcx, portion, platform type, platform lipping, code wasnot included for this attribute. platform width, cortex type, dorsal scars, distal termination, thermal alteration, wear pattcrn, edge Art.$uct Function angle, and each artifact was mcasured for width,length, and thickncss with calipersin millirnetcrs. Each artifact A list of variable codes was used to classify each was also examined withan SOX binocular microscope artifact by function, which categorizes and describes to identify retouch and wear patterns. Lack of tools. Many artifacts didnot have a function other than diagnostic material for sitcs in this areahas made it debris from core reduction. Biface flakes wcre difficult to use othcr sites to comparcdata on any level. dctermincd byusc of a polythetic setof conditions. If The artifacts recovered fromboth sitcs in this report are 70 percent of the conditions werc met, the flakc was no exception as the material quality of the majority of considercd a biface flakc. the lithic artifacts was poor. Evidenceof a tcmporary site use was established forLA 29363 and otherwisc Dorsal Cortex these sites have assemblages typical of sitesin this immediate vicinity. Data produced byLA 29362 and Cortex is measured in 10 pcrcent increments on LA 29363 were entered on an SPSS program and chipped stone artifacts.For flakes, thisis the amountof included all of the lithic attributes, field specimcn cortcx covering the dorsal surface. Cortex present on numbcrs, grid location, dcpth and level. TheSPSS platforms is recordcd elsewhere and was not included program was utilized to organize the data and generate in this estimate. For othcr chipped stone artifacts, the tables. including angular dcbris, cores, and formal tools, the

CHIPPED STONE ATTRIBUTES 77 amount of cortcx covcring all surfaces was estimated dcfinition of biface flakes and suggests a greater or and rccordcd. lesser degree of rcduction.It may also suggcst which matcrial typcs werc selected for tool production. Portion Distal Terminution This attributc refers to thc partofthe artifact that is rcpresented. By definition, angular debris and cores are This category applies only to flakes and includes whole. It is impossible to tcll whcther thcse typcs of fragments as well as cornplcte artifacts.By examining artifacts were fragmented during or after reduction. the distal tcrmination of flakes and by comparing Flakes and fonnaltools can be whole or fragmented. rnatcrial quality and flake breakagc with distal termination patterns,it is possiblcto make assumptions Fluke Platform Types about material quality and reductive technology.

This attribute records the point of impact on wholc Thermul Alteration flakes or proximal fragments. Platforms that werc modifkd to case flake removal were scparated from Lithic materials are sometime modified by heating at those that wcre not, and those types of modifications high tcmperatures for sevcral hours. This form of were recorded. processing causes a realignment of the material's crystalline structure and can somctimes heal minor flaws like microcracks. Heat treatment is most often performed on cryptocrystallinc matcrialslike chert and Platform lippingcan bc, but is not always indicationan can be very difficult to detcct unlcss mistakes arc of soft-hammer reduction. While hard- hammcr madc. While color change and increased luster rcduction can produce occasional examplesof lipped (waxiness) are sometimes useful in determining platforms, a high perccntage of lippcd platformsin an whether an artifict has been thermallyaltcred, they asscmblage in addition to a predominantly diffused cannot bc used with confidencc unless the full range of bulb ofpercussion can suggestthat most reduction was colors and lusters are known. accomplishcd by use ofa soft ratherthan hard hammcr. A platform is considcred to be lipped if the striking Weur Patterns platform pro-jects over thc ventral surface theof flakc. Lipping is an observable eversion, which overhangsthc The use of a toolor a piecc of debitage or core an as vcntral surfacc; oftcn, a concavity is crcatedby the informal tool can result in cdge attrition, producing bulbar scar, which lics underneaththe lip. patterns of scarring indicativc of the way in which the tool was used. Purposcf.ul modification of an informal tool to improve itsuse pcrformance is also included in this category, as long as the modification is marginal This attribute measurcs the width of thc striking (extending less than one-third of the way across the platform at the pointof' impact on whole or proximal artifact face). fragments. Platforms were mcasurcd to onc-tentha of millimeter with a calipcr to help detcrmine qualiticsof Edge Angle matcrial types. This is the angle at which opposing surfaces meet at an edge. In gcneral it is recommended that only thc edges of formal and informaltools be measured; this provides Thc typeof cortex prcsent canbe a clue to its origin.A an accurate estimate ofthe range of angles that were watenvorn cortcx suggests that wasit transportcd by preferred for USC. Sevcral mcasuremcnts (3-5) were water and thatits source was most likelya gravcl or taken and averaged. Each angle was measured in cobblc bed while a nonwaterworn surfacc indicatesthat degrccs witha goniometer. an artifact was obtained whereit outcrops nahlrally. Raw Materials Dorsal Scurs Raw materials for LA 29362 and LA 29363 wcre This attribute determines thc presence of opposing collected from the area labeled the "Mcscalero Plain" scars on the dorsal surface on flakes only. Thc presence (Recves 1972), which lies between the Pccos River of scars or lack thereof may helpin determining thc

78 CARLSBAD RELIEF ROUTE Valley and the Llano Estacado. This gently sloping also an arca of Alibates chcrt. Matcrial qualitiesrange featurc consistsof deep, sandysoils, which are covercd from fine-grained chertto coarse-grained quartzite that by eolian sand, caliche, and gravel. Thcse gravels were typed as chert. This illustrates not only which contain chert, quartzite, and other siliccous matcrials materials wcre being selectcd for tool reduction, but probably derived from the Ogallala Formation. also the difficulty in detecting wcar patterns on thc Hummock and dune areas are common.Many small quartzitc. Also at LA 29362, 1,149 platforms out of sinks and playa lakes serve as water catchments. chert, limestone, quartzite, rhyolite, igneous, and Whercas most of the material rccovercd from LA chalcedony and 2,264 distal tcrminations werc broken 29362 and LA 29363 had waterworn cortex, this in manufacture further demonstrating the poor quality suggests that a large water catchment arca existeda in of the materials available. dcpressed area just south and westof LA 29362. Platform widths wcrc also takenon whole and Chert, quartzite, silicified wood, and other proximal flakesto determine material quality.Out of materials cxhibit waterworn cortex canand be assumed 7,363 flakes collected, 513 platforms wcre collapsed to be local. Waterworn cortex counts forLA 29362 and 102 were crushed, which representsa 5 pcrcent were 7,374 and nonwaterworn werc 415. Waterworn loss in the total due to poor material quality (withsome lithic counts for LA 29363 werc 505 and element of human error). Further, the distal nonwaterworn were seven. Nonlocal materials terminations forLA 29362 show that, of thc artifacts recovered included Tecovas chertand obsidian. The that had a termination, 2,264 were broken in one exampleof obsidian collccted exhibitssomc hazing manufacture and629 had snap fractures fora total of and is probably from theLas Cruces arca viathc Jernez 23.6 percent of the artifacts being incompleteduc to Mountains (Regge Wiseman,pers. corn. 1998). Thc poor material quality. Tecovas chert originates from Turkey, Texas, whichis

CHIPPED STONE ATTRIBUTES 79 X0 CARLSBAD RELIEF ROUTE DESCRIPTION OF CHIPPED STONE DATA-TROJAN HILL by James A. Quaranta

Geologic Deposition of Materials Ckrlsbad area indicate that quartzito nodules have a tcadcncy to sl1:ittcr uncontrollably. Neither the skills 'I'hc dcbitage assemblagefrom Tro.jm IIill is comprised nor tools of the knappers apprcd to affect this of 12,269 pieces of lithic debris of which the vast propensity (Kat7 andKatz 19XSb:S9). majority is miscellaneous chert (n = 6,485) and The lowest man value for whole flakc length is in yuartzite (n = 4,82X), along with lcsscr amountsof the the miscellaneous chert class (mean = 2 I, 16, n = 1,958, lithic materials listed in Table 38. Virtually all olthcse sd = I0.47), where thc reduced mean length is a likely lnatcrials were geologjcallytransported to the site as mnnikstation ofthe smallersize ofthc parent material, wash from the Ogallala Formation at least 200,000 ix., chert cobbles tend to he smaller than quarttzile years ago. They were subsequcntly covered by a cobbles on site. The tighter range around the mean stratum of caliche caprock, from which they more value likely reflects its 1norc consistent knapping recently eroded outwardly. Virtually all of thc lithic properties. material at this hilliop site that has cortcx exhibits An even tighter range of values abouta class mean waterworn cortcx (95.5 percent), supportillg this cxists within the ched class that has been visually determination. idcntilicd as Sat1 Andrcs chert. Thismaterial type may thus exhibit even nmrc consistent flaking qualities, Table 38. Lithic Materials frorn LA 29362 which would makc it a likely tnatcrial to be curated. The tighter rangc of values in Sa11 hndres chert wholc llakcs may be due, however, to ii later stage of reduction rcprcscntcd. This would again seem to indicate that the Sat1 Andrcs cllcrt was a more highly regarded chert. The sample size of Sat1 Andrcs chert is snlall (n = 26) so the relevance or tllcsc statements cannot be proven with any statistical contldencc. Silicilicd wood, utilized to a small extent (n = 14) had n lncan length v:~luc falling between tho San hndres and misccllancous chert values. It also had a tight rangc about its mean, RS did chalcedony flakes. Chalccdony wholellakcs (n= 26) h:id tho lowest me:itn length value ( 16.41 rnm) or material types that arc rcprcsclltcd by at least 10 spccimens. Scc 'l'ablc30 for tabulation ol' whole flake length mean and standard cieviations for all material types. Mean width and thickness valuesof whole flakes by Inaterial type exhibit similarpatterns (Tables 40 and 4 I) to those viewed in Table 39. Generally, larger, more variable tlalcc dimensions appear to correlate with Core Flake Matcrials-Knapping Properties the coarser-graincd materialtypes, and smaller, more consistent flakes arc derived from materials thatarc Where tnatcrial types are reprcscntcd by at least ten line grained.C.'asual observation or tllc material typ specimcns, we call see that the highest mean valuc Ibr on site would reveal that gravels of finer-gnlincd whole-flake length isin quartzite, with a mean value of materials are consistently smaller. 26.75 mm. 'This material type also exhibits the largest dcviations from its mean value (sd = 13.62). The Material Type-Temporal Implications largest mean length value for this material class is dw, in a11 like1 ihood, to the larger size ofthc parentcobbles. The proposition that the use of high quality lithic The morc variable flake length is most likely :i materials is a characteristic of the carlierprehistoric manifestation of its inconsistent Ilaking propcrtics. periods was cxaniincd in a regional study of the Previous obscrvniions on knapping quartzite from the Tahlc 39. Whole Flake Lcngth (mm) ‘l’ularosa Basin (Cannichael 1986). In order to do this, Cannicllael developed a lithic divcrsity index (LDT) where: LDl = numbcr of chert colors /total types of lithic lnaterials 011 site (Carmichacl 19x6). An initial investigation ofa snnzple of the lithic population liom LA 29362 (Trojan ITill) gives an LIII 01‘0.74 (where n = 220); it appeared in subsequent analysisthat virtually all ofthe variability in chert color was subsumedwithin this sample. In Cannichacl’s 1986 study, differenceof‘ mcms tests showcd Paleoindian and Archaic assclnblages to bc statistically different from one another and both to be statistically dislinguishable from latcr prchistoric pcriod assemblages, whcrcthere is an increase in coarscr (nonchert) materials. The most statistically significant differencc bctwcclz ad.jaccnt time periods is between thc Archaic and Mcsilla

I I I periods, i.c., the pattern ofdccreasing lithic quality is most pronounced through the preceramic (Archaic) into Table 40. Wholc Flake Width (111~11) the ceramic (Mcsilla) periods.l’hc incan LDI for the Mcsilla is noted as 1.00; and after snbdivisiolz of this period, Cannichael notes a mean LDI or 0.90 for thc “latc” Mcsilla phase. The Tro-jan Hjll LDI is quite low (0.74), and may indicatc that this lithic scatter wasformed subsequent to the introduction olpottery in the region (designated as the Globe phase). This supposition hassome support in that the projectile point round was an arrowhead, a few pottery sherds wcrc found, andradiocarbon-dated features, although not in direct association with excavations blocks, but onthe hill, yicldcd Globe phase dates. The low lithic diversity index lzcre is further highlighted considering the fact that color variation of chert nodules n~ybe great even when from the same 18,117 3272 Y.HU outcrop; this divcrsity could be reasorled to be cspccially great in geological secondary dcposition, Table 41. Whole Flake ‘I’hickness(mm) here a wash of Ogallala gravels, but the LDI is still quite low.

Material Type-Functional Implications

Of chcrt types being worked, the majority were from ~hcimncdiate locale. but a fine-quality gray San Arldres chert, not visible in the gravels, and a nonlocal chert visually identified as Tecovas, wcrc encountered in the debitagc assemblage, as was one s~nallflake of obsidian (Fig. 70).These aspects, in regard to material type, attest to thc site function asbeing somcwhat more than simply a quarry. Additional aspects ofthe dcbitagc assemblage from this lithic scatterwill givc crcdcnce to this interprelation, altllough undoubtedly thc predominancc of exposed. serviceable lithic materials liom which to quarry was of primc consideration to the prehistoric occupants of this locale. ". __ I- activity or special activity site, which disregards this Lariability. An attempt will bc made lo deconstruct aspeck of 1,A 29362, and tn examil~cand rcconstruct tllc formation processes that fonned this lithic scatter.

Fl,zke Portions

In order to examine aspects of tnanufacturing technology between the most dcnsc artifact clusters (i.c., cxcavation blocks) a chi-square test was Spatial r)isaggregation of Site Components and pcrforrned bctwccn flake portions andarcas ofthe site Functional Implications (Table 42). The chi-squarevalue was significant at thc IO percent level. As can be sccn from the table, the Tlu-ce block cxcavation arcas wcre superirzzposed upon largest positive deviation from expectcd (16.04893) is Lbmd in Arca C, wholc flakes. Sullivan and Roml the most visibly dense clustcrs of artifacts on the hilltop: Arca A, the most westerly hlock; Area B, the (1985) state that debitagc assemblages resultinghm most eastcrlyblock; and Area C, the middle block (Fig. an emphasis on corc reduction generally tend to bc 4). Although there was sotne eolian deposition, dominated by complete flakcs and debris,whjle those essentially all recovcrcd artifacts can be classed as produced during tool manufacture yield high surface Gnds, lying within a few ccntirnetcrs of the perccntagcs of proximal a11d distal Ilakes. It is caliche caprock. No nahlral stratigraphic break existed suggested that although core reduction occurrcd in within the 1i.w centimotcrs of eolian deposition areas other than Arca C:, these relations arc obscured overlying the caliche nnd hence artifacts wcrc grouped duc to the presence of other reduction stagcs and contrasted according to block unit only, represented. It seems probable that Area C, having cmph~sizinghoriznn~tlrather than vcrticnl stratigraphy twre wholc flakes than would normally bc cxpccted, in of the site. In thcsc analyscs an attempt was made to addition to its slnaller overall assemblage sizc, was a cxarnine this lithic scatter with regard to occupational locus or focused activity, principally if not entirely or organizational variability and to move beyond the devotcd to core reduction. administratively useft11 functional taxon 01' lirnited-

Table 42. Flake PortionIArca Associations

CHIPPED S7%INE DA 72 "TROJAN HILL 8.7 Table 43. Flake l'urtiotl (Chert)/Area Associations

When this same analysis wasrestricted to chert, the etnploycd hcrc, however, are intcndcd only for intersite 111111 hypothesis of no association could not be comparison. Percentages ofdorsal cortex presentwere disproved at a strong enough statistical conlidcllcc. recorded [or flakes in estimated 1 0-pcrccnt intervals; IIowcvcr tllc strollgcst positive deviation l'ronl these small increments were unlikely to produce bias expected was once again in Area C, whole flakes when the distributions are cxamincd in regard to their (value = 12.5 I) following the pattern of tllc previous cumulativc percentages in cacll dorsal cortex class. walysis (scc Table 43). Restricting the analysis lo Instead of employing the data in a categorical hshion quartzite materialproved uninformative, likely duc to with arbitrary cut-oll' points implicitly rclatcd to its somewhat random flakingproperties(Katz and Katz reduction stage, thedata were ranked and cach area's I9XSb). distribution was compared for its "iit" to the other arcas. A Kolmogorov-S1llinlov goodness-of-fit test was pcrfonncd comparing the dorsal cortcxdistribution for A wide variety of categorics has bccn cmploycd in the each area's Ilakcs. No statistically significant allalysis 01 dcbitagc presumed to llave been produced dil'l'ercncc between assemhlrtges would ilrlply that they by core reduction. Sullivan and Rozen ( 1985) explore were derived from thc snmc parent population, ix., a thc goncral mcthodological problcms commonto nlany population of comparable dorsal cortex distribution and of' thcsc norltool dcbitagc taxa. Often three basic thus likely, one with similarlithic rcductivc processes categories are subsumed withinthe overall typology: represented. In like fashion, a significant difference primary, secondary, andtertiary flakes. They state that would imply that the distributions werenot derived dcbitage analysesemployingreduclion stagcategories from the same or comparable parent population, and based on arbitrary division of cortex are compromiscd would thus likely reprcscnt diffcring reductive by the consequencesof operaticma1 problems. Hecausc processes. the proportion ofcortex that dciines :t specific dcbitagc Table 44 shows the cumulativc pcrccntagesin each category is not standardized, prirnary or secondary dorsal cortcx class ror flakes in Areas A, R, and C. flul

, 0 _17"" . , , , , , I 201 0 10 20 30 40 50 60 70 80 90 I00 dorsal cortex '"L dorsal cortex Figure 71. Ogive of dorsal cortex distributions.

Proportionally less of the Area C flak assemblage contains flakeswith minimal amounts oi'dorsal cortex, or tnorc clearly stated: Arca C: has proportionally morc ~ = Area A flakes with greater alnoullts of dorsal cortex. This "." =Area R would appear toindicate that carlior stagcs ofthc lithic 100 A reductive process arc rcprcsclltcd in this spatial clustcr. hrcas A and R have proportionally111orc flakes with less cortex than hrcn C, indicating further reduction occurring in these locales. The examination of this ranked data (dorsal corlcx) lends support to previous inlcrprctations based on the categorical data (flakc portion).

Table 44. Kolmogorov-Smirnov C;oodness-of-Fit Test for Cumulative Percentagcsin Cortical '"1 Classes by Area 00 0 10 20 30 40 50 80 70 80 90 100

Cortcx Arca A AIW~n Area C dorsal cortex 10 41.X 45.8 j2.3 20 53. I 52.2 37.5

Table 45 trcats thc data in similar fashion but restricts the analysisto chert flakes. Thesame patten1 is once again clearly demonstrated, with the Arca C subassemblage significantly different from that of thc other areas(C: vs. A sigllificant at the0. I percent Icvcl; C vs. R significant at thc 0.1 percent level).Diffcrcnccs as large as these would only bc arrived atby random processes one timc out ofcvcry 1,000 trials. The ogive ofthis data set (Fig.72) once againpaints a clear visual picture of the distributions and highlights Area C's distinclncss. Table 46 and Figure73 agaill treat thc data in like fwhion, this time restricting lhc analysis to quartzite fl;tkes. Once more similarpatterns arc seen in tllc data. excavation area. Neither was there Iound in this set although thequartzite flake data set for Area C is excavation area any lithic material that could be rnarkcdly smaller than that of the other arcas. The considered of nonsite origin. San Andres chert, ogivc ofquartzite llakc dorsal cortex percentage shows chalcedony, rhyolite, and unidentified igncous Area (:'to be visually distinct in the samc manneras thc materials were present in the other two artifact scatter chert flakc data set and thc data set ol' all i-lakes in a11 arcas but wcrc absent in Area C. Tlzc San Andrcs chert material classes. The K-S value dcrivcd from the table. is deduccd to be oI nonbilltop origin; it is uncertain howcvcr, only shows a statistically significant whether the chalcedony, rhyolite, and igncous arrivcd diIl'crcncc between Area C: and Area A (I percent on sitc geologically or culturally (Figs. 7 1-73), Area C level). also lackcd the visuallyidentified Thunderbird rhyolite and thc obsidian lound in Area A. Succinctly, 14 lithic Tablu 45. Cunlulativc Percentagesin Cortical rnatcrial types were encountcrcd in Arca A, ninc in

Classes (Chert) hy Area Arca 13, and only four in Arca C.

Shattcr-Angular Debris

Thc Area C subassemblage had the highest interassemblage percentage of angular debris (45 perccnt),just marginally greater than that of Area B (43 percent), but ten nwrc than that found in thc Area A subassemblage (35 percent). Area A contained X3 percent of tllc total site's angular debris, obviously thc main site locus of lithic rcduction. The higher intcrasscrnblagc proportion of shatter in Area C does not indicate that it was tllc area ofmost intensive core reduction on site. It simply adds credence to thc previous assertion that activity in this localc was quite restrictcd, whereas other activities are in evidence in Tahlc 40. Cumulative Percentagesin Cortical the other localcs. most notably in Area A. Classes (Quartzite) by Arca In experinzcnts onthe productionoflithic dcbitage from stone loo1 manurxture (Pokotylo 1978) it was noted that thc technical aspects of manufacturing a biface are substantially different Irom thosc dreducing a corc. It is adduced that the blank is held in the hand I I whereas a corc is seated on the ground, and this was the procedure employed in the experiment. More control is obviously achieved in the process of reducinga blank and less shatter is produccd. In the experiment, 48 percent of the dchitage from reducing a core was classed as shatter. This compares quite well with our percentage of angular debris fron1 Area C? and Area H (45 perccnt, 43 perccnt). Inthe replicative knapping experimentofa blank, 22 perccnt of the dcbitage wasclassed as shatter. Arca A of LA 29362 had 35 percent of its asscmblape comprised of angular debris or shatter. The lesser Utilized Debris-Exotics interasscmblage proportion of shatter in Area A mu-y be indicative ofthis later stage oIrcductionaccon1panying Additional support to the postulation that Arca C primary core reductio11in this localc. This was tllc only activities wcrc ol' limited scope and duration maybe area on site in which bifacial tools werc found found in the Iact that out of thc three dense artifact (excepting a projectile point with an impact fracture in clusters only Area C was devoid ol'utilizeddebris. No Area C). artilhcts with use-wear or rctouch were found in this Table 47. Platform ?'ype/.Material Type Associations

Significant at the .OS pel-crnt lcvcl

PILl(/hnns cxists bctwccn site areas and pla~lon~ltypes (and ultimately reduction scr~ucnccs).is thus restrictcd 10 the It has bccn noted that". . . more scars arc cxpcctcd on chert nlatcrial class. This simplistic heuristic does not flake platforms at late stagcs [ol'rcduction I as :I resull show any evidence or association between any of the ofplatfonn preparationto better control flake removals threc ma.jor platform types with excavation nrens . . ." (Pokotylo 1978). If it can be reasoned that (Tablc 48). mnultiple platform scars arc indicative of grcatcr Regarding all platforrr~-rernn~llt-bcaringchert reduction proccsscs than cithcrsingle facet platforms or flakes, ~1 small pcrccntage of crushcd plnlfonns was . cortical platforms, parallel reduction scqucrlcc patterns found in Area A and in Area 13 (2 percent in both). No , arc in widcncc bctween the site's ttrcas. As common crushcd platforms were found in Area C'. "This fault sense would indicate, the more coarsc-grained occurs whcn the knappcr dclivcrs a blow to a sharp, maiirials should IX uscd less than the finer-grained acutc platform. The platform being fragilc, materials, and it stalistical association does exist disintegrates , . . . Crushing is n problem more c01111~1011 between platli-mn typcs and material types (chi-square toward the cnd of the thinning prc~~ss. . ." (Lord significant at thc 0.05 levcl), with tnore singlc-lhcct I993:35,36). The absenceofcrushcd platforms in Area platfonns on quartzite flakes and fewcron chert flakes C: provides supporting evidcncc for the lack of late- than would nonnally be expected (Tablc 47). This stagc reduction in this localc, whereas it apprs analysis, which aims to discriminatcan if association

Taldu 48. Platform TypdArra Associations I

CHIPPED STONE LIATA-TR(3.JAN HILL 67 that some late-stage reduction is representcd in Area A unlil

Although thc thrcc lithic clusters exhibit diffcrcntdata patterning, it is clear that corcreduction is tl~cprincipal activity reprcscntcd by the overallsite dehitagc. '1'hc great rnany corcs and tested cobbles attest to the sitc's primary iirnction as thatofa quarry. Eroded out from the caliche caprock, a lnultihlde of serviccablc lithic materials were readily available on thc surface ofthe hill. These rounded cobbles. as previously stated, arrivcd as wash from the Ogallala 1;olmation. at least 200,000 years ago. The most abundant natural and culhmlly altcrcd materials on site are chert and quartzite. It has also been observed that thcrc isdifferential spatial patterning ol'thc debitagc with regardto st:tgcs ?'hroughout all sitc nrcas, 159 local chcrt, 91 quartzite, oflithic reduction represented, and that othcr activities 12 siltstone, :md 3 San hdres chert cores were found, beyond corcreduction arc represented. Since activities as were 1 chalcedony core, I silicilicd wood core, 1 beyond sinzplc quarrying episodes arc in evidence, rhyolite core, and 1 limestone core. Ofthese, 229 wcrc which noncthcless appear to bc subordinate activities, found in Area A, 28 in Area B, and 12 in Area C. Ofall an emhcddcd resource procurc~ncntstrategy was likely bifacial corcs from excavatcd nrcas, 92.2 percent ofthe etnploycd by the site's inhnbitants. The likclillood ofan total wcrc found in Arca A. Forty-seven hifacial cores embeddcd strategy prompis furtherinvestigation ofthe wcrc found within Area A, as opposed to 1 l'rom within lithic database. In thc nearby habitation site of Macho Duacs, we Arca I3 and 3 from within Area C. Considering the individual areas, 21 percent of the Area A corc encounter both chert and quartzite matcrials visually assernblagc is comprised ofbifacial corcs (11 = 47/229); identical to those found here (Trojan Hill).It appears that various lithic materials may havebccn transported 25 percent of the Area C: core assetnblage are of this type (n = 31 12); whereas only 4 percctlt of the cores away from Trojan TTiI1, although not in comparrhlc li-onl Area R (11 = 1/18) arc bifacial. Less strategic and proportions to thoserepresented on the hilltop. less intensive lithic reduction appears to be a correl:ttc Table 52. Stem and Leaf Plot (Chert Core Length, mm)

Area A 1s 2136788 3 I 1222233444456hh7777XX99999Y9 40000000001 112222233334444455500~~~i777~~~~ 501222333344555555h67777XX 6013356677778999 71 244O7XXX9 X0 I4 9

Area B Arca C

1 2258 3 41 13569 5034s 6 5 0 X 9 7 6 x 9

Table S3. Stern and Leaf Plot (Quartzite Corc Length, mrn)

Area A

1 2 3578% 403345St56677XX 5001 122222233345h67778XSX9~ 600012244555666677X 71 11 13457X!) 80245 903t) 10 2

Area B Area C 1 1 25 2 3 3 4274; 4 5 5 67 6 756 1233 8 8 9 90 10 IO lhc ogive (Fig. 74) displays distribution data for A chi-square test (Table 54) showed that therewas flakes ofall materials, for flakes ofchcrt,and for flakes an association between material type and flake portion of quartzitc rcgarding dorsal cortcx cover. As can bc on (he overall site (0.05 percent Icvcl). The material seen, thcsc distributions virtually overlie each other, type that contributed most infinnation was the implying that thedcgrcc to which a parent cobble was quartzite. -1hc flake portions that contributed the most worked MWS not dependent upon material type. If'largc inf'onnation were mcdials, wholes, and laterals. Of' amounts of quartzite were not transported away but these, only the nlcdial and latcral flakes exhibit positivc MVW worked on site as exhaustivcly as was the chcrt, deviations from expected values in quartzite, i.c.,more then it is likely that this material scrvcd a vital functioll of' those than nonnally expected. in the contcxt ofthe prehistoric use of this sitc. What is likcly represented in the llulnerical dah is an aspect of lithic rllorphology/lithic technology that was appreciated during analyscs but which was not 120 amenable to incorporation into the existing lypologies. .- chert What 1 believe is represented by the preponderance of all material 100 -" types lateral and medial quartzite flakes is R citms core kchnology mosl nlanifkst in this material. Many

80 quartzite flakes that horc the diagnostic double-faceted ventral side with functional edge opposjte to the cortical backing(wcdge-t;hapcd) wcrc labeledas lalcral Ilakcs. (Most chcrt flakes thatwere labeled as lateral, on the othcr hand, appear to have been fr:icturcd at the platforms due to imperfections or errors in forcc and do not seem morphologically or tcclmologic~dlydistinct. The majority of quartzite lateral llakcs were 20 morphologically distinct, confonning only to visual and descriptive information of tlakes derived from 0 ,-.I- bipolar pcbblc technologies ((3)odycnr 1974:74: 10 20 30 40 50 60 70 80 90 1 I dorsal cortex Schiffcr 1976: 104- 105,125;C:'resson 1977:27; Stafhrd 1977:2X). Many quartzite medial flakes also had two Figwe 74, Ogive ofdistril~utioas,,flake duta by mutevial type. 7'ahle 54. Flakc Portion/Material Type Associations ventral sides with cortex 011 a lateral margin. As recognizable lunalc ilakc fon11, called Orange Peel Solberger and Patterson ( 1976) point out, thc Flakes (OPFs) by Stafford. Cioodycar (1974) and mechanics of this technology are such that lat-gc Schiffcr (1976) describe the tnanuf:,lcturing process of amounts of unutilizable debitage ~rccrcatcd, likely these flakes in some detail ;~ndstate that one of the represented here by many block-like medial quartzite main advantages ofthis flaketype is the natural cortical flakes. The statisticshighlight associations that hint at backing ofthis type of flakewhere thecortex is usually this technology. A very large number of these flakes on the wargin opposite thefunctional edge (Fig. 75). could not be properly assigned toa class that would The natural backing supplied by the cortex would reflect their morphology or the manufacturing provide some comfort when using the flakc as an technology. Many flakes bore attribute states that itnplclncnt fix chopping or cutting. It would thusmakc obfuscated tllcir distinctness whcn incorporated into the a fair quality cxpcdicnt tool. existing typological framework. 1 believe that the numerical data i.s displaying a pattern related to citrus corc-bipolar pcbblc tcchnology, and lhat it is greatly Itnl/~,~.-rcpresclltillgtllc amount of these flakes andthe manufhcturing behavior that it rcprcscnts. Bipolar, citrus core technology was apparently used by tllc prehistoric idlabitants ofTrojan Hill; but due to the above-mentionedoperational problcrns, wc cannot quantitativelyassess ihc dcgrcc to which it was cnqdoycd. We caninvestigate, however, thebehavicml correlates of this manufacturing technologyand its Figure 75. 0rarigepeel.flakes. possible role in a11 cmbcddcd rcsourcc procurernent strategy. Bipolar technology may bc characterized as less efficient than other tcchnologics, as it produces much I3ipolar Ilnking is itscll'a lncans to an end.. . unutilizablc dcbitage and many OPFs without the This results in citllcr a ready-made cutting fimctional edge, nevertheless it can be tcclmically tool or a like-preform . . . Economy of time, advantageous. The desired flake ronn has a functional encrgy and nlatcrials is tlle real issue here (not edge and hasa natural backing that would facilitate its "true bipolar flaking"). . . Pebble resourccs use in heavier-jobs. 71hcscilakcs would also tend to be and bipolar reduction skills provided the larger than other flakc types as they makc we of the logical solution for the acquisition of entire brcadth 01. the cobble, again indicating hsks workable materjals, as well as a time and whc1.c a larger tool may bc preferred. "Jt isobvious that energy saving technique in lithic removal oftlakes frompebbles in this fashion will not rL.duction.(Cressotl 197727) accomplish anybasic tasks of tool tnanufacturc such as decortication or thinnjng e *'' (Cjoodyear I974:74), so it would seem that the production of cxpcdicnt tools Cresson (1977)states that in the castem UnitedStates, may be a correlate of this technological behavior. the archaeologicalrecord is ovcrllowing with artifxis Additionally, "Anyone interested in lithics, who has reduced froin pebbles and that a high incidence of tried to rcducc R quartz or quartzite pebble knows the specific pcbblc (quartz, quartzite,sandstone) resource considerable amount oftime spent preparingthe piece using groupsprcdomillatc the cultural milieu. So much to be bifacially thinned" (Cresson 1977:27). Since this so, he states. that perfectly good and abunda~ltnatural coarsc-grained material docs not exhibit the consistent lithic outcrops occurring within 60 to X0 were km conchoidal knappirlg properties of chert, it would not neglected by mnc groups. In desert environments, appear to be an cffcicnt usc of energy toattempt to which lack the usual fcaturcs of lithic resource reduce this material bifacially. The more rupid availubility (as do coastalplains), cultural adaptation reductive technique, which achieves its usc- would mandate that use is made of every applicability sooner regardless of greaterwaste outputs, cnvironmcntal resource, including pcbblcs. would be the preferred technique for an expedient tool, It is accepted thatbipolar rcduction of cores is not especially if expediency is planned as part of an a precisely controlled technique (asnoted by Stafford embedded strategy where suitable low-grade lithic Solbcrgcr and Patterson ( 1976) state that large 1977). mntcrinl is rcndilyavailable (Fig. 76). amounts ofunutilizable debris would be cxpcctcd with this technology. Noncthclcss, it is evident that some degree of control was attaincd, which yielded a

9 2 CARLSHAD Rk;LIEF ROUTE TJtilization of the Lithic Artifacts and Functional lntcrpretation

Across the site, 49 picccs ofdebitagcwere tabulatcd as 11:iving a culturally modiiled etigc angle. (Xthcsc, 45 wcrc simple core flakes that had bccrl utili7ccl by the Figure 77. Utilized artifacts jkonr 7hjrm Hill; (a-11) prehistoric inhabitants to fulfill task rcquircments. scraprs, (c) scraper/gravev. Virtllally all utilized debitagc was noted in the fine- grained material, specifically chert. Thisis likely due to the claritywith which this material exhibits signs 01 ancicnt behavior,and surely over-represents activities associated only wilh tllismaterial. Utilization ofcoarse- grained materials, tlloagh expcctcd, is often difficult to conlidently discern. Nevertheless, a second look at some of the quartzite OPFs rcvcaled what appeared to be use wear (examplesare FS 435, FS 7 I, scc Fig. 76). As mentioned, thesc larger, coarse-grained, though sturdier flakeswould be more ideally suited to rougher tasks. thc site evidently not occupied for As was l 3 crn extended occupations, it sccnls plausible that thcsc ~ ~~ ~~ roughcr flakes fonncd part ofthe inventory ofmatcrials Figure 78, Sifuce [email protected] Trojm Hill. nccded for plant-gathering as part of an crnbeddccl In addition to thc greater numbersof corcs reduced strategy onthis "quarry" site. here, the largest amount of utilized debris was Thc smallercllert flakes were less suited to thcsc recovered from this cluster (Figs.77 and 78). The 45 implicd rough tasks. Tllcir size and weaker functional picccs of utilizcd debris appear to pcripherally ring an edges made thcm ill-suited to tasksthat required more area some 15 m wide. If this pattern is rcflccting brawn. Nevet-thclcss, there is concrete evidence thal cultural transformations, and the Ionnation oi'thc lithic small chert flakcs were employed in site tasks. No scatter hcrc was fonncd over a brief time period,an utilized flakes at all were found in Area C:, as stated hypothcsis could bc formed that states people were previously. In Area R, four utilized flakcs were found, engaged in other task activities pcriphcral to thc as was the projectile point that manifested an impact primary core reduction :ma and wcrc occupied hy fracturc. Possibly. this indicates that R meat package (hew tasks at a safe distance from the flying shatter of was prnccsscd here. In this tcrrain thc likely hullling primary core reduction. strategy is I'orager-style hunting. Thismay, thus. dso Virtually all utilized debris or1 site manifested reprcsc1lt one comporm1t of an embedded resource unidirectional cdgc wear (Fig. 79). This is usually acquisition strategy. taken to indicate thc transversal rnovenlcnt of thc Area A displayed themost site activity ofall areas. artifact while engagedin the task that left these traces. This movement is normally :issociatcd with scraping activities. It is possible that cithcr plant or animal (or both) materials were scraped with Illis movement. Forty-two of these artifacts with unidirectional edge wear have edge angles associated with heavy duty cutting or scraping activities, according to the edge angle framework, postulated by Willnscn (1970:7O- 71). Only fiveof the utilizedartifacts had cdgc angles that could indicate butchering within Wilmscn's framework. The heavy-duty scraping evidenced through the chert artifacts, coupled with the fimctional implications andprobable edgc wear of' the quartzite OPFs, and the crushing andstep fractures (Le., possible use-wear) along the bifacial margin of some cores provide strong cluesthat site activities at Trqjan Hill encompassed a range of activities, likely as pait of an cmbcddcd resource procurement strutcgy. The prepondcmncc of exposed serviceable lithic materials was evidently of prime consideration to the Tools prehistoric occupants 01' Trojan Hill, The very large number of cores attests tothe main sitc function as No tools wcrc encountered in the Area C asscmblagc. being that of a quarry. No area on site had a llakc Only Area A and Area B contained artifacts that could asscnlblage dorninalcd by late-stage lithic reduction be classed as formal tools. Area A containcd three characteristics. However, differential spatial patterning bifaces: FS 30, a late-stage quartzite bifacc (43-by-26- was evident bclwccn the site areas' subassemblages. by- I5 mm);1's 1 17, ancarly-stage quartzite bifxc (SX- Statistically significant differences inthe assemblages by-49-by-21 mm); and PS 232, a middle-stage chert wcrc found that indicated nrganiLational and biface (5.5 -by-42-by-20 mm). Area B also contained occupational varinbility within the sitc. three hifaces: t;S 540, a middle-stage chertbifacc (X Flake portions were scen to be rclalcd to site area. by-19-by-9 mm): FS 551, an early-stage cherl bilhcc The Area C asscnlblage had ;I grcatcr proportion of (33-by-22-by-1 I nun); andPS 742, ;inolhcr carly-stage whole ilakcs, which is often lalccn to be related lo chertbiface (30-by-24-by-8 mm).Additionally, a small primary core reduction. This area also manifestcd a pro-jcctile point was encounteredin Area B, in the same greater proportion of highly cortical flakes than the I-by-1-m grid unit as the early-stage bil-ace FS 742. other arcas, with the same technological implications. The stem of the projectile point wasbrokcn. The distal Area C had thesmallest overall assemblage,and lacked end also was broken, manifesting an impactfracture. nonlocal lithic nmterials, as were found in the other site ;mas. Area C was also devoidoi' lhlizcd debris, as Ilammerstones opposed to Area A and Area H.No crushcd or abraded flake platforms were encountcrcd in this locale. The Remarkably, only seven harnnlerstones were identified implications of thcsc data appear to jndicatc that, from the site. These all derived from Area A. Fivewcrc although an early reduction stage focus is manifest sitc- quartzite; two were chert. wide, this area had a particularly limited activity One-llundrcd-Glty-nine tested cobbles were regimen. collected from the sitc. These are cobblesfrom which More reduction occurred in Area A than elscwhcrc up io two flakes had bccn removed, as though the on sitc, as attested to by its largcr overall assemblage material was tcstcd for suitability and cobbles that and large number of cores. IIowcvcr, a more well- appear to havebeen intentionally broken in half. It is rounded activity rcgimcn is noted for thisarea than for surmised that manyofthe cobbles with only onc small the other areas, especially Area c'. Area A had the ilakc rcnlovcd wcrc cxpcdicntly used hamnlcrstoncs lowest angular debrjs to flakeratio oi' nlt sitc areas. that llad been Inisclassif-icd. Ratios ofthe other areaswcrc similar to those obtained Of note, the specimens that were classified in replicative work for primary core reduction. "I'hc correctly evidently bore sufiicient signsof battering ratio obtained from Area A approachcd that of blank and were derived frotn the area with the most intensive reduction from replicative cxpcrimcnts. Less focus on simply the earliest reduction stageis noted in this area.

94 CAKLSBAD KELllil~KOU'I'E The assemblage from this 3rca appears to indicate The obvious, primary site function was for lithic reductio11 along a greater continuum than in the other procurcmcnt. Nonetheless, subsistence activitieswere areas. Both crushed and abradcd flake platfornls wcrc in evidence and aspcctsofthe lithic technology pointto cncountered in the Area A asscmnblape, often taken to a planned expedient Ilakc technology, whcrc citllcr he related to late stage lithic reduction. The unrctouchcd core flakes or lunate, orange peel tlakes distributions of whole chert flakes between the sitc scrvcd task needs without further modification. areas alsoindicntcd that Area A was unique in that it f-xpcdicnt tccllnologies tend to bc associatcd with low had a significantly greater propottion ol small flakes tool replacement costs.This site obviously had low tool than elsewhere on site. Arcit A also exhibited thc replacement costs.A rcsidcntial camp that h:td this sitc largest diversity ofhigh qualitylithic materials and had within its loragjng radius would also have low tool a great nunlbcr ofutiliyed tlakesin its asscmblage. replacement costs andwould likely havean expedient The Area B flake assemblagewas significantly less technology rcprcscntcd, as well. One such sitc within the foraging radius (LA 29363) will he treated later. cortical than that of Area C:. 7'110 chert whole flake asscnlhlagcs betweenArea K and AI-ea A. regarding The plarmeti expediency and the dearth ofnonlocal length and thickness,were also seen to be signiticantly lithic debitage seem to point toa somewhat low level different. TheArea 13 asscmhlagc did not have as great of tnobility within the mobility continuum. The low a proportion of small thin llnkcs as did Area A. Sonic lithic divcrsity index valueobtaincd lor the overall site exotics were found in Area B, but not the full range as implicd both this and that occupationlikely occurred found in Area A. An equal proportion or cnlshed subscyuent to the introduction ofpottery in the rL,''~~1011. plntforms were found in Arca I3 as in Area A, although Additional lincs of evidence also indicate this. no abraded platforms were noted. Fourutilized hkcs Intra-assemblage data indicate orgat1iz:ttional were recovcrcd in this locale, as opposed to 45 in Area variability within thc sitc, likely related to occupational A and 0 in Area C.'. Most importantly, an arrowhead episodes. Palimpsest occupations would only lead to with an impact Il-acturc was found within I 111 of a higllcr LDI values, being a ubiquity indcx, and would utilized flak in Area R. The possibility that a meat not bias the indcx to jtnply low levels ofmobility. The pickage had been proccsscd here is likely. low LDI is seen to bc indicative of decrcascd mobility If the procurement of raw material is usually during the period of sitc occupation. embedded in thc collection ofsubsistenccitem inlow- It is exceedingly remote that intrasitc differences energy groups (Rinford I977), then the tnobility in the assemblage compositions olTrojan Hill could stratcgies ofthese groups would affect the encounter of have arisen through noncultural transfonnatioa raw materials. Data from Trojan Hill indicate that processes, field methods, or contemporaneous, spatially subsistence activities had occurred011this "quarry" site. contiguous activities. Artifact scattersform in modules Procurcnlcnt of lithics wcrc apparently emboddcd ar1d these :tbut one another spatially ratherthan ovcrlap within subsistence itcm rtcquisitioll or processing. ((hnilli 19x9). Disaggrcgating the LA 29362 artifact Forty-ninc pieces ofutilized debitagc were recovered. scatter, by the occupational episodes and organizational Virtually all of these had cdgc anglcs indicative of variability, allows for greaterinterpretive potentid in heavy-duty cutting or scraping.Bifacial cores and the acldrcssing the issucs of prehistoric economic systems. lunate orange peel flakesbore evidcncc ofpossible LISC- A possibly related component of this system is thc wcar RS well. nearby Globe phase site, LA 19363, to be treated later. '3 6 C'ARLSBAD RELIEF ROUTE DESCRIPTION OF CHIPPED STONE DATA-MACHO DUNES

Geologic-Cultural Depositionof Materials = 22.83 mm, sd = 15.28). This very likely reflects differences in the sizes of the parent cobbles and The lithic debitage assemblage of Macho Duneswas aspects of material quality. Chert cobbles tcnd tohe comprised of 1,230 picccs of dcbris, 75 percent of smaller andhave more consistent fracturing propcrtics, which was in tllc nonspecific chert class. Quartzite as discussed in the previous lithic analyses ofTro.jrtn comprised 20.8 percent ofthe assemblage and siltstone Hill. Descriptive statistics regardinglength, width, and 2.3 percent. The remaining 1.9 percent of the thickness by material types are foundin Tables 56, 57, asscmblagc encompassed small quantitiesofboth finc- and 58. grained and coarse-grained materials(Table 55). The Of note, the mean length values for whole lithic materials appear to be of a local origin. The quartzite and chert flakes are lowcr atthis site (22.83 variation in the cherts appears to be similar to that mm, 15.78 mm, respectively) thanthey arc at Trojan which is subsumed withill the sample population from Hill (26.75mm, 2 I, 16 mtn). Matcrials had apparently the nearby TrojanHill site. The quartzite and siltstone been brought to MachoDunes in a reduced statc. also appear to be visually identical to that on the aforementioned hilltop site. These materialsmay have Material Type-Temporal and Mobility becn transportcd from that localc to Macho Dunes, but lnlplications that is dirlicult to ascertain. The materials arc, however, cvidcrltly derived fromthe same geological formation. As pcrfonncd in the Trojan Hill analyscs, thc lithic As in the assemblage from Trojan Hill, virtually all diversity index for this assemblage was tabulated, with materials with cortexare waterworn (98.6 percent). a sample sizeof200. The Macho Dunes'[,Dl equaled All of thc lithic dcbrison Macho Dunes had been 0.94. This value coincides well with the tncan LDI culturally transportedto this locale. Althoughthe two value obtained for "latc" Mesilla phase sites in the sitcs trcntcd in lhis report arc inclose proximity, they Tularosa Basin of 0.90 (Chrmichael I9M: 104). Thc occupy different geological strata. MachoDLIIXS is tcmporal implicationsof the LDI, in this case, arc well situated on the Mcscalcro Pediment and is supportcd through radiocarbon determinationsof thc characterized by a lack oflitlzic material. site's i'caturcs. The lithic diversity index was primarily designed Table 55. Lithic Material from LA 29363 as an aid in identifying prcccramic components.It was soon noted that this index might also function in measuring the relativc tnobilityof adaptive systems. Pwccnt Carmichael (I 986: 187) suggested that the progressivc Chert 75.0 75.0 dccrcasc in LDI values through timc reflected Tecovas chert decreasing mobility of the groups producing the sitcs. Sntl Andre& The deposited by-products created through resharpening oi'curatcd toolsby highlymobile peoples would contribute to high LDI values due to the juxtaposition of materialscoI lected in dilfcrcat arcas. Rcduced mobility or reduced curation ol'bifaccs wonld lead to low LDI values. Thc LDT obtained from the Macho Dunes assemblage, as well as corroborating tcmporal evidence, appearsto reflect this rclativclylow level of mobility. A dcgrec of sedentisrn is noted [or the site by the habitation structures andsmall storage Core Flakc Material-happing Properties features, and it appears that r-sclectcd flora were utilized, which would indicatc that a "collector" Only the material types of chert and quartzite arc strategy is represented rather than a more nomadic represcntcd by at lcast ten whole flake specimens (n= "forager" pattern. The lithic and feature data are 271 and 270, respectively). Chert has a lowcr incan corroborative in this rt,:p'4 cct. length and smallcr standard deviation (sd) than quartzite (chert: x = 15.78 mm,sd = 9.76; quartzite: x 59). The chi-squarc test rcvcalcd an association sigtlilicant at the 2.5 percent Icvcl. The flake portion that contributed the most infonnationto the chi-square statislic was the distal portion Lion1 the Macho Dunes site. The corrcspondingpositive deviation indicatcsthat more distal flakes are cncountered here than would normally be expected. Sullivan and Rozcn(198s) slate that debitage asselublages produccd during tool manulhcture generally yield high percentagesor distal 'I'ahle 57. Wholc Flake Width(mni) and proximal flake fragments, whereas core reduction produces more wholc flakes. The second largest contributor to the total ch- square lies in the lateral flake class, where therearc less than would normallybe expected liom Macho Dunes. The preponderance of lateral flakes at Trojan Hill, Inany of'which appcar to be lunate orange peel Ilakes, has already been discussedin thc previous andyscs of 'Trojan Hill. These flakes arc correlated with initial Table 58. Whole Flake Thickness (mm) stages of reduction. (Coinciding with that, we iind more wholc flakes than expectedat Trojan Hill as well, also related to primary reduction.) The medial portio11 at Macho Duncs is thc third tnost significant contributor to the total chi-square value. The corresponding positive deviation indicatcs that more of thcsc flakesarc found at MachoDuncs than would normallybe expected. A higher proportion of broken flakes is likely related lo the tnechanjcal Corc Flakes"Uchnviora1 Implications failurc of' thin flakcs that separate into several pieccs during bifilcc or tool manufacture (Spcth 1975: 13). It appears that Ihc relatively greater proportionof lateral flakes, often OPFs, and whole flakes, is indicative of A contingency table contrasts thecore flake portiolls of core reduction beingofprimary importance on Trojan tllc assemblagc from Macho Duncs withthat 01 the Hill, as ample evidence hasalready indicated, whereas nearby Trqjan Hill site. These two sites appear to be tllc rclatively grcater proportionof broken flakcs from functionally distinct, although theymay have operated Macho Dunes may implya greater emphasison tool as components of it single system. A chi-square tcst manufacturc andlate-stage rcduction. was performed to investigate whether any association existed bctwccn the flake portions and thesitcs (Table

9s CARLSBAD RELIEF RWTE Table 60. Flake Portion (Chcrt)/SiteAssociations

Contribulion I A29362 0.340933677 0.034ROh.379 0.801 804076 0.437598307 0.2451378461.Xh028033.5 LA29363 2.276848953 0,232446581 5.354hhR347 1.03709K352 2.Y22401458 12.42346419 2.h17782631 6,156472423 0.26725296I.RX223h24R 3.3h0000264 14.283745

Signilkant nt the I percent level

A similar analysis was conducted where thetnaterial similar materials mightbe encountered clscwherc in the was restricted to chert in order to verify that the vicinity that would be dcrivcd from the same geologic patterns observed werc nota manifestation ofdiffcring formation and would have undergone similar material proportions011 the two sites (Table 60). Once depositional processes. Thesc parcnt cobbles would again the chi-square value indicateda high probability then likely be of similar dimensions to the cobbles of association bctwccn flakc portions and the sites found on Trojan Hill. So, cvcn ifthe materials were not (significant at the 1 perccnt levcl); similar patterns were transported from Trqjan IIill to Macho Dunes, the observed in the data. More medial flakes than expected diffcrcnccs in thcir thickness, length, and width wcrc cncountercd in the Macho Dunes assemblage, distributions should indicate that a latcr stagc of more lateral and whole flakes than cxpcctcd were reduction is represcntcd at Macho Dunes. As noted encountered in the Trojan Hill assemblage. Thc prcviously, casual observation and corltrollcd implications, again, of these associationsmay be that cxpcriments have indicated that wastc flakcs more tool manufacturing or late-stage reduction was systelnatically dccrcasc in size from initjal to final occurring at Macho Dunes, as opposed tornore core stitgcs of lnanufachlre (Newcomer I97 1 ). The rrtnkcd reduction at Trojan Hill. continuous data, of flakc length, width and thickness, along with the categorical data of flake portion, corroborate infcrcnccs that core reduction was not a Fluke Dimensions (MdwDzrnes Twjjun Hill) versus principal focus of activity at this site. Later stagos of reduction or tool manufachlre wcrc lhc focus of thc Ir the higher proportion of broken flakes at Macho lithic industry on MachoDuncs. Dunes is related to mechanicalfailure ofthinncr flakcs, as Speth( 1975: 13) would postulate, then a significant Table 61. Chert Flake Thickness differcncc in thickness of flakcs should bc discernible hctwccn the two sitcs. To invcstigate this, a Kolmogorov-Sminlov goodness-of-fittest was run on thc chcrt whole-flake tllickncss distributions (Table 6 I ), As cxpcctcd, a significant dill'crcncc (.01 percent level) was discerned. Furthertests were also run onthc length and width distributions of thcsc chcrt wholc- flakes (Tables 62 and 63), which again evidenced statistically significant differences betweenthe two site 30 99,s populations, at the same significance lcvcl (0.01 35 I I 9Y.Y percent). The data indicate that chcrt hkos from Macho Dunes are thinner, shorter, and narrower, than chert flakes from Trojan Hill. The lithic material from Macho Dunesis visually KS value significant a1 tl~c.01 pcrcenr level ideniical to that found on Trqjan Hill. Nevertheless,

CHIPPED SlONE DA TA "MACHO D CINES 99 Table 62. Chert Flake Length rI 120

I IOU

80

30 91.1 83.9 35 94.8 90.0 40 YS.0 91.8 45 97.4 96.8 50 99.3 98.3 r 20

0 4- 0 10 20 30 40 50 60 70 RO 90 100

dorsal curlex

Figure 80. Ogive of dorsal cortex distrihutiorz by site. 'Iahle 63. Chcrt Flake Width products as opposed to thosc produccdthough primary core reduction wouldbc expected to have less cortex.) Widlh (mtn) Macho Dunw 'I'ro,ian Hill Percrnl rtwent The grealcst differcncc between the two sitcs is in thc 5 10.0 I .4 "0 percent dorsal cortex class"; 40.2 percent of the I I 10 I 46 L7 I 21.6 chcrt flake assemblage from Trojan Hill lies withinthat 15 65.7 48.7 class, as opposed to 72.8 perccnt of the chcrt flaltc assemblagc from Macho Dunes. A Kolmogorov- ZU 80.4 (39.9 '2 5 Yl.1 83.4 Stninlov test furlhcr indicated that the differcncc between these populations significant (0.0I perccnt 3 0 95.6 91.(i was level). A differencc such as that would only occurby 98.9 95.n 35 chancc in 1 trial in 1,000. 40 08.9 Y7.6 Strong evidence is advanced that, although the 45 YY,O Y8.9 same lithic materials were utilized, thc flake populations from tho two sites arc truly distinct. This is

Table 64, Cumulative Percentagesof Chert Flakes in Cortical Classesby Sitr

Cortex 'l'rn/an Hill Macho Dunes 0 I 40.2 I 72.8 As further prooflhal the industrics ol'the two sitcs had I 10 I 47.0 I 77.4 I different emphascs,the dorsal cortex percentagesofthe 20 80.2 53.0 flakc populations wcrc cxamined. Thc ogivesof thc 30 57.8 83. I dorsal cortex distributions from Trojan IIill indicated that all materials firom that site wcrc cqually workcd. This was not thc caseon Macho Dunes, as will bc discussed. This analysiswas thus restricted to the chcrt material class. Observation ofthe cutnulativc pcrccntageof each silc's distribution, in regard to cortex class (Tahlc 64 :md the accompanying ogivc,Figs. 80 and 8 I ), rcvcal that Macho Duneshas a proportionately greatcr amount offlakes with minimalcortcx. (Tool manulhcturing by- well illustrated by the dorsal cortexdata, ;IS well as by the data concerning flake dimensions and flakc portions.

An investigationof the main extantplatfornl lypcs had previously been undertaken on thc chcrt flakes fro111 Trojan Hill. That analysis could find no association

- = Area B between the platform types and the artifact clusters - - =Area C within that site. A similar analysis was undertaken "" = Area F between thesetwo related siks, and in this instance the probability of an association existing betweenthe flalcc platform types and the sites was significant (0.05 percent level; chi-squarc value 5 1.34, 2 = dfj. Thc valucs from Table65 indicated that thcrc wcrc 2ol0 fcwer 0 10 20 30 40 50 60 70 80 90 10 cortical platforms and morc singlc-facetplatforrns than dorsal cortex would normally bc cxpcctcd at Macho Dunes. This evidence again points to a later stage of reduction cxxxrring al Macho Dunes than atTro.jan 1 lill.

Table 65. Platform l'ype/Site Associations

Significant lllc .OS pcrccn~IcvcI Intrasite Assemblage Comparisolls further manipulation and subdivision of the datasets. Nonetheless, thcsc excavation areas do display The lithic data amply demonstrate intersite differenccs differences in their components, which, together with between MachoDuncs and TrojanHill. Although lithic the lithic data may aid in understandingthc organi- materials are ovcrwhclmingly local in both instances, zational and occupationalvariability orthe site. clcar difrcrcnccs arc seen inthe characteristics of the assemblages, as outlined above. These aspectsrclatc to Core Flakes reduction stages and implicitly to site functions. A further attemptwill now be made to examine the spatial and organizationalvariability ~lithinthe Macho Duncs site. Areas choscn for block excavation were whcrc As an initial step in examining the spatialvariability cither topographic featuresor surface manifestationsol' within thc site, a contingency table was set up that cultural materials indicated a likelihood of contrasted the assemblages of the chcrt flake portions encountering prehistoric cultural remains. Theseblock by the six areas. A chi-square test was performedon excavation areas arc thosc used in this intrasitc this data, which was significant (2.5 percent level), assemblage analyses. Timc constraints did not allow indicating an association existing bctwccn the portions Table 66. Churt Flake PortidArea Associations

Signllicnnt at the 2.5 pcrwnl lcvel ofthe tlakes and thc different sitc arcas (Table66). The ...... + .A 350 flake portionthat contributed thc most information to II I thc total chi-square value was the wholellakc category. Arca F and Area B contributed almost exclusivelyto the chi-square information within this class. Arca F 5. exhibited a lack of these flakes whereas Arca R exhibited moreof thcsc flakes than would norrnallybc +"".+"".+"".+"---+-----i-----t---.~,-+.l~~1 'C cxpected. -380 760 -140 -20 100 220 340 m0 rnncip;,lA~,~ I

+ ...... Flukc Length I Chert flake length data havc bccn weakened somewhat by reducing it to catcgoricalform for importationinto a chi-square contingency table ('Table 67). The data have been grouped into interval clnsscs (less than or equal to 1 cm, 1. I cnl to 2 cm, 2.1 cm to3 cm, grcatcr than 3 cm) and examined with regardto thc sile arcas. Patterns wereagain obscrved which mirrored those of the other data sets. The chi-square value obtained through the tablc was significant (0.05 perccnt Icvcl), indicating a probability of an association existing 520 bctwccn length class and sitc area. 'The area that contributed the most inronnation was AreaB, with negative deviations in the smallest length class and Figure 82. Scatter plot of cnrresponclence analysis positive deviations in the largest flake length class. jhkc length dura. Area k', the sccond most important area, only had a

102 C'ARLSKAD RELIEF ROUTE positive deviation in the smallest flakelength class. In statistically significant. A Koltnclgorov-Sminlov test other words, there were more large flakesthan would indicates that this difference is signilicant at the 5 normally be cxpcctcd li.om Area I3 md morc small percent level (Table 69). flakcs than would normally bc cxpccted fromArea F. The dirfcrcncc in the flake length populations These same patterns arc disccmiblc in the graphic (wholc chert llakcs) bctwccn AreaH and Area F likely display of output from thc correspondence analysis reflects aspects ofthe behavior of thc site's inhabitants (Fig. X2), which wasperformed on the length data. In use of the landscape. Both arcas may exhibitpalimpsest the correspondence analysis of chcrl wholc flake occupations; ncvcrthcless, prehistoric bchaviors length, 84.9 percent of the total inertia was rcflcctcd in focused in Area F wcrc evidentlyrelated to the sniall the first principal axis. Along this axis, the variable features (hearths, small pits, roasting features). No "less than or equal to 1 cm" has thc lowest value (-278) evidcncc of any large-scale feature wasround in this and plotsat thc cxtrcmclcrt orthc graphic, thevariable area; no large storage featuresor structures were found. "grcatcr than 3 cm" has the highest valuc(728) and Activity here was apparcntly focused around the plots at the extreme right of the graphic. thermal features. The relativepreponderance of small Considering the excavation arcas, it is observed flakcs in Area F may reflect either thcir suitabilityfor that the twolargest absolute contribution values to the foodprcparation tasksor later stagc flint knapping/tool first axis are from AreaR and Area F (579,233). These maintenance activities ccnteredaround these thcnnal are the most noteworthyareas rcgardirlg flakc length. features. The other areashave low valucs and do not contribute Area E, in contrast, exhibited cvidcncc of Inore that much infonnation. In term ofthe variable space, varicd activities. Here habitation structures were found, we obscwc that Area F plots to the extreme Icft, and along with other fcaturctypes, and virtuallyall of the Area R to the extremc right in Pigurc X2 (Table 68). ground stone artifacts from the cntirc siteand most of Area F is dominatcd by smallflakes, Area K by large the cerarnics wcrc derivcd from this locale. A widcr flakes. range or activity rcgimens appearsto have occurrcd in Of thc chert wholc flake population from Area B, Arca B than in Area F. The lithic data support these 18.2 pcrccntis subsumed within the "less than or equal inferences from fcalurc typc and artifact type divcrsily. to 1 cnl" length class whereas45.2 pcrcent ofthc chert Less spccializcd activity regimens arc m evidence whole flake population from Area F is within that around the mainhabitation locus. Area B. length class. The difrwencc in thcsc populations is

Table 67. Length Class (Whole Chert Flakes)/AreaAssociations

Area A Area B Arcn C Area I) Area E Arctl F How 'l'ofal Cbi- Square

I Value "" . Observed 0-10 wn 10-20 111111 63 20-30 inn1 22 +2 I 14 4 4 61 I49 24.68325041 60.291 87396 26.S04 14504 64.7396351 (> h.2719734h6 15.3?006633 3.540630182 5.648424544 hl I40 1.316749585 7.7n81xmh 2.4')5854063 -1.73Y635158 -4.271913406 -1.320066335 -4.648424544 0.4593608 I8 -4.648424544 0.0702431~8 0.~8545~617 o.mo30hm 0.04m6178 2.YOY731266 0.11374462 0.059509737 2.4984724X3 3.274604R44 3.644422897

I 1 I I Table 68. Correspondence Analysisof Flake Length Data Decompositionof Inertia by Principal Axis

Units Related to First Two Principal Axes

Vari>ll>lesRelatcd to First Two Principal Axes

KtY: Quality quality of reprcscntation Mass mass of thepoint Inertia incrtia in full spacc Cor dative contribution Ctr absolute contribution

‘I’able 69. Percentage of Whole Chert Flakes< or perccnt of the inertia) can be explained and displayed = 1 mrn b Site Area through the first two axes. Thc table relating the variablcs to these lirst two principle axes indicatcs that J most of the variablcs are wcll represented on the two dimensional plot. l’hc “0 percent cortex class” is the most wcll-representcd variable point in two- dimensional space (notcd through high-quality and inass values). This variahlc also has largcst the relativc contribution to the first principal axis. The largcst spatial separation of variable points alongthis axis in two-dimensional space is bctween low dorsal cortex cover groups (20 perccnt, 0 percent, 10 pcrcent) and high dorsal cortex cover groups (X0 pcrcent and 90 pcrccnt). This indicatcs thata plot of the units (i.e., cxcavation areas) along the first axis should discriminate fairly well bctween the asscmblages that Dor:snl C’O~~lCX could hc classed as highly or minimally cortical. Thc largcst spatial separation of units along thefirst axis is ln order to gain an appreciationor thc similarities or bctween Arca B (+359) and hrcas C and 1; (-267,-202). tho subasscmblages with rcgard to dorsal cortex cover, Sincc distancesin thc multidimensional spacc ofa a correspondencc analysis was conduckd on the chert correspondencc analysis arc arrived at rhrough flakes (‘l’ablc 70). This technique was chosen due thc to calculation of a chi-square metric, an examination of simplicity of thc data rnanjpulation, the ease of thc thc related chi square contingcncy tablc was intcrprchtion, andtime constraints. undertaken. This table (Table 71) has conjoined The table of decomposition of the inertia by the variable groups in order tomakc the sparse tablcmore prirlcipalaxesrevealstllatmostofthevariabiliry(70.32 robust. Thc chi square valuc arrived at through this

104 CARLSBAD RELIEF ROUTE table (27.21, I5 dl) is significant (5 percent Icvcl) minimal dorsal cortex cover. indicating that thcrc is an association existing bctwccn Additionally, Kolmogorov-SmirIlov goodness-(>[- the cortical groups and theexcavation arcas. The area fit tests were performed on the Area €3, C md F thal contributes themost information to the chi-square distributions (Table 72), which had appeared value is Area H, which exhibits the largest tablc noteworthy from the correspondence and chi-square contribution in the 90 pcrccnt to 100 percent cortical analyses (Table 71, Fig. 83). Statistically, thc category, corresponding with a positive deviation from difikrences between Area B and thc other two the expected valuehcrc. The next largestcontributing assemblages are significant: Area H vs. Area C (5 area is Arca I:, which has negative deviationsin all percent level ), Area B vs. Area F (2.5 pcrccnt level). classes cxccpt for the0 percent to 20 pcrccnt cortical No significant diffcrence wasnoted bctwcen the dorsal class, where it hasa high positivc deviation.This samc cortex distributions between Arcas C and F, only pattern is observed for Arca C. The chi-square tablc, betwccn these areas andArea E. The ogives of Arcas related to the correspondence analysistables and plots, 13, C, and F (Fig. X I) provide a graphic displayofthcse clarifies the observed relationships bctwcen the areas. data which highlight the above findings. Area R is The Area R chert flake assemblage is more associated distinct fromArea F and Area C. The implicationsof with highly cortical flakes than thatof Area C or Area this distinctness will be treated shortly. F. These two lattcr areas have assemblages with

'lable 70. Correspondcnce Analysisof Cortcx Data Decomposition of Inertia by Principal Axes

Principal Axis inertia Pcrccntagc Cumulativc Pcrcentaxe I 0.046202 SO hi, 5O.hh 2 0.017Y30 I Y .60 70.32 3 0.0 I4405 15.74 N.1 1 4 0.00Y621 10.5s Ph.h(i 5 0.003044 3.34 100.00 0.04 1203

Units Rclatcd to First '1.~0Principal Axos

Variablcs Rclated to First Two Principal Axes

"KEY: Qu;ility quality ofrepresentation Cor relalive contribution Mass mass of the point Ctr absolute contribution Inertia inertia in full spaco

CIIIPPED STONE DA7'A"MACHO DUNES IO5 Table 71. Chert Flake Cortex/Area Associations

Significant at thc 5 pcrcent level

Table 72. Cumulative Percentagesof Chert Flakes in Cortical Classes, by Area .D

.E Cortex pcrccnt percenl pcrccnt 0 (13.4 82 7'4.1 10 hl.2 85.2 84.0

4 -1x 78.6 02. I 7 0 80.2 Yl.8 94.7. 80 84.7 91.8 '45 9 0 90.1 91.8 96.4 100 I00 ... IO0

1 70% An exllmination wasundertaken on the subasselnblagcs within Macho Dunes with regard lo the categorical data -60% ol' platform type. A chi-square analysis contrasting + ...... +. .+...... + ...... + .....+ ...... +-1100 -100 EO 220 3% IUU -200 F~#nc#OrlAx85 1 54U plailonn types (cortical, singlc facet, and multifacct) Figure 83. Scatter plot of correspmdence ana[ysis of cortical data. 106 CARLSBAD RELIEF ROUTE Tahlc 73. Platform Type/Area Associations

93-. .. ".I . . 41 I 85 -30.4 I E 184817 I 13.6308UO05 I ?R.2591623 17.25795812 26,59685864 127 43-091 18.9973822 24.09424084 37.06X06283 117 62304 I". . . I 39.88481675 IS.989528X I 8.371727744 I 17.3S6020')J 1n.t,17sn1n5 16.33507853 78

I I I

Nor signiticant at the IO percent level with site areas was performed(Tahlc 73).This analysis could discern no association bchveen the mjor platfom typcs and the sitc areas. By extrapolation, no area had an association related to a reduction stagc, as could be evidenced through the variablc plallonn type. As previously detailed, the site as a wholc does not have a lithic assemblage with an emphasis on prirnary- stage lithic rcduction. Nor docs it appearto have had an emphasis on very latc stages of reduction, as rnultifacctcdplatl'oltns wcrcnothighlyrepresented, nor were cxhaustcd tools found. This appearsto hold tn~c for all ofthc individual site areas. Variations within the lithic subassemblages of Macho Dunes appear to represent diffcrenccs ofdcgrcc ratherthan differenccs rclatcd to major stages in reduction bctwccn thc site arcas.

A correspondence analysis was clnployed to initially investigate the relationshipshctwccn the chert flake popl~lalionsof thc sitc arcas withregard to platform width (Fig.84). The datawere weakened somcwhatby reducing thc continuous data to a categorical form. Nonclhclcss, evcn with this reduction of the strength of Figure 84, Scatter plot of correspondence unalysis the data, notablepatterns are evident. of plat$wm width. The firsttwo principal axes in this analysis contain 79.53 percent ofthe total inertia, indicating that most of relative nor absolute contributions to this axis, nor do the variation in thedata set will be well represented on they contributc as much inertia to the fdl variablc two-dimensional plots.Thc tablerclating the variables space. Simply, the first axis displays the most to the axesindicates that the variablcs contributing the infbnnationregarding platform width. Thc largest flake most infmnatim to the first principal axis are theless platforms plot to the extremeleft ofthc graphic display than or qual lo 1 inm class andthc greater than or (highest negative values alongthe x-axis), the smallest equal to 5 1nm class (thc most cxtreme classes). The platforms plot to the extreme right (highcst positive intenncdiatc class widths do not exhibit such large values alongthe x-axis).

CHIPPED STONEDATA-MA CHO DUNES 107 Area I3 has a very high negative x-axis coordinate Area C: with rcgard tothcir cumulative pcrccntages in in relation io thc other units (-517). It is spatially the cach platfbmwidth class witha Kolmogorov-Smirnov most scparatcd along this axisfrom the othcr points, goodness-of-f3 test (Tablc 75) we see that the prtttcnls and is relatcd to platform widths greaterthan or cqual observed in thc correspondcncc analysisplotsare valid. to 5 mm. Arca F plots at thc other extremc of the No statistical significancc is noted betwecn AreaC: and graphic display (+275), more closely rclated to Area t;, as in the cortex data.Statistical significance is platfonn widths less than or cqualto I mm, thc most noted, howcvcr, between thc distributions of Area B different from Arca B along this axis. and AreaC: (2.5 percent level) and betweenArca B and Thc variable that contributes the most information Area F (.0.01 percent lcvcl). lo chc second axis is thc 1.1 mmn-2.0 mm class, as can Arca B's distinctncss, especially in rcgard to Arcas bc observed from thc data on Tablc 74 and its C and F, is highlighicd throughthesc invcstigations of accompanying graphic. The spatial unit thai lics platform width. Strikingly, thew data mirror those of farthest away from the other variables along this isaxis thc flake portiondata, the flake dimcnsion data, and the Area C ("454). cortex data. Clcar patterns arc cvident. Examining thedistributions ofAreaB. ArcaF, and

Tahlr 74. Correspondrnce Analysis of Platform Width Data Decomposition of Inertia by Principal Axes

Principal Inrrtia Perccntagr Cumolativc Axis Prrcenlagr 0.08671 2 58.75 0.030bh3 7Y.53 0.020048 13.58 03.1 1 0.009 149 99.31 0.001015 0.6Y Ioo.00

Units Related to First 'Two Principal Axes

Variables Rclatcdto First 'Two Principal Axes

KEY : Quality quality ol'rcprcscntation Mass IIIBSS of thc point Inertia incrtia in full space C: o r relativc contribution Ctr absolute contribution

108 CARLSKA13 KELTEF ROUTE Table 75. Cumulative Percentages of Chert Flakes ccntral hcnrth, exhibits similar ratio patterning. Again, in Platform Width Classes, by Area this may be indicative ol‘ more rounded activities focused around living quarters. A chi-square contingency table (‘I’ablc 77), with counts orchert and quartzite angular debris and flakes, indicates an association existing between thc classcs and the excavation areas(0.05 percent lcvcl).The most significant row variable is quartzite angular debris, the most important column variable is Area H. Area H exhibits morequartzite shatter than would nonnally be expected, having a very high positive deviationfrom the cxpcctcd value. The next most important row variable is chert flake, which exhibits the highest positive deviation inArea F, closely followedby Area E. The chi-squaredata support theinitial inspection of ‘I’hc main lithic material type that was utilized the above ratio data and the conclusions drawn from throughout the site was non-specific chert. Area B, them, notably concerning Area B. Area F, and Area E. however exhibited the lowest percentagc ol‘ this material type. Area B’s artifact assemblage was Corcs comprised of 64.1 percent of this material type. The other areashad chert pcrcentages hctwccn73.1 percent Twenty-three coresin total wcrc obtainedfrom thesite. and XS.3 percent (Area C, with a well-defined No area was devoid of a core (Fig. 85). The areawith structure, had the second lowest percentage,73.1 ), The the most corcs wasArea B (n = I)).Eight of these were Area B subassemblage had the highest percentageof chert. None ofthese wereclassed as either unifacial or quartzite material: 30.2 percent. Evidently, tasksin this bifacial cores and hence could not be conceived as area did not require fine-grainedmaterial as much as in blanks. Of the three corcs found in AreaF, all of which other site areas. Thisagain points to more rounded were chert, one was classed as unifacial oneand was activity regimens occurringin this locale. classed as bifacial. These charactcristics imply a planned strategy of flake removal and they could ShatterlAngulnr Debris conceivably be designated as blanks. This again ties into the interpretation that limited, later-stage lithic Ratios of shatter to flakes byexcavation arcas arc reduction occurredin the Arca F locale, focused around presented in Table76 for all material types, for chert, small, thermal I‘caturcs. Less specialized lithic reduction occurred in the vicinity of the households, and for quartzite.Arca 1: exhibits a very high flake ratio in boththe conjoined material group andthe chert. Arca E. Less specialized lithic reduction occurredhere, This islikely due to an emphasis ontool maintenance with less planning of flake removal evident on the or later-stage reduction focused around the thermal cores. fcaturcs, as previouslyindicated by small whole-flake length data. Area E exhibits similarly high ratiosin all Table 76. Angular Debris to Flakc Katio classes. This may be due to the limited activity evidenced hereas well, which is similar to that ofArea I All Material I Chcrt I Qt~arrzire Area A I 12.7 I 125 I I:2.2 F. Area I? is somewhat spatially separated, presents a later radiocarbon dateon its singlethermal feature than thosc or most of the dated feahlres of the main sitc locus, and likely reprcscnts an ephemeral campsite Area D 1:2.8 1 :?.

CHIPPED STONE DAI’A-MACHO DUNES 109 Table 77. Angular Debris and FlakeIArea Associations

Area A Area R Arcs E Value 23 131 44

I067

10.18431584 I 43.02155576 15.14245548 I ~5.52108716

-S.l2371 134 -2.001874414 r 0,339457238 2.406386468 7.89513075Y 3.12~149013 3.074079311 16.260020~4 4.470578391 I.Y3472%288 25.23537715

4.378390372 0.061 574119 1.533103273 0.133575027 10.08638463

8.178810483 2.25400533 9.4h8287915 1 8.053769093 I I I 5Y.482918 I I wI I I Significant at the .05 perccnt lcvel

Fig. Mi). On the tlooror this same structure a highly C.hnclusions modified projectile point was encountered. The modified shapeof this artifact implies that it no longer lntrasite Assrmblage Vuriabiliy served its original functionbut was likely intcndcd for scraping activities. Macho Dunes hadspatially separated artihct modules likely related to differcnt occupational episodes in many ofthe cases, and often expressing organizational variability. The simple angular debris to Ilake ratio within the major material types has highlighted thal Area F had a preponderance ofchcrt flakes as opposed to angular debris ofany material. Considering the grcat number ofthermal fcaturcshere, thismay be indicativc of tool mailltenartcc or late stage reduction occurring around hearths. Area E had a similar ratio to that obtained fromArea F. This arca hadonly one thermal featurc and probably representsan ephemeral campsite locale, where a similar stageof lithic reduction was reprcscntcd. In conlrast, Area E, which had well- defined habitation structurcs had the lowest angular debris to flake ratio. Area C also had a well-defined The remaining two lithic tools cncountcred onsite structure and manifested a similarratio of shatter to were derived from Arca F. These were a latc-stage that found in Area H. A signifkant chi-square value bifacc ol'chcrt, anda middle-stage bifacc of siltstone. had corroborated that evidence and examination of the Ncithcr of these exhibited evidenceof use-wear. contingency table had indicated a preponderance of quartzite angular debrisin Area B. This coarscr grained material had the lowest replacement cost and was probably suitable to the more general site tasks, focused around theliving quarters. Arca B was the area with the most cores on site. No unidirectional or bifacial cores were cncountcrcd thcrc. Strategic rcductionolcorcs docs not appear to be represented here, but a more generalized form of reduction, possibly on an “as needed” basis. Of the thrcc cores cncountcrcd in Area F, one was unidirectional and one bifacial, indicating a more strategic plan of flakc removal, likely rclatcd to the tlmnlal fcaturesthcrc.

The probability of associationan existing between I I the portions of the flakes andthe site areas was also Figure 86. Tools from MUC~~JDunes. significant. Area B again stood apartwith more whole flalccs than would nollnallybc cxpcctcd, as did Area F material may bc sccn to be related to the more with less of these than would nonnally bc cxpcctcd. generalized activities that would occur within the High proportions or whole flakes arc taken to be vicinity of a household. However, with significant related to the earlier stagcs orreduction. differences noted bctwccn their distributionsof cortex Chert, whole flake length, was alsoseen to be cover and platform width, it may bc that occupational rclatcd to site area. Again thepattern indicates more variability is represented. Additional nonlithic evidence advanced reduction occurring in the Arca 1: locale points to the same conclusion. where there was a preponderance of small flakes. A Area A and Area E likely represent ephumcralusc significant difference was noted betwccn this area’s of this landscape. Both areas are focused around a length distribution and thatof Area H. single thcnnal rcaturc, bedrock hearths, and have small Data on flake cortcx cover also noteda significant lithic assemblages dominated by small chert tlakes. diffcrcncc bctwccnArea B and Area F. A significant Statistically significant differences betwecn thcsc diflcrcacc was also discerned between Area I3 and assemblages and those of the other areascould not be Area C:. Paralleling this evidence was the platform detected due to the small sample siLcs. Area D width data of chert flakes. Again significant differences contributed little information to most of the analyses. were sccn bctwccn Area B and Area F and betwccn The area was not especially high in artifact countand Area B and Area C. was of such a large size that multiple prehistoric Area R’s uniqueness is highlightcd through these artifact modules likely obscuredrelations. investigations. A generalized activity rcgimcn is The salicntdata of theseanalyses point toa likely represented through the dcbitagcby a greater breadth functional difference in the use 01 the landscape, in the reduction sequence continuum than in other notably between Area B and Area F. Additionally, the areas. In special contrast is Arca F, with a more narrow significantly different distributionaldata bctwccn Area range ofrcductionrepresented. Thereduction of blanks B and Area C imply a temporal difference betwccn or tool maintenance activities arebehavioral correlates these two habitation loci. Other areas (notablyA and E) of the debitagc fromthis area, where therrnal features also likely represent different occupational episodes arc most manifcst. although the lithic data alone cannot discern this. Palimpsest occupation is probable in Area I?. The simplistic examination of the ma-jor extant Noncthcless, with its distributional data indicating platfonn typcs by area has not shown any valid more narrow, later stage lithic reduction, runctional association. No area was dominatcd by primary implications still seemclear. The debitagc isrelated to reduction (which would be noted by a prcpondcrance the thermal features. Later stage controlled flaking, of cortical platforms) as opposed to very late stage producing smaller piecesof debitagewith little cortcx reduction (where multifaceted platforms would bc present. is represented here, either dueto the cornnzunal more cvidcnt). Generally, thelithic assemblage of the aspect ofknappingaround a fire or due tothe necessity site lies within an intermediate range of the rcduction of small, sharp flakes in foodpreparation tasks rclatcd continuum. Unrctouchcd flakes werelikely thc dcsircd to the features. tool form as there was a lack of othertool types, and The distributional differcnccs bctwccn the only four bifacial reduction tlakcs cncountered. assemblages of AreaB and Area C may be less rclatcd Specialized, diverse tool forms are generally not sccn to functional implicationsthan to temporal ones.Both to bc required by hunter-gatherers engaged in areas represent habitation locales. These areashave the subsistence activitics in low-altitude desert highcst proportions of quartzite material. This crudcr environments (Torrance 1983). This appears to be the

CHIPPED STONE DA TA-MA THO DUNES I I I casc for LA 29363, Macho Dunes. Retouchon lithic thinner flakcs from a 1atcr reduction stage. artifacts was noted in only four instances(the projectile Flake dimensions also statistically corroborated point, the scrapcr, and the two bifaces). Use-wear was this line of evidcncc. The populationof wholc chert apparent on only fourflalcc specimens, and may have flakcs from Macho Dunes was thinner, shorter, and been duc to the lack of heavy-dutyactivitics occurring narrowcr thanthat from Trojan Hill, although the lithic on site. Hunters and gathcrcrs in envirormentssuch as materials appeared visuallyidentical. this arc primarily dependent upon small animalsand The data on cortex distributions again pointed to plants. Nelson ( I99 I :12-25) presents data suggesting the same conclusions, where a significant difference stone tools werc not necessarily required tohunt or was noted between the populations of the two sites. process small game. This postulation, coupled with a Thc Macho Dunes assernblagchad signifkantly fewer low- energy cost of‘ acquiring lithic materials, picccs of chipped stone with cortex present thanthat presupposes to a degrcc that an expcdicnt flake from thc ncarby quarry site, TrojanHill. lcchnology will be most represented. Easily acquired, The simplistic examination of extant platform serviceable lithic materials were wcll within thc types, which could not discriminate an association roraging radiusof this site (for example, Trojan Irill). between types (ilnplicitly related to major reduction It is thus inferred that expedient, unrctouched llakes stages) within the Macho Dunes’ site arcas has, served mosttool needs at this sitc. The lack of visible however, noted asignificant probability ol’association use-wear may indicate noninlcnsiveuse of individual between platform types and the two sites. Macho flakes; subsistence tasksthat might haveutilized lithic Dunes had fewcrcortical platforms than Trojan Hill malerials may not have been intonsivc enough to and more singlc-facetedplatfonnsthan would normally manifest indisputable use-wcar. Additionally, bc expected. O’Connell (1977) noted that stone tools in these In general, the Macho Dunes lithic assemblage cnvironmcnts wereprit-narily formaintenancc activities reprcscnts a noticeablylater stage ofreduction than that rather than for subsistence activities. Subsistence from TrqjanHill, although thelithic materials appeared activities would be most in evidcnce around habitation visually identical. Materials atTrojanHill are gcnerally stnlctures. derived from that locale; those obtained from Macho Maintenance activities, such as the cutting and Dunes had cvidently bccnculturally transported to the shaping of wood, might occur in tnorc direct site. Macho Dunesdid not exhibit evidenceof heavy- association with procurernenlof this material. duty activities occurring on sitc, as could bc cvidenccd Within the site, flakc tools were likely the norm through theI ithic materials. In contrast, a grcat numhcr and reduction seems to have been within the of unidirectionally utilized llakcs were cncountercd intermediate range of the continuum. Contrastingthis with edge angles irldicativc of heavy-duty scraping sitc to the nearby ‘Trojan Hill sitc we have seen activities from the ncarby Trojan Hill sitc. diffcrences. Additionally, quarrying episodes arc in evidence on Tro.jan Hill, which likely formed only one aspectof an cmbcdded resource procurclncnt strategythere. The lithic diversity index for Macho Dunes The examination of’ flakc portions has shown a obtaincd a value that indicated occupation datingto the significant diffcrence betwccnthe sites. Proportionately Globe phase. Radiometric data had substantiatedlhis more wholc flakes, associated with primary reduction, index by placing the habitation loci within the Sam occurred at Trojan Hill, as did lateral flakes, which time period. Apparently, Macho Dunes and possibly again may be related to this stage (OPFs). More the nearby Trojan Hill sitc wcre occupied during the medials were derivedfrom Macho Dunes than would Globe phase. However, temporal association cannotbe normally be expcctcd, likely related to the breakage of determined.

I 12 CARLSBAD RELIEF ROUTE MACHO DUNES GROUND STONE

James A. Quaranta

The ground stone assemblage from Macho Duneswas Grain size was dctcrnlined with the aid of an comprised of 65 artifacts. Fivc sandstone types have AmericaniCanadian Straligraphic Card. Grain size been differentiated, somewhat subjectively, due to the Ihroughout all artifact categories was either fine or presence of various inclusions and the color of thc mcdium-grained, with about two-thirdsof all artil'acts binder. Diffcrcntial degradation due to the clcments or in the fine-grained category.Manos wcrc represented heat and the uptake of coloration from thesoil were in all nlatcrial classcs cxcept for the limestone/ confounding factors. There wasalso some overlap in dolomite. Mctate fragments were representedin all ol' grain sizc catcgorics within thc subjectively arrived at the five sandslonc typcs (Fig.87). In manos there was typcs. Nevertheless, two major types were notedin a 4:l ratio of finc to mcdium-grained materials; in equal proportions. Twenty-four specimens wereof a mctatcs the ratio was slightly less than 2: 1. The somewhat dense reddishsandstonethat obtained much prcpondcrancc of smaller grainsize in mano spccimcns of thc rcd coloration from the binder. This typc appears may have functional implications. Thc finer grained visually idcntical to that which is found in outcrops sandstonc lnatcrials also seem to have greater nearby, most notably at the Mississippi PotashMines compaction ofthe grains andless binder bctwccn them. along U.S. 62/1XO. Within this material typc, thc ratio The implication of this is that grcatcr use life was of spccimcns with finc-sized grains to rnediurn-size possible fromthis material than fromless consolidated grains was 2. I X: I. Twenty-four specimensof anothcr materials. This is important for manos which arc said sandstone typehad a yellow ochre-colored bindcr and to exhibit a muchfastcr wear rate than metates, about appeared somewhat inorc friable. The grain size four timesthat ofmctates. Textureis thus seenLo bc an distribution within this material typc had a ratio of important charactcristic related to the durability or 1 .A2: I in fine-graincd to medium-grained specimens. ground stoncartifacts. Durability of nlano~sccins to bc Small numbers of artifacts were encounteredin the a factor in the selection of thc sandstonctype. otherthree differentiated sandstone typcs. Additionally, Different characteristics are sought in metates quartzite was used to a small extent (n = 2) as was where use-wcar docs not scemto be a prime elemcnt in limcst(~~~c/dololnitc(n = 1). the cxpcnditure ofthese items. It is thc maintcnancc of thcsc items rather thanthe use of thcsc which gcncrally - ." .. appears to remove mctalcs fromtheir original systemic context. Restoring the rough surface of metates, by pecking with a hammcrstonc, seems to have been coInn1on practice, as cvidcnccd through ethnographic rcsearch and the archaeological rccord. It may also have played a role in crcating the very fragmentary nature of many ground stone assemblages suchas this one, as it is at this stagc that most breakage occurs.FS 333, for example, showsa l'uwow of pecking nearand paralleling the fracture line whcrc the artifact broke. Pecking was notcdin many instanceson the surfacesof the grinding stoncs. Duc to the degradationofmuch 01' the sandstone material, and the difficulty of ascertaining pecking on coarser materials, quantification of this techniquc secms unjustified. Nonetheless, pecking wasin cvidcncc on ground stone from this sitc. A higher proportion of the metrites than the manos was comprised of the coarscr matcrials,as prcviously mentioned.Thew coarscr grained materials would not needto be repecked as oftcn and wouldbe

0 3 i/l/ lcss subject to the breakagethat appcars more relatedto maintenance than to the use of thc objccts (Schlanger

MACHO DUNKS GROUND STONE I13 of manos: 47.1 percentcircular, 35.3 percent oval, 5.9 percent subrectangular, 11.9 percent irregular.) No wholc metates were found. All were created from tabular sandstone. It appearedthat 22 of the 32 (69.0 percent) were derived from slabs less thanor equal to 5 cm in thickness. Five appearedto be derived from thickcr slabs, 5 to 10 crn thick; the preform thickness of the remaining five could not be deduced, beyond that they were derived from tabular sandstone. Within mctate fragments,54.5 pcrccnt displayed the same slight production input (<50 percent modification, 33.3 perccnt indeterminate modification, 3.0 percent no modification, 9.1 percent greater than 50.0 pcrccnt modification). Tabular sandstone was available to the site inhabitants and required little production input. Grinding, pccking, and flaking wereutilized to shape the ground stoneartihcts. Although production inputs were low, energy costsof acquisition wcrc obviously high enoughthat materials wcrc substantially recycled. Many artifacts had a secondarily utilized surlhce: 13 artifacts had a secondary use-surfacc firnctionally indicative of a one-hand 111an0, 3 indicative of a Figure 88. Om-hand nzalzns from Macho Dunes. nondifferentiatcd metatc, 1 of a basin metatc, 1 of a I99 1 :462). It is quite likely then that the prehistoric slab metate, and 8 had a secondarily used side of jnhabitants ofthis sile had a predilection forthe coarser indctcrminate hnction. Tour of the seventeen whole grained sandstonesto be used ror their rnctatcs and thc manos had cross sections that were designated as liner grained smdstones fortheir manos. "airfoil shape," which implies substantial use-wear. All manos from Macho Dunes were of B onc- Two wholc manoshad a convex-concave profile, thc handed type (Fig. 88). The majority ofmano specimens side with the concavity derived from the artifact's (40.6 perccnt, n = 13) wcrc dcrived from the reddish primary use as a mctatc beforc fragmentation and sandstone, whichis visually identical to that from the recycling into a onc-hand mano, another had an tnost local outcrop. The mean length, width, and irregular-concave profile for the same reason.At least thickness values for wholc specimens arc as follows: live manos rccovered from the site had evidently 9.67 cm, 1.44 cm, 3.75em. Length of manos hasbcen served as metates beforc fragmentation and recycling. seen to hnvc subsistencecorrelates. IIard's ( 1990:141) Manos and metatesthat are brokcn during maintenance study identified a strong corrclationbetwccn the degree may be stored for some future use, put to some other of agricultural dependence and average manolength. use immediately or kept in use as grinding tools The patterning in his data indicates that the average (Schlangcr 1991 :463). The ground stone assemblage length of manosis smaller than 19 cxn for groups with from Macho Dunes appears to be similar in these low agricultural dependcncc (0-20 percentofthe diet). regards; the ground stone artifacts do not appearto As can be seen from the whole rnano length data lrom have been discarded ilnmedjately after breakage. this site, the mean length is substantially smaller than Ethnographic and experimental evjdcnce suggcst that proposed as the cut off point for lowor absent that one-hand manos are designed for generalized agriculturrd dcpcndcnce. ?'he absence of cconomic pounding work, pigment grinding,and rood-grinding pollen grains or phytoliths noted from the scans or tasks that employ a "short, often circular grinding slidcs dcrived from artifact washes may bc motion with the pressurc exerted into thc body of the substantiating this lnano length data. It is unlikely that mano" (Lancaster1986: 182) as wcll as being used for cultigens werca locus ofthe grindingactivity here and hide working (Adams 1988). Many artifacts bore may not havebeen present at all. striations visible under the binocular microscope Wilhin manos,73.0 percent had productioninputs indicative of this type of movement, in which 76.0 that displayed less than50 percent modification firom percent ofthe striations were classedas crcsccnt-like or their parent forms. Generally, they wereshaped into a circular. Considering only themanos, 8 1.O pcrcent of roundcd or oval plan-view shape without modifying their cross section except throughuse wear. (Plan view

114 THE C'AKLSEAD RELIEF ROUTE thcsc artifacts with visible striationshad cresccntic or circular striations. All manos were of the onc-hand variety and wcrc likely uscd in a rotary or pseudo- rotary rashion as evidenced by the striation patterning, and as Lancastcr (1986) states, is the general motor movcmcnt employed.This striation patterning has been noted as indicative of an emphasis on the grinding of wild foodstuffs as opposed to cultigens. Of thernetatc liagmcnts that had striations, 81.0 percent bore crescent-like or circular striations. The only othor striation pattcming visible on the artifacts was that which was classed as random. These, too, may have been related to a rocking, pscudo-rotary use but oftcn had striations supcrinlposcd so that it was difficult to ascertain the individual stria. 0 3 cm The depths of the grinding arcas of the metatc LA fragments wcrc measured when feasible. The nondifrcrentiatcd metate fragmentshad a mean depth I of I .S 1 cm. The Incan depth of those that had bccn Figure 90, Limonite pigment. designatcd as basin rnetate fragments had a Incan depth 01'2.43 cm. An artifact that had bccn designated as a stone artifact. thrcc morc pieces of mincral pigment lapidary stone had a concavity of 0.9 cm wcrc found, which would likely be classed geologically as limonite. Onc piccc (Fig.91) measured 30-by-20- by- I7 mn. This piece wasfound within the deepbrush structurc in Area €3,wherc thc palette with the piglncnl adhesion was found. It exhibited a 2 mm concavity in a roughly 15 lnm square area; two other arcas on the

Figure 89. Metattj,fragment withlintonitc pigment.

Two hgmcnts,one from a metate and one which had hccn designated as an indeterminatc fragment,had adhesions. On the merate fragment, which had been fhnd in the deepest stnlcturc in Area B, a chunk of 0 5 cm limonite pigment was found (Fig. 89). Opposite - surfaces of this slab had been utilized as H metatc or palettc. Tllc pigment is adheredto the secondarilyuscd side at the impact point offragmentation of the artifact. This item may or may not have been used for the grinding of vegetal nlattcr prcviously, but it is likely that at least by the time of fiagmcntation it had been relegated to nondolnestic duty. It was quite thin and evidently had a precarious potential use-lifeat the time it was uscd for crushing the pigment. Thc white sandstone material of this rnetateipalettc is also the mosl friable and weakest of all the sandstone types (Fig. 90). Figurc 91. Metate-palette from Macho Dunes. In addition to the ochre encountcrcdon the ground

MACHO DUNES GROUND STONE I15 surfacc of this mineral pigment appear to show material. grinding wherethe pigment wasnot gouged but ground Subsistence evidence was sought through an onto a surracc (14-by-13 mn; 7-by-7 mm). examination of the pollen and phytolith assemblages Another chunkofmineral pigment (FS604) found from the surfaces of the grinding stones. Six artifacts just outside of the same structure, appeared slightly had bccn washedto rctricvc pollen and phytoliths from hardcr and had a yellow and black bandedhard rind. the interstices of the grinding surfaces. Thc washing The piecc of pigment measured 43-by-27-by-I3 mm. and collection waspertonncd in-house;the extraction It seemed to have amore rriablc, redder interior that was performed by Texas A 81 M University. The was mostly removed rrom a gouge.It measured 34-by- analysis was pcriormed by Dr. Kichard Holloway of 17-by- 12 mn. Quaternary Services. After extractionof the first two One filrther chunk of mineral pigment was found samples, it was decided to proceed only with the on site (FS 72) and had a slightly different aspcct. It phytolith extraction fromthe remaining l'our samples. seemed hardcr and more orange. Limonite mineral Both the pollen andthe phytolith assemblages wcrc concretions arc found in the soils of this region; judged to be in poor condition and it was deemed however, the close association with features, the unlikely toretrieve countable asscmblages. prcscncc of pigmenton ground stone, and the gouging Thc phytolith assemblages showed evidence of and grillding evident on the minerals themselvesattest deterioration. Due to thisdetcrioration, the lilcclihood to the prehistoric use of thcsc lnilzerals on this site. of obtaining an accurate interpretation was seen to be minimal. No cconomic taxa (i.c., corn or cucurbits) I were seen to be present. The assernblagcs appeared to consist entirely of grass phytolith assemblages. Although this is not quantifieddata, the evidence may be pointing againto a lack ordependence on cultigens and :t focus on wild fbod-stuffs, as both mano length :mi striation patterning seem to indicatc. As mentioned above, virtually all ol' the ground stone artifacts from this site were composed of sandstone. Various sandstones outcropin Eddyand Lea counties. The prcscncc of various types and the recycling of these materials maybe indicative ol' the mobility strategies of the prehistoric site inhabitants. Although sandstones outcrop slightly beyond the foraging radius, material acquisition costs were great enough that substantial recycling of materials was 0 3 cm practiced and materials other than the mostlocal ones were present. 'These facts may indicatc that this site was - part of a seasonal round. Materials from somewhat distant areas apparently arrived at thesite in a recycled form. Additionally, virtually all of the ground stone Thc other ground stone fragment withan adhesion materials on site bore evidenceof cxtcnsive use-wear. was classed as indetenninatc but wasin all likelihood It is exceedingly doubtfulthat a site in this cnvirontnent part of a metate. It appears that this fragment was could bear permanentor long-term human occupation. secondnrily used as a hearth stone, as heat-related Therefore, the extensive use-wearon the ground stone changes in the formof differential reddish coloring and artifacts and the recyclingof thesc may speak to somc crazing appear on the artifact. Camilli et al. (1988% dcgree of site rcoccupation. Grinding stones from 24) report finding mano and metate fragments in regularly reused sites should be more extensively hearths and roasting pits where they evidently shaped than those from sites that were used functioned as heat-retaining elements. Along withthe fortuitously. Mctatcs from rock shcltcrs, forinstance, heat-rclutcd changes,the artifact bore a 1-cm-sq patch appear to be generally unshaped, whereasvirtually all of carbonized resin(Fig. 92). The extant grinding area metate edge fragments fromLA 29363 were shaped on this artifact measures about I cln sq. The adhesion and smoothed. This initial production inpui may be is not on this surface, but ratherit is on a fragmented related to a planned site rcoccupation and intended side of the artihct. Evidently, the mucilagcthat had curation of thesc materials. adhered to the stoneis related to the cookingof vegetal Curation of' ground stone is expected when site matter rather than to the grinding of' any vegetal reuse is anticipatcd. Curation refers to a "strategy of

116 THE CARLSBAD RELIEF ROUTE caring for tools and tool kits that can involve advanced burned, hcat-rclatcd changes did not appear evidenton manufacture, transport, reshaping and cachinge .l.but.l thc artifacts. It would be supposed that the limonite need not include all of these dimensions" (Nelson pigmcnt would be blackened, at least, if it had 1991x52). It appcars that all 01 thcse dimensions are, undergone burning withinthe structure. This was not however, represented from the ground stone the case. Thcse linesof evidence give support to thc assemblage from MachoDuncs. hypothesis that this structure, subsequentto burning, Under conditions ofanticipatcd reuseofa site (Le., was used to cachc ground stone matcrials. Perhaps this part ofa seasonal round), grinding tools wouldbe used method ofcaching materials in a burned context served to the point of being expended. Additionally, ground to mark the cache pit, and combated somewhatthe stone artifactsarc often sccnas sitc furniture (Binford strong eolian processes that could have hidden the 1979; Camilli 1989) and are often in left their systemic materials were they simply lefton the surface. context or cached. Interestingly, muchof the ground Gvidcncc is advanced that supports a planned stone material recovered from this site comes froin rcoccupation of LA 29363, likely as part ofa scasollnl within the pits of the bumcd brush structures. Possibly round. The caveat which mustbc addressedis that the that is because these artifacts were in left their systemic sizc ofobjccts influences their visibilityand hcncc thc context. This may bc true to some dcgrcc of the likelihoodthat they wouldbe scavcngcd prchistorically artihcts encountered in the structure from Area C (5 or taken by modern collcctors, biasing our sample. nrtihcts), and possibly from the trenched structurein Grinding tools are often large enough to bc quite Area B (10 artifacts), where some relations are visible on sites, making them subjectto recycling and obscured due to the trenching. But strikingly, 23 reuse whether this reusc was anticipatedor not. The ground stone artifacts wcrc recovered within thc hypothesis advanced of planned rcoccupation, structure in the northcast corner ofhrca B! This is an nevertheless, has fairly strongsupporl in the contcxtual enonnous quantity to be in systemic context in it data ofthe 23stone artifacts andthe 2 picccs of ground stnrcture that measures less than 3 m in diamctcr. pigment retrieved from thcbrush structure in Area E. Furthermore. although thc structure was evidently

MACHO DUNES GROUND STONE I I7 1 I X TIIE CARLSBAD RELIEF ROUTE MACHO DUNES FAUNA

Susan M. Moga

A varietyof spccics wereidentified in the small faunal Table 78. Taxa Kecovered from LA 29363 assernblage fromLA 29363. The site was dividedinto six excavation unils (Areas A to F), which includes structurcs and katurcs. Thefauna recovered from each of lhcsc units will be described in thetext.

Mcthodology h complete analysis was perf'onncdon the 413 bones recovered from the Carlshad Rclicf Route faunal assemblage. The bone was dry brushed for cleaning and infonnation was recorded by codedvariablcs. The variables included site, field spccimcn (FS) and lot numbers; count, taxon, clement, side and portion, age of thc animal, environmental, animal and thermal alterations, and evidence of human processing or modification of the bone. Idcntilication of the bone was madeby utilizing the OAS comparative collections. Sourccs on New Mcxico fauna (Bailey I97 I; Findley ct al. 1975; Degenhardt et al. 1996) wcrc consulted to determinc which species inhabited the :wa. Descriptions are provided for the fauna reprcscnlcd in the assemblage.

Tuxn Recovered

The bulk ofthe asscrnblagc consists of desertcottontail and woodrals. A variety of species occur in the assemblage ('Table 7X), but some species are rcprcscnkd by only one fragmented element. Each taxon is supplied with a description, common name, count, and minimum numberof individuals (MNI). lndctcrminate Taxa. Unidentiiiable small mammal remains encolnpasscd17.4 percent (n= 72) of the t-umdasscmblagc. The bones wcreeither so minutc or severely fragmcntcd that they could notbe identified to specific spccics and wcrc size categorized. Thc Ruhidac 'True toads b I .5 Mussel Mussel 2 .5 majority of this group, 80.6 percent, wcre mature Total 413 100.0 individuals, and X.4 percent were irnmaturc. Thcsc clcmcnts consist of long bones, blade fragments, one D@odomys cf. ordii (Ord's kangaroo rat). Thc rib, and a few vertebrae. Only six bones were kangaroo rat inhabits a variety of geographiczones and represented in the small to medium, medium, and is extremely common throughoutNew Mcxico. Their medium to large unidentifiable lnalnmal categories. pelage varies with soil color and their physicalsize Spernzophilus vnrieatus (rock squirrel). A increases in the northern and castern partsof lhc state. mature innominate was the only squirrel in the Their breeding season appears to vary belwccn assemblage. These animals prcfkr to inhabit rocky counties (Findley et al. 1975:174-175). Two malurc terrain and nest beneath boulders. Theyrarely hibernate and one juvenilc individuals are reprcscntcd by in southern New Mexico andthe young arc born in portions ofa scapula, tibia, cranium, and mandible. spring (Findleyet al. 1975: 125- 126).

MACHO DUNES FA [JNA I1Y Pevotnyscus sp. (rnicc). New Mexico provides a specific species and wereclassified accordjng to size. ha bitat ror eight typesof P~~~clmq~scu.~.four ofwhich arc These rodents rangcin age from youngto juvenile and distributed throughout Eddy County: Per.om,ysclrs maturc. er.c.micms (cactus mouse), Perom_vsars mnnicuhtus Sylvihgus auduhonii (desert cottontail). The (deer mouse), Pernmyscus lezlropus (white-footed cottontail encompasses thelargest number of bones (n mouse), and I+rom,v,srus hoylii (brush IIIOLISC).The = 151 or 36.A percent) collected in the assemblage. only distinguishing differences between speciesis the Cottontails are distributed throughout the southwest shape and lcngthoftheir tails and pelage color, which and lhcir breeding season varics with increasing often changes with seasons (Findlcy et al. 1975:201- latitude. Gestation period rangesfrom 25 to 35 days, 215). Thc internal skeletal strllchlre is cvcn more with a litter size of approximatcly three lo five difficult to diilkrcntiate,theref'orc, these species were individuals (Chapmanet al. 1982:94). At this extremely recorded as Pc.rwnysru.s .sp. Nine bone hgments: high rate of rcproduction, thc human population is cranial, mandible andlong bones, possibly from three providcd with a renewable meat and pelt resource. individuals, one mature andtwo juvcniles werenoted Approximately four to t?ve malure individua l,,5 two in the assemblage. immature, and one juvenile are represented. A large 0nyr.homys leucogaster (northern grasshopper number (n = 22 or 14.6 percent) of cranial frnglnerlts mouse). Common throughout New Mcxico, the and vertebrae (n = 38 or 25.2 percent) along witha grasshopper mouse is often associatcd with (kt's diversity of other body parts reveala valuable role in kangaroo rat (Findlcy ct al. 1975:227). Fivc bone the subsistcncc practicesofthe local inhabitants. Only frilgmcnts of one mature grasshopper mouse was three boncs exhibit breakage attributed to proccssing, identibed. two are burned, and thirteen provide evidence of Neotoma sp. (woodrats). Two woodrat speciesare animal alterations. known to coexist in the Carlsbadarea, tllc Neotoma Lepus calijhrnicus (hlack-tailed jackrabbit). micropus (Southern plains woodrat) and Neotornrl Jackrabbits arc found throughout New Mexico in ulhigulu (white-throated woodrat), (1:indley et al. regions that propagate succulent vegetation (Findlcyet 1975241 ). The only distioguishing difference bctwccn al. I975:93-94). Thereis also a relationship between thc two specicls is in the first upper molar (MI) latitude and litter size: the lowerthe latitude, the longer (IIoffmeister and De La Torre 1960:477).Some ofthe the breeding season (Chapman 19x2: 129)."l'hcre were woodrats in the assemblage arc quite large in size and fewer jackrabbitsthan cottontails recovered, Only two more robust than the specimens in the OAS immature long bones fromone individual and a variety colnparative colleclion. None of the woodratspecies of bone fragments from one mature individual were arc known to hibernate, and more than one litter is recovered. None of the jackrabbit boncs exhibit raised in a season (Bailey 1971: 180). Five mature processing, animal or thcnnal alterations, but four jndividuals and threejuveniles are represented inthe bones showed environmental changes. asscmblagc. Thirty vertebrae (35.3 percent) were thc Canis lntrans (coyotc). Coyotes are found in a largest numbcr of elements recovered, along with wide variety of habitats in New Mcxico, but most twelvc ( 14.1 percent) cranial fragments, eleven commonly in the grasslands (Findleyet al. 1975:281). mandibles ( 12.9 pcrccnt), several long bones, and A proximal portion of a mature radius was the only various other fraglncntcd elements. Only five (5.9 coyote remnant retrieved. Whether the animal was percent) woodrat bones revealed evidence of animal taken for its pelage, utilized as a food source, or the alt-eralion;four (4.7 percent) exhibited rodent gnawing elclncnt was cawicd to the site by a scavenger is and one (1.2 precent) displayed a carnivore puncture unknown. mark. Canis I~pu.s(gray wolf). Known to have ranged 1:'rethi:on dorsntum (porcupine). Porcupines throughout New Mexicountil the end ofthc nineteenth reside in a1 most everyhabitat in New Mcxico (Findley century, the gray wolf was eliminalcd to protect et al. 1975:273). Breeding season occurs from domestic livestock (Findley et al. 1975:283-284). A September tlzrough November, and birthinghas been complete dna wid1 heavy evidcnce ofroden! gnawing recorded inlate April to early May, and some aslate as was in the assclnblage, but like the coyote radius, its August (Dodge 1982:358-359). Two long bones horn utilization is unknown. Taxiden (badger). nistributed overthestate, 0111: rrlahlre individual and three innominate fragmcnts tuxus from a juvenile were in tllc assemblage. the badgcr is more nulnerous in valleys, which are Roderrtin (rodcnts). Fifteen small rodent boncs inhabited by their prey, ground scluiwcls and prairie and two nicdium to large-sized rodent boncs were dogs. Hibcrnation is unlikely in lllc lower valleys oT ~ollcctcdbut wcrc too fragrneoted to identify to a New Mexico and breeding season occurs in early surnmer, with an offspring rate of four to five young assemblage. Becausea variety of frog and toad species (Bailey 197 1:343-345). With onlythee vertebrae, it is inhabit the area, further identification was impossible. diflicult to determine its usage. Bufonidae (true toads). Four toad species occupy BodRisotz (cattle/hison). A maxillary molar from the Pccos Rivcr drainage:Bz~fi7 cognatus (Great Plains a cow or bison was rccovcred from Lcvcl 1, which toad), Bufo rkhi1i.r (green toad), Bz@ punc~atus(red- consists of a maximum depth of only 10 cm, so the spotted toad), and Bu[n sperinsus (Texas toad). The tooth could have been from a contemporary incident elenlents from this taxa are comparable to the family and was envelopcd in the churned-up soils. Bufonidae, but not to the species level. One vertebrae Altcmativcly. the tooth is historic and is the remain of fragment, thrcc innominate fragments, and two a cow or small-sized bison. The dentitioll of these fragmcntcd lcmurs represent one mature and one species are similar and it is difficult to distinguish inmature individual. Thcsc creatures are most likely between them, cvcn when evaluatingtheir variability in intrusive and not a segment ofthe prehistoric diet. size, Bisonhad been rocordcdus :I common sighting on Unionidae (freshwater mussels). Only two the Eastern Plains of New Mcxico in historic times minute pieces of musselshcll wcrc recovered from the (Findlcy ct al. 1975:335). site. Three species, 0ylonniu.c tampiromsis, Aves (birds). One small bird coracoid fragment Popnnius poppi, alzd I,umpsilis teres, have been and two medium-sized bird long bones, a juvenile recorded from the Rio Grandcin New Mexico;ml arc l1umerus and a long bone fragment froin a mahlre expected to occur in the Pccos River (Murray 1985:A- individual that werc only 25 to 50 percent present, 24). These fragments could belongto any ofthe above could be identitiedas bird. species. If there wcrc a larger number of shells, we Eggshell. Eight fragments of eggshell were could assumethcsc mussels were being consumed, but collected andthe specimens were severely fragmented. the shell may have been used for ornamentation. Ruten sp. (hroad-winged hawks). A portion of a carpornetacarpus closely rcscmhlcd hawks nativeto Provenicncc Divisions New Mexico, but could notbc matched to a species. This group is charactcrizcd bytheir broad wings, wide The Macho Duncs sitc was divided into six scparatc tails, and soaring movements (Ligon1961:63). excavation areas, A to F. In Area A bedrock wasjust ~SClUhJ~ySsp. (sliders or cooters). Two spccics below thc surfice of this unit. It produced 78 lithic that reside in thc lower Pccos River drainages, artifacts and one sherd, but was completely void of Truchemys scripta (slider) and Psruhnys gorzugi faunal materials. (western river cooter), are commonly mistakenfor one Area €3 exposed tlrcc structures and had thrcc another (Degenhardt et al. 1996:102-103). A plaston bones. A .jackrabbit cervical vertebrae was rccovcrcd fragment and two marginal carapace fragments so were approximately 6 m west of Structure I. A turtle similar to the above species, they could not be carapace li-agment was collectedfrom a pit (Feahlre 3) distinguished. Whether tllcsc turtles were utilized for in thc southeast quad of Structure 2, and a jackrabbit their meat or their shells f'or ornamentation, such as ulna on thc northwest exterior of Struchlrc 3, ;1 brush rattles, is uncertain. structure. Trrrapenc nrrzatu (Ornate box turtle). Found Area C contained the highest number of faunal along thc lower Pecos River drainage,box turtles arc lnatcrials among the five areas. Located between large also known to occupy a variety of habitats, including sand dunes,this structure was offered some protection grasslands with suitable soils for burrowing from the clcmcnts.Two hearths were excavatedin this (Degcnhardt ct al. 1996:104-105). Two carapace area, but only three of the 35 I bones (85.0 percent) fragments wcrcidentified and bothare void of human collcctcd revealed evidenceof burning. A wide variety processing. Ofcrcahlres (Table79), all srnall in stature, were either Sauria (lizards). With hundredsof lizard spccics consumed or utilized to some extent at this seasonal occupying various habitats throughout New Mcxico campsitc, Dcscrtcottontails, which wcrc abundant in and the lower Pccos River drainage, one niaturc the area, yieldcd the largest number of bone(n = 13 1 or fraglncntcd mandible, was difficult to assign to a 37.3 percent), and the amount of cncrgy spent to specific species. It is doubtful this specimen was capture them was probably minimal. Woodrats(11 = XI utilized as a food source. or 23.1 percent) and other unidentifiable small Salietttu (frogs and toads). A fragmented radio- mammals (n = 60 or 17. I percent) were also high in ulna, void of any alterations, was noted in the numbers and easily accessible for consunlption.

MACHO DUNES F'AIJNA 121 Table 79. Frequencies hy Areas at LA 29363

Several features were located outside thestructure Taphonomy in Arca D, including thrcc thermal fcatures. Surprisingly, only1 bone out ofthe 46 collcctcd from Taphonomic variables arc nonhumsn processes that this arca, was recorded as lightly scorched. Nineteen affect the condition of' biological remains (Lyman unaltered bones wcrc retrieved from a posthole 1994: 1). Some oflhcsc variables include honealteration (Featurc 02).Once again, thecottontails (n = I8 or 39.1 by animals andthe environment, which were notedin pcrccnt) and small mammals (11 = 11 or 23.9 percent) the Carlsbad faunal assemblage. Animal alterations dominate the faunalartifacts recovercd fromthis arcx involve gnawing, puncture marks, or scatology. Arca E contained a hearth, situated directly on the Scatological boncs are thosc that have been passed bedrock, with firc-cracked rock scattering the through lhc digestive tract of an animal. The stomach immediate vicinity. Onlytwo mussel shell fragments acids produce smootha polished surface and roundthe were collected lkom this 44-by-60-111 surface-stripped edges of the bone. Only seven (1.7 percent) bone unit. It is possiblc thcsc thermal units were utilized for fragments appear tobe scatological, five cottontail, onc purposcs other than food processing. small mammal, and one srndl 10 medium-sixd Fifteen features wcrc recordedin Area I;, seven mammal. A maxillary molar from a cow or bison thermal features, six storage pits, and two postholes, revealed a carnivore puncturc mark. Rodent gnawing which may have been utilized as drying racks. Faunal enco~npasscs 3.1 perccnt (n = 13 boncs) of thc counts in this area are minor: a partial skeleton of a assemblage. A mature gray wolf ulna exhibits extreme mouse, two cottontail boncs, and onesmall mammal gnawing over thecourse of the element and gnawing element. was also noted on one small marmnal,[bur woodrats, and scvcn cottontail bonc fragments. Weathering results when bone has been with a spiral fracture, and a cut off portion of an environmentally altered through moisture. sun innominate, all of which are desert cottontail bones, cxposurc, or temperature. The degreeof weathering and the largest taxonin the assemblage. depends upon the amountofexposure time, but the rate Butchering as defined by Lyrnan (1994:294) is "the of decomposition will vary betweell environmental human reduction and modificationofan animal carcass rcgions (Marshall 1989:19-20). Only nine elements into consumableparts." This process provides evidence exhibit a range ofWeathering, two through prolonged of cut marks, gouges, and scratchmarks from stone cxposure to the sun, one had pitting, four wcrc tools upon the bone. This assemblage was void of any exfoliated, and two throughroot damage. intentional butchering proccsscs as well as evidence of Splitting can result from environmentalprcssures, tool manufacturing. such as ice, water, or soil weight upon the bones. Trampling uponbone by humans or largc animals, such as cattle, will also producc longitudinalsplits (Marshall 19X9:19-21). Nearly all the bone displaycd some The smallnumber (11 = 413) of bone recovered from degree of splitting or fragmentation through the Crirlsbad site supports the hypothesis of a scasollal environmental factors. Only three bones in the camp. The ma-jority of thc animals (n = 322 or 78.0 assemblage appear to be altered through human percent) arc mature in age, with a drastic drop inthc endeavors. number of juvcnilcs (n= 44 or 10.7 percent), andonly 21 (5.1 percent) are immature. Thc prcscncc of Processing immature jackrabbits, cottontails, woodrats. and various rodents suggests the season of the campsitc to Processing is observed throughburning, breakage, and be from latc spring to late summer. butchcring of an animal for consumption. personal It is possible the inhabitants may rcliedhave more adornment, or utilitarian purposes. Only fourbones in hcavilyonvcgetal food gatheringand processing rather this assemblage exhibitedtwo degrees ofburning, light than fauna. Or, the fauna acquired for consumption (tan/brown) and graded (tan to black). The light could have bccn tosscd into the stewpot andthe lncd (tan/brown) burning can result when meat isroasted on remnants disposedof outsidethe cxcavation area. Even the bone. Graded burning occurswhen a portion of though the number of bones collcctcd was small, the meat is removed from thebone and if this barren bone range of taxa was diverse. The highest lrcqucncy of is thermally uxposcd, it becomes blackened. Excessive bonc from a single taxonis dcsc1-t cottontail (n= 15 1). heat carbonizes the collagen in the bone and the then woodrats (n = 85)- and unidentifiable stnall clcnlcnt turns black. Continuousheat will oxidizc the rnarnrnal (n = 72). The remaining 26 taxa produccd carbon and bonc will become chalky, white,or calcined between 13 and I5 fragments each and some categories (Lyman 1994:384-385). This proccss usually takes werc as low as only oneelement per taxon. place after the mcal is consumed and the refuse is The same variety oftaxaand low numbers ofbone tosscd into the fire. in this assemblage is comparable to thetwo extensive Brcakage and frachlringof bone can occur during excavations near Carlsbad, at the Brantley Dam and thc cooking process from theextraction of grease and Reservoir Project, conductcd by Gallagher andKearden marrow (Marshall 1989: 16), a process which can (1980: 156) and Katz and Katz (19XSa:A-135). Small produce a variety of fractures, such as transverse, numhcrs ol dcscrt cottontail, jackrabbit, woodrats, spird, orlongitudinal on long bones dependingon the rodcnts, turtle, and shell were found along with a method utilized. Not all fracturing is human induccd, variety of burned ring middens or nlescal pits. The much ofit can be from environmentalfitctors, as noted purpose for this typc of seasonal camp was for the in the taphonomysuction ofthis tcxt. Only three ofthe gathcring and procurement of mescal plants, which 41 3 bones in thc assemblage exhibit intentional may have been thefunction of the Carlsbad site. breakage. A tibia with a longitudinal break, a humerus I24 CAKLSEAD RELIEF ROUTE BOTANICAL ANALYSESAT MACHO DUNES (LA 29363)

Pamela J. McBride and Mollie S. Toll

Introduction rurtipendula (sideoats grama), Muhlmhcrgia sp. (muhly grass),Srtariu sp. (bristle grass). Spoxahollts Flotation analysis wasused toinvestigate plant resources cyptundrus (sand dropseed), andScirpu,y sp. (sedge) exploited in the eighth and ninth centuries A.D. at were some of thc grasscs and grass-like plants ibund Macho Dunes, east of Carlsbitd, New Mexico. This growing on the site. extensive seasonal base camp includes remains of 5 brush slructurcs and 35 thermal and storage features Mcthods spread over nearly 5 ha of sand dunes. Thesite location provides habitation and workspacc protected from The 39 soil samples collected during excavation were prevailing windsby a high ridge onthe north. A total of processed at the Museum of New Mexico's Office of 4 I flotation samples taken from 31 proveniencesin Archaeological Studies by thc simplilicd "buckct" Areas 8-F offer insight into the local plant food products version of flotation (see Hohrer and Adam 1977). collected, processed, and consumed. Carbonized wood Flotation soil samples averagcd 2.8 liters in volume, from 38 flolation samplcs and 30 radiocarbon samples ranging in size from0.4 to 7.1 liters. Each sample was provide quantificationol'scvcral taxa from the abundant immersed in a bucket of water, anda 30-40 second local woody vegetation exploitedfor fuel. interval allowed for settling out heavy particles. Thc Macho Dunesis situated in an ecotonal area where solution was thcn pourcd througha line scrccn (about ChihuahuanDcscrtscnlb fromthe south intergradeswith 0.35-nun mesh) lined with a square of"chiffon" fabric, the lower, drier reaches of Desert Grasslandto the north catching organic materials floating or in suspension. (Dick-Pcddic 1993:107). A remarkable diversity and The fabric waslifted out and laid flat on coarse mesh density of shrubscharacterize vegetationin 1he immediate scrccn trays, until the recovered material had dried. environs of LA 29363 (Toll and McBride 1997). Many Each sample was sorted using a series of nested shnh specimens are robust (astall as 2 in, and 3-4 min geological screens (4.0,2.0, I .O, and 0.5 mm mesh), diameter). Rh~rsmicroph-yllum (little-leaf sumac') appears and then rcvicwcd undera binocular microscopc at7- to bc dominant and specimens of this taxon are truly 4sx. glorious here. The significant prcscncc of Juniperus Flotation data are reportedas the number ofsceds monnspv-rna (onc-scedjuniper) is characteristic of the per liter of. thc original soil samplc. Samples upper elevational limits of Chihuahuan Desertscrub conipletely emptyof floral remains (FS 209 from Area (Brown 1994:175). Several other shrub componentsof B, and FS 732 from AreaF) are not listed in the tables the local vegetation arc Characteristic of cithcr Desert of flotation results. To aid the reader in sorting out Grassland orC:hlhuahuanDcscrlscrub.Ofchcsc, Prost~pis botanical occurrencesof cultural significance fromthe glandzrlosu var. Torre.vunu (honey mcsquitc) is considerable noise ofpost-occupationalintrusion, data abundant. Scattcrcd Atr@lex cancsccns (four-wing in tables are sorted into categories of "Cultural" (all saltbush),Acrtciu constrictn (whitethorn acacia), Yucca carbonized remains) and "Noncult-ural" (unburncd &tu (soapwccd yucca),Gutiprrezia sp. (snakeweed), materials, especially whcn of taxa not economically and Mimosu hiunctfkru (wait-a-bit) form the balance of useful, and whcn foundin disturbcd contcxts together the shrubs observed growing the in inmediate site area. with modernroots, inscct parts, scats, or other signs of Herbaceous plants included mclnbcrs of the rcccnt biological activity). composite family such as Zinniu ucerosu (dcscrt Charcoal specimens takenfbr radiocarl>on dating zinnia), Mucltu~rantherulunucrtzfbliu (tansy aster), wcrc cxamincd by snapping cach piece to expose a Hyrnenopupplrs sp. (wooly white), Llyssodia acerosu fresh transverse section, and identified at 45x. All (ncedlclcaf dogwced), and Parthenium incanum charcoal fromeachsarnple wasidentified and separated (mariola). Croton sp. (doveweed),Dithyrtrru wislizrnii by taxon with the exceptionof very small fragmentsof (spectacle pod). Brigonurn annuum (wild buckwheat), charcoal that were impossible to identify. Identificd Sol&unt elacgnifidi~rm (nightshade), and the charcoal from each taxon was weighed on a lop- introduced species, Pegunum hurmulu (African rue) loading digital balance to the nearcst tenth of a gram were also part ofthc herbaceous understory.Bouteloua and placedin labeled foil packets. Low-power, incident light identification of wood specimcns does not often

BOTANICAL ANALYSES ATMAGI10DUNES I25 allow spccies- or even genus-lcvcl precision, but can widespread use of goosefoot lcaves (Kirk 197O:SO). provide reliable information useful in distinguishing The presence ofjuniper sccd fragmcntsin Pit 9 could broad patterns ofutilizalion of a ma-jor resourcc class. be an artifact of using branches for fuel (cones or Charcoal from the 4-mm and 2-tnm screens of berries charred along with the wood).It is less likely flotation samples was identified, to a maximum of 10 that thew seeds derive from processing the fleshy pieces from each scrccn size. Each piece identified was cones ior food, asaromatic rcsins gencrally relegate in the samc manner as radiocarbon wood specimens. these structures to use as medicine, seasoning, or Wood data were reportcd separately forflotation and starvation food (Robbins et al. 1916:39-40; Reagan radiocarbon specimcns, as specimen size often has a 1928:158; Swank 1932:50; Castetter 19353-32). significant effect on species composition. Goosefoot, purslane, andmollugo (carpetweed) arc thc dominant charred seed species from other proveniences Results in Area I3 as well, with juniper occurring in one other sample (Pit 3 in Structure 2). Area B Juniper and mesquitc wood wcrc recovered in similar perccntages by weight in flotation samples; Samples were examined fromall three structures, and both were idcntified in every sample from Area B two cxtramural pits associated with Structurcs1 and 2 (Tables X3a and 83b). A wood that compares favorably (Tables 80-82). Storage Pit 9 (FS 54 and 55) yielded with Mexican crucillo (a member of the buckthorn thc greatest number of charred seeds including family) comprised7 percent of‘thc Area 13 sample by purslane, gooscfoot, carpetweed, juniper, and an weight; rccovery was restricted to Storagc Pits 9 and unidentiiicd seed. Documented cconomic uses of IO. A mall amount (<1 perccnt) of unknown weedy annuals like goosefoot and purslanc seeds nonconifer wood was recoveredfrom the generalfill of abound in [he ctlmographic literature. Castettcr (1935) Structure 2. The distribution of radiocarbon sample describcs the usc ol‘ these seeds as a ground meal, wood is close to that of flotation wood with mesquite either catcn as gruc1 or combined with olhcr food such having a slightly higher percentageof total weight than as corn meal and made into cakes. Carptweed may juniper (Tables84a and 841)). have been boiled as a pothcrb, similar to the

Table 80. Flotation Results, Area B, Structure 1, LA29363 (Macho Dunes)

I26 CAKLSBAD RELIEF ROUTE ‘I’ahle81. Flotation Results, Area B, Structure 2, LA 29363 (Macho Dunes)

I80 Ni22 I E I80 Ni 220 E FS 378 FS 306 FS 346 FS 265

Tahlc 82. Flotation Kesults, Area B, Structure 3, LA 29363 (Macho Dunes)

Fill FS 610

0.5

1 1 0.5 0.7

0.5

BOTANICAL ANALYSESATMACIIOLUJNES 127 ‘I’ahle83a. Species Composition of Flotation Wood Charcoal, AreaB, Structure 1, LA 29363 (Macho Dunes)

Hearth

4/0. I 2kO.I 181’0.4

‘I’ahle 83b. Spccics Compositionof Flotation Wood Charcoal, Area B, Structures 2 and 3,1,A 29363 (Macho Dunes)

X-M pit FS 265

Ioio. J

20/0.3

Table R4a. Species Compositionof Charcoal Submittedfor C-14 Analysis, AreaE, Structure 1, LA 29363

128 C‘AIILSBAD RELIEF ROUTE Table 84b. Species Compositionof Charcoal Submitted forC-14 Analysis, Area B, Structures 2 and 3, LA 29363

cf. Condatia Mexican crucillo was the dominant wood taxon identified in flotation samples from AreaC (‘Table 86). Srnall pcrccntages of One smlplc cxarnined from extramural Fit I I (FS 170) juniper andunknown nonconifcr were recovcrcd. The containcd the only charredprickly pear seed rccovered remaining 23 percent was comprised or probable from thc project (Table XS). The fruits or tunas of Mexican crucillo wood, which was recovercd solely scvcral gcneraof prickly parwere eaten rawor boiled from the extramural hearthand storage pit. Pcrccntages by many Nativc Americangroups (Castetter 1935:35). of radiocarbon wood taxa display the same (;ooscloot is the only othcr charred taxon that was distributions as thc flotation wood (Table 87). idcntiiied in samplcs lrom the Structurc 4 hearth, the south hair of cxtrarnural Pit 41, and Pit I 1. Mesquitc

‘I’nhle85. Flotation Results, AreaC, Extramural Hearthsand Pits near Structure4, LA 29363 (Macho Dunes)

Tablc 86. Species Composition ofFlotationWood Charcoal, AreaC, IA29363 (Macho Dunes)

I30 CARLSEAD RELIEF ROUTE Table 87. Species Composition of Charcoal Submittedfor C-14 Analysis, AreaC, LA 29363

FU~LII-~ llcalth (I'it 41) 'I'OtillS

1.1 4.8 3.7 9.6 18% 7.5 28.2 41.2 70""

Areu D was recovered in all five flotation samples from Area D (Table 89). Tracc amounts ('1 pcrccnt) of Posthole 62 produced charred goosefoot, purslane, saltbushlgreasewood, mesquite, and unknown juniper, an unidentiikd sccd,and three unidentified nonconiferwere also identified. The samples examined fruits (Table 88). Charred gooscfoot was rccovcrcd from Postholc 62 had only two small li-agmcnls of from two other pits, and Thcrmal Pit 51 yicldcd a juniper and one ofmcsquik, representing debris rather charrcdjuniper seed fragmentas well. Juniper charcoal than allowing for any identificationof the post itself.

l'ahlu 88. Flotation Results,Area D Extramural Features,LA 29363 (Macho Dunes)

tJnidrntifiable

'Iota1 C:ultural TAXA Total Cultural SELDS Noncultursl: tantha (Hiddenflower)

Total Noncultural SEEDS I Table 89. Species Composition of Flotation Wood Charcoal, AreaD, LA 29363 (Macho Dunes) Arm E plant after maturation (preventing their loss before harvesting [Doebley 19841) may havc made dropseed Two samples from the hearth were analyzed and grains a valuable resourcc. Mesquite was the dominant charred gooscfoot, purslane, and dropseed grass seeds wood burned in the two isolated hearths in Arca E wcre recovercd (Table 90). Although dropseed grass (Table 91). A smallamount ofjuniper (5 pcrcent) was grains are very small,the positive qualities of abundant also recovered. seed production and the rctcntionof the grains by the

Table 90. Flotation Results from Extrtlmual Features, AreasE and F, LA 29363 (Macho Dunes)

AI- Hearth FS 531

0.8 0 4

2.7 I I n 0.8 4

2.4 I I ~ 0.5 +[wig I N 3 2 0 6.3 2.3 0

Tablc 91. Species Compositionof b’lotation Wood Charcoal, Area E, 1,A 29363 (Macho Dunes)

Area IT wood from the area (Tablc 93). Mesquite, thc dominant wood typc by weight, was identified in all but twoof Cllarrcd gooscfoot seeds wcrc rccovered fromGvc of the pits, while junipcr was recovcrcd from all pits thc ninc samples yieldingcultural remains (Tablcs 90 cxcept the roastingpit (Pit 3 I). A small percentagc(x1 and 92). Other chamd cconornic taxa include purslane, percent) of unknown nonconifcr wasalso prcscnt in Pit juniper, and carpctwced. Mexican crucillo was 29 and Pi1 32. Mesquite was also the dominant wood identificd in four ofthe eight pits cxcavated in Area I:, taxon idcntified in radiocarbon samples, followedby colnprising 27 percent by wcight of the total flotation juniper and Mexican cnlcillo (Tablc 94).

132 CARLSRAD RELIEF ROUTE Table 92. FIotation Results,Area F Extramural Features,LA 29363 (Macho Dunes)

W Feature

FS No. Cultural: Chenopodium (Goosefoot) Portulaca (Purslane) Mollugo (Carpetweed) Unidentifiable

Table 93. Species Composition of Flotation Wood Charcoal, Area F, LA 29363 (Macho Dunes) Table 95. Ubiquity of Carbonized Plant Remains by Area.

a total occurrences + number of samples. total number of seedsiliter + number of samples. +

I30 CARLSBAD RELIEF ROUTE Tahle 97. Charred Hcrnains of Economic Plant Taxa from Formative Period Sites in the Cllihuahuan Desert Biotic Community

Ratscar'

Location Carlsbad Wcstcrn Basin l'ularosa

PERFNNIAIS mesquite + + + t I+ pricklypear + + + + +I I I I I I

panic grass I I I I I

+ I I I + + + + + I I I

+ I+ +

CULTIVARS

BOTANICAL ANALYSES AT MACHO DUNES 157 Table 98. Utilized Wood from Formative Sites in tbu Chihuahuan Desert Biotic Community

NONC'ONIFERS:

Macho Dunes shares with other Fonnative absent even where rnesquitc is the predominant period sitcs of the Descrt Grassland-Descrt Scrub fuel wood (for instance a1 4.84 M and St. Theresa, zones a predorninancc of goosefoot and purslane as well as Macho Duncs; see Table 98). Prickly among economic weedy annuals in thecultural pear and hedgehog cacti are low frequency but asscmblage. Perenllials vary considcrably widespread, in contrast to remains of the leaf according to local abundance. Mesquitc food succulents (agave, yucca, and sotol), which are rcrnains have never bccn recoveredin abundance in coocentrated infoothill habitats. ihosc archaeological sitcs, and are frcqucntly

138 CARLSHAD KELIEF ROUTE PALYNOLOGICAL ANALYSIS

Richard Holloway

Mcthods and Materials and bleach, which have been delnonstratcd experimentally to be destructive to pollen grains (Holloway I98 I ). From each pollen sample submitted,25 grams (g) Ten grams of soil wcrc subsampled and placedin of soil were subsampled. Prior to chemical extraction, beakers where the carbonatcs wcrc rcmovcd using three tablets of concentrated Lycopodium spores (batch hydrochloric acid.The samplcs were rinsed, scrccncd #307XA2, Dcpartmcnt of Quaternary Geology,hnd, lhror~gh 150 p mesh screen and a series of short Sweden; 13,500 f 500 marker grains per tablet) were centrifugations wereinitiated in orderto remove acids added to each subsamplc. The addition of marker and facilitate the removalof clay particles. This step is grains pcnnits calculation of pollen concentration values and provides an indicator for accidental necessary since phytoliths smaller when 2-3 p arc rarely identifiable or valuitble. The samplcs wcrc destruction oi'pollcn duringthe laboratory procedure. sonicated using a Delta D-9 sonicator for a periodnot The samples were treated with 35 percent exceeding 30 seconds and severalshort centrifugations hydrochloric acid (HCl) overnight to removc wcrc performed. carbonates andto release theLd)mpodium spores li.om The residues were transferredto glass IO0 mllubes their matrix. After neutralizingthc acid with distillcd and Schulzc's solution (42 pcrccnl nitric acid and water, the samples were allowed lo scttlc for a period potassium chlomtc) was addcd to the samples. The of atleast thrcc hours bcforcthe supernatant liquid was samples were placedin a boiling water bath for about rcmovcd. Additional distilled water was addcdto thc 2-3 hours or until all organic traceshad hecn rcmovcd. supernatant, and the mixture was swirled and tllcu The samples were centrifuged and rinscduntil neutral, allowed to settle for 5 seconds. The suspended flne and then rinsed with 5 percent KO11 lo rcmnovc any fraction was decanted through 150p-mcsh screeninto additional humatcs. After additionalrinsing, the residue a second beaker. This procedure, repeatedat least thrcc times, removed lighter matcrials, including pollen was transferredto a 300 mItall glass beaker, and thc samples were fractionated ina watcr column. This grains, from the heavier fractions. Thethe nlaterial hctionation resultcd ina scparation ofparticles into a was conccntratcd by centrifugation at 2,000 revolutions 3-2511 and a 25-150p-size range. The residues wcrc per minute (RPM). transferred to 1.5 ml tubes and as much watcr as The fine fraction was trcatcd with concentrated possible was removed in prcpwhon i'or heavy density hydrofluoric acid ([IF) overnight to remove silicates. scparation. Zinc bromide(S.G. 2.38) was addcdto the After completely neutralizing the acid with distilled samples and centrifuged at high speed 10 for minutes. watcr, thc samples were treated with a solution of Thc lighter. phytolith fraction was collected and this Darvan, and sonicatedin a Delta D-9 Sonicator for30 step was repeated. Very few phytolith spccimcns wcrc seconds. The Darvan solution was removed by repeated present hmthcsc samnplcs. washing with distilled water and centrifuged (2,000 The phytoliths wcrc rinscd and transferred to RPM) until the supernatant liquid was clear and absolute ethanol for curation.A single drop containing neutral. This procedure removed t?ne charcoal and phytoliths was addcdto a cover slip and it was allowed other associated organic matter and efTcctively to dry. A drop ofMeltrnountadhesive (refractiveindex dcflocculated the sample. 1.539) was added to the coverslip and a permancnt The samples were dehydratedin glacial acetic acid slide was then made. in preparation for acetolyis. Acetolysis solution (acetic anhydride: conccntratcd sulfuric acid in 9:l ratio) Pollen lhtruction Methods following Erdtman (1960) was added to each sample. Ccntrihgc tubes containing the solution wcrc hcatcd in Chemical extraction of pollen samples was conducted a boiling water bath for approximately8 minutes and at the Palynology Laboratory at Texas A & M then cooled for an additional 8 minutes before University, using a procedure designed for semi-arid Centrifugation and removal of the acetolysis solution Southwestern sediments. The method, detailed below, with glacial acetic acid followed by distilled water. specifically avoids use of such reagents nitric as acid Centrifugation at 2,000 RPM for 90 seconds dramatically reduced thesix of the sample, yet from periodic examination of the residue, did not remove condition, that I suspect it was a modern contaminant. fossil palynomorphs. No pollen was present from theotherpollcn extraction, Heavy density separation ensued using zinc The phytolith residues were likewise in very poor bromide (ZnBrJ, with a spccidc gravity of 2.00, to condition. The phytolith concentrations on the slide rc1novc much ofthe remaining detritus from the pollen. were very low and I anticipatedthat standard counts of The light fraction was dilutcd with distilled water 300-500 specimens would be difficult, if not ( 10: 1) and concentralcdby centrifugation. The samples impossible to obtain from thcscslides. Additionally, wcre washed repeatedly distilledin water until neutral. the phytoliths showed strong evidcncc ofdeterioration, The rcsiducs were rinsed in a 1 percent solution of which was likely due to the local soil chemistry. potassium hydroxide (KOFT) for lcss than one minute, Howcvcr, given thehighdegree ofdeterioration, which which was effective in renloving the majorityof the had occurred to these assemblages, thc likelihood of unwantcd alkaline soluble humates. obtaining an accurate interpretation from these The material was rinscd in Ethanol (ETOH) assemblages was minimal.Hascd on this information, staincd with safranin-0, rinscd twice with ETOH, and I advised that no further analysis be conducted on these transfcrrcd lo 1-dram vials withTertiary Butyl Alcohol samples. Sincethis was thc ikst set of samples, J also (TEA). The samplcs wcre mixed witha small quantity suspectcd that recovery was notlikely to improve by of glycerin and allowed to stand overnight for the processing and cxlractioll of additional samples. evaporation ofthe TBA. The storagevials were cappcd 'Thcrcfore, none of the other samples was processed. and wcrc returned to Muscum of New Mexico at the While in many areas of the arid Southwest, completion of the project. phytoliths arc thought to preserve better than pollen A drop of the pollinifcrous residuewas mounted assemblages, phytoliths arc not indestructible. The on a microscope slidef'or examination underan 18-by- deterioration and destructionofpytoliths appearsto be 18 mm cover slipscaled with fingernail polish. The correlated with highpH values ofthe scdiments andthe slide was examined using 200x l00x or rnagnjfication presence of free iron and aluminum oxides(Pcarsall under an Aus-.lcna Laboval4 compound microscope. 1988). Wilding et al. (1977) and Pease (1967) have Virtually no pollen grains were recovered from thc two both reported on a phytolith plI solubility curve. samplcs cxuactcd.A singlc grain ofpinits was present Basically, thesc authors suggestthat the biogenic silica but was in such pristine condition that it likely go back into solution under conditionsof highcr pH, represented modern contamination. which is usually accompanied by thc presence of carbonates withinthc assemblage. While thepH liom Discussion the study area is not significanlly higher than li-om other archaeologicalsites, it probably falls in the range After extracting only two samples eachof pollen and of 8-9. Severe deterioration of phytolith assemblages phytolith matorial, Texas A & M University advised generally occurs atpH levels abovc 8.0. However, the me of the extremely poor shape thc of assemblages. In soil conditions from the projcct arca wouldlikely be their opinion, there waslittle likclihood of obtaining sufficient to weather the assemblages. Thus,it is likely countable phytolith or pollen assemblages from the that the local soil conditions have actedin this instance continued extractionof samples. Artcr examiningthc to significantly alter the phytolith assemblages. It pollen residue from these two samples,it was dccided should be remembcrcd, however, that not all soil to procccd only with the phytolith extractionsfor the conditions withpH values between8-9 would result in rcmaining four samples submitted a as test. There was deteriorated phytolith assemblages. For this reason, virtually no pollen presentin these residues. A scan of analysis of lhcsc types ofartifacts should be continued om slide indicated only a single Pin14.s grain was in order to individually assess the potential for prcscnl on the entireslide. This grainwas in such good phytolith recovcry. DATJNG THE CARLSBAD SITES

Yvonne R. Oakcs

While the lack of large, in-depth block studies for soulheasternNew Mexicois discouraging, a nurnbcrof small-scalc excavations producing sitc data that arc increasing our knowledge of the region far beyond what existed even 15 years ago. Until recenlly, the dating of southeastern New Mexico sites was almost a fortuitous occurrence, considering the dunal environment and often wind-blown site surfaces. Today, because of improved archaeological methodology, excavatorsare opening up broader areas of sites and taking potentially datable samples fromthe small hearlhs and pits that fi-equcntly occur. The Chrlsbad Kelief Route project openedup large areasof the two excavated sites, linding roastingpits, hearths, storage pits, and brushstructure floors that allowed for Figure 9312.Middle Archuic (3000-2100 KC.) recovcry of abundant datablematerials. rudiocurhnn dates. Twenty radiocarbon dates wcrc obtained from Macho Ilunes (1,A 203h3), ranging from the Middle Archaic to ca. A.D. 1000 (Table 99). Several arcllaeomagnetic samples werealso taken liom the sile, but the loose, sandy composition of the soil did not allow for valid dates.No cultural features were present at Trojan IIill (LA 29362) to yield absolute dales; however, two nearby hearths produced datesof cal A.1). 1000-lO55 within a I-sigma range and an intercept date of’A.D. 1025 andcal A.D. 1150-1245 in a I-sigma range andan inlercept date ofA.D. 119S. However, thcsc may or maynot be associated with the occupation at ‘l’rojan Hill. ‘l’he four piecesof ceramics on the extensively excavatedsite likely do notcorrelate with its use as a quarry sourcc. The radiocarbon dates for Macho Dunes indicate periodic reuseofthe sitc arca through time (Table100). ‘l’oday the environment is dunal; however, an abundance of mesquitc, little-lcaf sumac, and othcr desedic vegetation is more plentiful on the site than in surrounding areas and botanical analyscs indicateit may havebeen like this through titne. ‘l‘hePecos River is also within easy walking distance(6 km), which may have been an additional factor in site selection. Reuse ol’ site areas is not at all uncommon in southeastern New Mexico as seen in Figure 93. kighty-threc available regional radiocarbon dates were plotted for thc specific periods that Macho Dunes was occupied. Besides Macho Duncs, several areas that exhibit consistent reoccupation through time includeHrantlcy Reservoir (Gallagher and Bearden1980; Katz andKatz 19X5a), the WIPP site area (Lordand Reynolds 1985; Figure 9.3~.Eurly Gldw (A.D. 650-750) rudiocurhon dares. radiocarbon datescome from tnany different locales on the site. The Middle Archaic date of cal 2570-1950 B.C. with an intercept date of 2280 R.C. is dcrived from Koasting Pit 31 in Area F. No ceramics were associated with this feature. A Late Archaic dateofcal 235 B.C.- AD. 75 with an intercept of 2 1013.C. is from a singlc hearth at the wind-scoured crest of the site ridge in Area A. No other cultural features were nearby; however,an F.1 Paso 13rown and a South Pecos Brown sherd were found 10 n1 downslope to thenorth and south ofthe hearth, but presutnably not associated. Formative period sites usually arc idcntificd by having ceramics in correspondence. Hearth5 I in Area D and Thermal Pit 34 in Area F are the two carlicst Figure Y3d. Middle Globe(AD. 7.50-800) Formativo pcriod features on the site, dating between radiocurbon dute.s, A.1). 650 and 720. '['hey arc located on opposite sides ol' the road and Hearth 5 I has 5 El I'aso Brown sherds within 5-6 111 but their relationship to thc hearth cannot be confirmed. Therrnal J'it 34 is in a 77-sq-ni area that contains 12 pits ol'various types including roasting pits, hearths, thermal pits of unknown use, and possible storage pits. In this area, 39 shcrds wcrc also rccovered. Thirteen of these were within 4 of the pits. A contour plot was generated showing locations and area of conccntration of the shcrds (Fig. 94). The spatial patterning indicates no shcrds can bo directly associated with Thermal Pit 34, wllicll dates to cal A.D. 660-760 (I -sigma deviation)with an intercept date of A.1). 670, 'l'heruforc, it docs not appear that pottery was present on the site in the Early Formative period (Early Globe phasc). A very mobile subsistcncc pattern is suggested for this time period at Macho 1)uncs with only Hearth 5 I and 'l'hertnal Pit 34 present. It is xtually possible that Late Archaic Ilunter-gatherer adaptations havepersisted to this time on this particular Staley et al. I996), the southern Guadalupe Mountains site and these cultural features arcnot representative of (Miller 1994), and the Loco IIills area (Oakes 1982, the Formative period at all, but rather LateArchaic. 1985). Table 99 suggestsa hiatus ofpcrhaps 100 years on While the number of dates are expectedly lessfor the site betweenA.D. 660 and 750. However,between the Middle Archaic pcriod between 3000 and 2200 approximately A.D. 750 and A.D. 800, Macho Dunes B.V., occupation evenat this time is spread throughout experienced its greatest period of occupation. Ninec" the C:trlsbad Basin and surrounding arcas(1;ig. 93a-e). 14 dates from sevcn difftrent cultural featureswere Katz and Katz (1 993: I 13) previously suggesteda gap obtained. Five of these are from the four datcd brush in occupation between 3200 and 1700 B.C. but this structures, suggesting an encampment of enough seems to have been Glled in by our data andthat 01' duration for occupants to have needed brush structures othcrs in the region. We can only speculate that thc with hearths for shelter. excavation of hroader areas of sites will greatly Brush Structure 1 in Area B is theearliest feature increase the total of dated sites, includingthe Middle dated to the Middle Globe phase (AD. 750420)with Archaic. Most datcd sitcs in Figure 93c-e relate to the an intercept datc of A.11. 760 and a 1 -sigma range of Early and Middle Globe phases but, bccausc ofthc low cal A.D. 680-790. Two other dates forthis feature sample size, may or may not correspond to actual (Table 99) arcfrom storagc pits and not considered :IS population counts at that time. reliable as hearth dates. One datc overlaps A.D.the 760 Macho 1)unes is an extensive sitewith discrete date and the other is over 100 years later. Twelve utilized surfaces spread out overthe landscape. The ceramics are presentwithin Structure I; five brown

142 CARLSBAD RELIEF ROIJTE G

0 1 meter U contour Interval =

Figure 94. C'wlturuljeufwres in Area F wifh locutions ofcernmics plotred in one-frequency contours.

'I'ablc 99. C:-14 IMcs for Carlshxd Proirct

Arca Bcta Agc B.P. Calibratcd I- Calihratrd 2-Sigma Intcrccpt Contcxt Kumbrr Six:m;1 h 1 10521 2 I60170 355-200 B ('. 380 -5 B.(~. I85 B.('. I learth 230-75 KC. B I 10084 I210 I.t)O A.D. 005-K80 A.D. 005-Oh0 A.D. 785 Plt Stmchlre 2 Storage Pit r3 1 IO98S 1 1 50+hil A.1). 885-9813 A.D. 770 tolo A.1). 890 PII Slruclurc I Storaac Pit

DATING THE C'ARLSBAII Sl'l't.:,s' I33 Table 100. Formative Pcriod C-14 Dates Ordered by Arca and hy 'lhle

"

C Thcrrnal PI! 21 Area F -

ware types are represented. The nearbyBrush Structure was recovered at a 3-m distance. Brush Slruclure4, 2 also dates to this same phase at cal AD. 685-870 approximately 15 m south of Hearth 4 I, contained 12 with intercept dates at A.D. 770 and 785. Fourteen sherds of 4 diffcrcnt types. It has a date 01' citl AD. sherds of tlzree different types were found in 760-880 with an intercept at A.D. 790. association. As a result of the relationship of shcrds There is a slight gap inC-14 dates betweenA.D. with these struchlres, we can conclude thatA.J). by 760 800 and 850. Four datcs arc subsequentlyreported at a minimum, ceramicswere present at Macho Dunes. between AD. 8S0 and 800. Of these, three arere-jcctcd. In Area I), Pit h 1 (possibly a small storage pit) is They arc from Structures 1 and 4 but were retrieved the only featurein the area of this time period at cal from fill, a storage pit, and an adjoining roasting pit. AD. 680-850 with an intercept ofA.D. 770. Five other Radiocarbon data from better proveniences in these pits md Brush Structure 5 are nearby,however. Only features suggest these two units are earlierthan AD. eight ceramics rcprcsonting four brownware types are 800. However, there is only one C- I4 date for Structure present. No sherds are associated with Pit 61; most 3 in Area B, from a hearth, andit has anintercept date seem to correspondwith the roasting pit directly to the of A.D. 895 with a 1-sigma range of cal AD. 780- northeast andwith two othersmall pits. In Area F, two IO1 S. If this date is correct, thcrcwould have bccn R thermal pits (23 and27) located 4m apart dateto this subsequent habitational occupation at this timcperiod AD. 750-800 timc frame with intercept datcs of A.D. on thesitc, indicating repeated reuse ofa specitkliving 770 and 785. Both pits contained ceratnicswith a few area on thcsitc. sherds alsoscattered close by. Aminorpost-A.1). 9SOutilization ofMacho Dunes In Arca C1, two nwre cultural fcahlres aredatod to is represented by three radiocarbon datesin Areas D, E, AD. 750-800, IIearth41 and BrushStructure4. Hearth and F, all derived fromsmall hearths orpits. In Arca I)+ 41 is isolatcd from habitation units and only 1 shcrd IIearth 50 hasa post-AD. 950 date of cal A.1). 7x0- 1025 with the intercept at A.D. 960. Hearth 50 is near more prevalenthere than in othcr areasol’New Mexico Hearth 5 1 (with an earlier date andno ceramics) and where two or maybe three different occupations also has no ceramicsin the immediate vicinity. The overlap, but in the southeast, there often seems tobe hearth could easily have been interpreted as five or six separate reoccupations of a singlc arca preceramic, ifthc C-14 date werelacking. In Area E, a (Staley et al. 1996:256). Therefore, dating ofnumerous singlc hearth was foundon the crest of a low ridge. No features on asite is an important consideration during sherds were in the area and without the radiocarbon excavations in order to isolate the variouscomponents date ofcal A.D. X85 to 1005 and an intercep1ofA.D. potentially present. 975, again, the hearth could havebeen thought to have The range of datcs obtained allowed for a been preceramic. In Area E‘, Pit 21 lies in a locale of determination of when ceramicsflrst appeared on the multiple pits ol’ various dates. Thispit dates to cal AD. site, which was about AD. 750. The minor occupation 890-1 010 withan intercept of A,D. 980. It did contain of Macho Dunes in the taler AD. 600s did not two different types of brown ware sherds. associate with any pottery.And the latest dates for the In conclusion, the recovery 20 of radiocarbon dates site, around A.D. 968,do not include any black-on- from the multicomponent site of Macho Dunes is white or corrugated pottery. ‘Thcsc findings are significant because a large number of samples were somewhat difikrent than what is normally thought ofas rctricvcd from what is often an unproductive the cxtcnt of ccramic occupationsin the region. Katz envirolment. The implication that the various dates and Katz (1993) indicate ceramicsmay begin as early represent six different occupational periods from as A.D. 500 and black-on-white pottery can showup by Middle Archaic to the Late Globe phase (2300B.C. to A.D. 750. Neither of these expectations weretrue for A.D. 980) is also very important. The pattern of Macho Dunes. Weresite occupants so localized intheir repeated occupation of a single locale is evident, not adaptations that trade with tnakcrsof black-on-whitc only at Macho Dunes, but in several othcr arcas of ceramics did not occur or were they basicallystill southeastern New Mexico mentioned previously.It is maintaining an Archaiclifcstylc with onlya minimum now up to future research to ask why consistent reuse amount of pottery, as has been elsewhere suggestcd ol s~meplaccs occurrcd. What wcrc the resourcesor (Ortkcs 1982:46; Stuart and Gauthier 19x3268; reasons that drew people century after century to a Sebastian and Iarralde 1989:92)’? particular place‘! This phenomenon may actually be 146 CARLSBAD RELIEF ROUTE SYNTHESIS by Dorothy A. Zamora

The main objectivel’or data recovery at LA 29362 and materials and variety ofreduction techniques. This area LA 29363 along the Carlsbad Kelief Koute was to also contained tnost of the smaller thinning flakes and answer questions on settlement and subsistence cores, suggesting thatmuch of the production of flakes behavior in the north end ofthe Trans-Pecos culture was laking place in this area. Area C, however, area andto place our sites within the cultural sequence contained all local materials and did not have any outlined by Paul and Susana Katz(19XSa). utilized artifacts as found in the other areas, suggesting a limited activity regimen (Quaranta, this volume). LA 29362, ‘Trojan Hill Four Playas Incised ceramicsfrom the same vessel were recovered from oneunit in Area A. Wilson (this During Ihc tcsting phase, ‘l’rojan Hill was thought to volume) believes that this a isolated incident,apassible have been an Archaic manifestation; however, the pol drop bypeople passing through. IIowever,this is excavations provided evidence that the site is an evidence that people were presenton the sitc during extensive lithic procurement area. Thereis no evidence A.D. I400 to A.D. 1500s. Also,two hearths found near that the site was occupiedby only one distinct group of the sitc could possibly be associated with Trojan IIill pcoplc, but rather through lithic analyses, by different and their dates place its occupation during the occupations through time. Formative period, ‘l‘rojan Hill coversan extensive areaand consists of lithic artifacts that suggest primary reduction and LA 29363, Macho Dunes Loot manufacturing activities. Not all of the site was collected; however, the sample recovered was The purpose of data recoveryat Macho Duneswas to substantial (N = 12,269 artifacts). There was a wide investigate problcmsposed by the environment, such as range ofmaterialbeing used. Chert(54.7 percent) and the acquisition of subsistence items. ‘l’hese specific quartzite (39.7 perccnt) were the most comtnonly environmental variablesin the Carlsbad region include utilized tnaterial type. Core flakes (60.0 percent) (1) season of’availability, (2) quantityol’biomass, (3) dominatc the assemblage, but there are all stages of accessibility of resources, and (4) density of manufacture. With such a large numberof artifacts participation population. recovered, thcrc is a relative small number of fortnal Specific questions relevant to the site (1are j site tools (.5 percent) in the assetnblage, which includes cl~onology,(2) site I‘unction, (3) subsistence, (4) site utilized debitage (n = 5 l), hammerstoncs (n- X), and cxchangc or mobility, and(5) chronological compari- core-hammcrstones (n - 1). sons with other sitesin the area. The absence ol’hearthsand storage pits madeit very difficult to place thc site into a spccific time Sitr Chronology period or phase.?‘he absence of these features didnot pennit us to collcct samplcs for macrobotanicat Macho Dunes dates mostly to thc Formative period, analysis, also the lackof fauna on the site leaves the and was occupied between A.D. 670 and A.D.950. question of subsistcncc and foraging pursued Howcvcr, two Middle Archaic componentsarc present unanswcrcd. As the site consists tnostly of lithic in Areas A and F. ‘The Archaic date of 380 t3.C. to 5 debitage, there isno real evidenceof when the site was B.C., intcrccpt date of 1x5 B.C in Area A is from a utilized. Doleman (1992) states thatif a sitc wasused small hearth, which is located on top ofa low rise and during the Archaic period, the artifact assemblage has beenbuilt into the outcropping bedrock.‘I’he other should reflect resource-specific procurement and Archaic date was retrieved from Area Fin a large, processing activitics. It does not secm as though the shallow roastingpit (31 j dating 2460B.C. to 2165B.C. lithic artifact assemblagecan dispute or agree withthis and 2065 B.C. to 2060 B.C. with an intercept date of statement. The assemblageis generally typical of all 2280 U.C. It was isolated to the west away fromthe periods and phases of southeasternNew Mexico. cluster of features under a sand dune. However, the There are, however, distinctions betweenthe areas dates from the radiocarbon samples suggests that (A-C) 01’ excavations. Area A contained rnost of the Macho Dunes was most frequently used during the artifacts, which included the ma-jority of thc exotic Globe phase, with minor activity both earlier and later.

SYNTHESlS 147 circular striations, suggesting wild resourccs were being processed rathcr than corn. The number of It appears clear that MachoDunes was a seasonal camp ground stone artifacts indicates the wassite a seasonal used lor gathering wild resources. Although each camp ofsome duration. structure contained interior hearths, the fauna and ethnobotanical data suggest that the sitewas occupied during the late spring to early fall. It is possiblethe that weather may have been cool during these limes, In the desed land in southeastem New Mexico, the explaining the interior hearthsin the structures. vegetation is vcry sparse. MachoDunes seems to bt: The faunal remains recovered from thc sitewas somewhat of an exception to that generality. Todaythe that of small rodcnt species. Therewas a lackof small vegetation on the sitc consists ol' several species, to large mammals implying that the inhabitants were including little-leaf sumac, one sccdjuniper,mesquite, rclying more on vegetal food gathering than on four-wing saltbush, white thorn acacia, yucca, hunting, snakeweed, wait-a-bit, desert zinna,tansy aster. wooly The ceramics from Macho1)uncs arc all brown white, dogwood, doveweed, buckwheat, nightshade, wares and theirlow frequency suggests theymay bc and African rue, 'The grasses present on the site arc associated with groups who are fairly mobile and rely sideoats grama, muhly, hristlc, sand dropseed, and on wild l'ood resources (Wilson and Hlinman 1991). sedge. Mauldin et al. (1 998) state that in the central Hueco The flotation samples that were collected from Holson, usc of ceramics was to hold reconstituted features produced purslane, carpetweed, goosefoot, and stored grains and thcy wcrc used in sites occupied dropseed grass.The most commonfuel wood used was during tlle winter and early spring months when stored mesquite andjuniper; however,there was minor useof food was in grcater demand.A more sedentary group Mexican crucjllo and four-winged saltbush.None of would produce more dccorative pottery and textured thcsc are domesticates, suggesting that only vessels. Wilson (this volume) statcs that sedentary procuremcnt of wild hodresources waspursued on the lifestyles result in an increased distinction 01' site. Ethnobotanical samples also yieldedno evidcncc decoration, paste,and fbrm between utility and paintcd ol'domesticates on the site. wares, rcflccting spccialized activities. Wilson further The data from the faunaimply that foodgathering, states that these activities, suchas boiling and serving not hunding, was the main function on thc site.The corn, reflect the increase relianceon agriculhlre, where remains contained mostly small animals,of which 59.6 more mobile groupsusc general vessels such asplain percent were rabbit and29.8 percent rodent; the rest brown wares. The ceramic assemblage exhibited were indeterminate fragments. None of the bones attributes generally relatedto cooking (Wilson, this exhibited burning and there isno evidence from the volumc). 'l'hcrc is no evidence ol'vessel production on fauna that indicates animal processing. the site. However,there has been evidenceof pottery production in some areasof southeastern New Mcxico (Hill I W6a). The chipped stone assemblageimplius that wild At Trojan Hill, seven Ilakes of' Tecovas chert were food gathering wasbeing performed at Macho Dunes. found. They all seemedto have come from thesamc Although there was lack a of formal tools, the tools (n artifact in Area A. It is possible that this material was - 4) recovered exhibited expedient use. One was a curated by someone passing through. There is no reworked chertpro-jcctilc point with a concave basethat cvidcncc 01' loo1 manufachlre or utilization of the has an Archaic-like appearance. The other is anchert 'l'ecovas chert.A single flakeofrecovas chert was also end-side scraper. The projectile point seems to be recovered from Area D at Macho Dunes. These few evidence of' curation. The tip was reworked and reuscd occurrences would not suggest trade; however,it does for other tasks besides hunting. The end-side scraper mean that this material type was broughtin from some was probably used on vegetal items. Included in the other place, suchas 'l'cxas, sinceit is not available in fornlal tool category was a chopper probablyused in the area or in southeastern New Mexico. Onc other cutting thick vegetation. None of the tools exhibited exotic material, smalla piccc ofobsidian, was found on heavy use implyingthat they were being expediently the site; however, becauseit was only one stone artifact used and then discarded. it is likely that no significant trade or exchange was The Lair number of ground stone from the sitc going on at the time the site was occupied. included tnostly fragmcnted manos and metates; Mobility is delined by Hitchcock(1 982:m) BS however, the wear patterns on the surfaccs cxhibited "the movetnent of thc co-resident group, includingboth

148 CAKLSRAD RE1,IlT ROIJTE producers and dependent, from one location to three postholes were found north of Structure5. This another." Nonsedentarylife is more broadly definedas area may haveserved as an exterior activity area with a short-term residence in a number of locations a ramada-type structurepresent. (Whalen 1994a). Whalenalso states that mobile people The interior hearths found on the site would are perceived as gatherers that collect and cook food suggest that the site was occupied during the cold without the assistance of ceramic vessels and wholive rnonths; however, the outside thermal featuresin Areas in small groups and maintain an egalitarian social A, D, E, and F would also suggest use during warmer culture. weather. The fauna recovered indicatethat the site was Mobility patterns are being studied more and more occupied during thelate spring to late summer. The by Southwestern archaeologists. C'ordell and Plog botanical data suggests summer to carlyfall. However, (1979), lor example, foundthat food production does the multiple dated occupations of thesite would allow not completely replace food collecting and mobility for. use during serveral different times of' thc year. continues in some areas into the late prehistoric The lithic diversity index (LDI) for the site periods. Wills and Windes (1989) state that the implies, at first glancc, that Trojan IIill is more Basketmaker I11 period in the San Juan Basin has sedentary and less mobile (Quaranta, this volume). basically a sem-agricultural adaptation, but relies Howcvcr, thcrc arc other considerationsto be taken heavily on wild resources.In the San Simon areaof into account, If this is a quarry sitc where materials are Arizona, Late Formative period sites arc seasonally being procured, [hen the diversity of the materials is mobile occupations (Gilman 1986, 1991 ; Minnis and not going to bc grcat. Also, location theof site affects Wormer 1984). Lightfoot's(1984) excavations at the the results of the LDI, because material availability Duncan sitc, a T.ate Formative occupation, recovered hcrc is low and the materials with the best qualityare ceramics, grinding implements, and variousfeatures; going to be selected. however, the analysis indicateda seasonal occupation. At Macho Ilunes, theLDI is higher thanat Trojan In the Palomas Drainageof West Texas andthe south- Hill, It would seem as though the LDI for Macho central Rio Grandc Valley in New Mexico, Nelson Dunes should have been lower. However, becausethis (1 984) believes there wasresidential mobility during is a seasonal habitation, it is very difficult to use the the Late Formative period in these desert regions. 1,131 results, since all the materials wcrc brought in It has been suggestedthat populations were highly from other sources. Thestructures and external storagc mobile, moving in and out of their regionsin seasonal and thennal pits would imply at least a seasonal rounds (Carmjchael 198 I, 1983; Lekson 1088). Hard habitation. l'husc arc not the types of features (1983b), however, al-gues for a biseasonal round for associated withlong-tem occupation. Whalen(1 994a) some the Formative period.As Whalen (I 994) points associates cachepits with highly mobile people, large out "Formative period populationsof the deserts of extanlural pits with seasonally tnobilc groups and South-Central New Mexico and Western 'l'exasrelicd intramural storage rooms with sedentary people. heavily on mobility to cope with the region's harsh 'l'he small assemblage ofcultigens recovered from conditions." Macho Dunes indicatesthat subsistence was bascd on At Macho [lunes it is evident that the site was not gathered wild resources. Harsh environmental occupied year round, but seasonally. The ceramics conditions also probably played a role in mobility prcscnt were broughtin by site occupants, no ceramic strategies in the area (Whalen 1994a). Becauseof these production occurred on the site. The lithic artifacts conditions, productivityofdescrt sccd-bearing plantsis wcre also transported, possibly from the nearby dependent on the rainfall pattern (Whalen 1994a) and procurement site of Trojan IIill. The ground stone people need to search out where thcsc resources are artifacts wcrc small enough that thcy were probably available. This adaptationis not part of a sedentary portable implements and were carried fromsite to site lifestyle, but one of mobile people whose subsistence until they wereno longer functional and then discarded. relies on food gathering. Johnson(I 989) also statesthat Several structural remains wcre found onthe site lnobility was the major strategy to cope with local along with both interior and exterior features. 'I'he difficulty or take advantageof distant opportunities. shallowness of the structures would imply that the buildings were possiblyhut-like and made of brush. There were only a few postholes present; however, rodent activity was very heavy, Structure 2 had an interior posthole almostin the center of the unit, In There is not much excavation data from large sitesin Area C, Structure 4 had three postholes, but they were southeastern New Mexico. Work by Gallagher and clustered in the southwest corner (Fig.36). In AreaD, Hearden (1980) and Katz and Katz (198Sa) inthe

SYNTHESIS 14 Y Brantley Reservoir area produced sites ranging from thc Middle Archaicto the historic Seven Rivers phase. 'l'he ceramics recovered consisted of mostly brown In comparing these sitcs with Macho Dunesit is not wares such as El Paso Hrown, Jornada Brown, and surprising that brown ware ceramics dominate the South Pecos Brown.El Paso Polychrome, Chupadcro assemblages. Thereis more ceramic variety duringthe Black-on-white, Gila Polychrome, Lincoln Rlack-on- later time periods, as would be expected.'l'hc lithic rcd, Thrco Rivers Red-on-terracotla, Seco Corrugated, artifacts on sites in the arcaconsist mostly of chert and and Corona C.ornlgated madeup a total of 15.0 pcrccnt quitrtzite debitage; however, more formaltools were of the ceramic assemblage. It seems as though thc prescnt on the WIPP sites tested by Oakes ( 198s) and earlier type sites liom the Brantley Reservoir project the Mescalero Sandssitcs tcsted by Staley et al. (1996). are also commonly foundat Macho Dunes. Chalcedony was present in larger quantities at Oakcs (1985) tcsted six sites located in a dunal Hackberry Lake ((lakes 1985) and at the Mescalero environment in the Hackberry I ,ake area. 'l'he setting Sands sitcs (Staley et al. 1996) than at MachoDunes. consisted of many hummocks and blowouts with high Although chalcedonyis available, it is a minor material artifact dcnsity. However, thc vegetation on thesitus type lbr Macho Dunes. Site habitantsat other localities was very different. lnstead of little-leaf sumac, are using it more. It is evident that most lithic debitage mesquite, and one-seed juniper, the vegetation is from local sources, with the exceptionof Edwards consisted of mostly shin oak and mcsquitcthat wcrc thc Plateau chert and Tecovas chert, which are nonlocal. main food resources for these sites. There was no 'l'he ground stonc trom MachoDunes consisted of evidence of shin oak pastor present on Macho Dunes; mostly one-hand manos and basin metate fragments however, mesquite was present and was being used for that exhibited circular striations. There were many fuel wood. indetemlinate fragments with circular striations. At 'l'he lithic artifacts from all of thesesites consistcd Hackberry L,akc the ground stone consisted of one- of tnostly core flakes and angulardebris. However, hand manos and basin and slab metales.The majority several tools were found such as scrapers, bifaces, of the artifacts were indeterminatc. The Mcscalcro choppers, hammerstoncs, and utilized cobblcs. The Sands sites contained mostly ground stone fragments, tnost commonly used materials were chert,quartzite, but there wcrcmetates identil'ied as slaband basin. All and chalcedony. the manos wcrc tfagmonted and not placed into The ground stonu on regional sites consisted of categories. Ground stone was madeof sandstone atall one-hand manos andrnetates. Thc mctatcs were mostly thc sitcs, except at the Mescalero Sands sites where indeterminate fragments, butthere were some slab and other materials wcrcprcscnt. basin mctates among the assemblage.A11 the grinding The grinding implements withinthe Carlsbad area implements were made from sandstone. seem to bcall small fragmentsthat have been usedto Dates from 19 radiocarbon samples ranged from their maximum potential. Hecause ground stonc ca. 2280 B.C. to A.D. 980 with rnost falling between material is so scarce, they are used to a point where A.D. 700 to A.D. 000. Staley et al.(1096) had dates they are no longer usablc. They arc like exhausted liom 1015 to 780 B.C. to A.D. 1545 at the Mescalero cores, where flakes cannot be taken offof the parent Sands pro-joct located 52.3km (32.5 miles) east ofthe material any further. When a mano becomes project area. The majorityof the ceramics were brown inoperable, it is sometimes used for anothcr task, such warcs; however, thcrc was a broad varietyof painted as ~3 sharpening tool for the metates. Some metates are wares corresponding to the C-14 dates. 'l'he lithic recycled as manos and sometimes as hearth stones. debitage consisted of mostly core f'lakes and shatter Manos are also usedas hearth stones. with alarge number of bifaces. Like the material from In comparing Macho Dunes with other sitcs it Macho Dunes thc predominant types wcrc chertand seems as though it lalls within the category of a quartzite. 'I'ecovas and Edwards chert, although in tetnporary or seasonal camp. What makesLA 29363 so small, quantities, were present on someofthe sites. The different is the fact that shallow brush structure remains ground stonc consistedof slab and basin mctatcswith arc present. There are recordedsites in the region that many indeterminate metate fragments. The manos were contain hearths, storago pits, and roasting pits: all fragmented and were not placedinto categories of however, none have recordedstructures, except forthe one-hand or two-hand manos. Most of the grinding Hrantley Kcservoirpro.jcct. itnplements were sandstone, butthere was also basalt, Each structurc contains an interior hearth with conglomerate, dolomite, andquartzite. several pits. Postholes are present, but are scarce, making it difficult to tind anypatterning. Hut why are thcrc interior hearths in the structuresthat are supposed

I 50 CARLSBAD RELIEF ROIJTE to be occupied during the late spring to early fall? site was occupied at different times of the year by There could be several sccnarios. Pcrhaps theclimate diffcrcnt peoplcs. It may haw served as a camping spot was much coolcr during the Globe phase than it is by people passing through throughoutthe ycar andalso today, or maybe the structures were protectionfrom the served as a place where wild fond resources were stifling heat that is common today. It is likely that he collected during the late sulnmerand early fall months. I52 CARLSBAD RELIEF ROUTE REFERENCES CITED

Adams, Jcnny I.,. Rellzled to Salt Dissolution in Southenstern 1988 Use-Wear Analysis on Manos and IIide- New Mexic.*o.Open File Report 74-194. U.S. Processing Stones.Jolrrnal ofField Geological Survey, Washington, D.C. Archnrology 15307-3 IS Bailey, Vernon Allen, M,S., and E. E. Foord 1971 Mmvvvds qf'the Sr1uthwc:strrn linited Stuks, I99 1 Geological, Geochemical and Isotopic with S1x.ciaI Rc;fiwncc.s to Nrw Mcxicv. Characteristics of the 1,incoln County Dover Publications,New York. Porphyry Belt, New Mexico: Implications for Regional Tcctonics and Mincral Nwhu-t, Harry W. Deposits. In NPW Mexico Gcwlogicnl 1974 Mcscalero Apache Subsistence Palterns and So&@ Guidchook, 42nd k'ield Conference Sociopolitical Organization.Apnchc h3rrn.s G~ologyqf thc, Sicvxr Blmcn, Swrunwnto, HI. Garland Publishing Co., NewYork. Cqitnn Rrrngcls, New Mexiro, edited by .lames M. Barker, BerryS. Kues, GeorgeS. Ratcho, DavidG., David I.,. Carmichael, Meliha Austin, and SpencerC'r. Lucas, pp. 97-1 14. Duran, and Margarct Johnson New MexicoC;cological Society, Socorro. 1985 Archncologicnl Invc.sti~rrtir)n.s (?f,Siites Locrrted Lrt the SouthernDoria Anu County Allen, M. S., and V. T. McLcmorc Airport Srwtcl 'lkrcw. NPW Mexico. Part I. 1991 'Ihe Geology and PetrogenesisoT thc Cultural Rcsourcc Management Division Capitan Pluton, New Mexico. In New Report No. 533. New Mexico State Mrxico C~cologicdSor:ir.l,y Guidebook, 42nd University, Las Cruces. Field C'onfl.rcnc.r Gtwlo~of the Sicrrll Rhnca, Srrcwmento, Cupitnn Rangrs, Nw Binford, Lewis R. Mmicw, cditcd by Jamcs M. Barkcr, Berry 1977 Ik'orty-scvcn 'Trips.A Case Study in the S. Kucs, George S. Austin, and Spencer G. Character of Sotne Formation Processesof I mas, pp. 1 15- 125. New Mexico the Archaeological Record. In Stonp I'ools Geological Society, Socorro. n.s Culturd Mnrkr.s, Clwngr, Evolutionrrnd Complexity, edited by R.V.S. Wright, pp. Anyon, Rogcr 24-36. Australian Institute of Aboriginal I OXS Arr.hrrr~ologi~:rrlTesting nt thr Fuiwhild Sitr Studies, Canberra. (LA 45 732), Otero Countv,New Muirw. Office of Contract Archeology, University 1979 Organization and Formation Processes: of New Mexico, Albuquerque. Looking at Curated Technologies.Journtrl qf'Anlhroyologica1 Research 35:255-273 Applegarth, SusanR. I976 PrL.historic [Jtili:ntiorr of the I:'nvironmcntof' I980 Willow Smoke andIhg's 'Tails: Huntcr- thr Kmtwn Slops qf'thc G'uaddlp Gatherer Settlement Systems and Mollntnins, Solrthr.nstrm NPWMcx-ico. Archaeological Site Formation. American Unpublished Pl1.D. dissertation, University Anliyuity 45:4-20. of Wisconsin, Madison. Bleed, Peter Atcn, Lawrence E. 1986 The Optimal Designof Hunting Weapons: I972 Evaluation of'thcr Cdtlrrrrl ResourctJsof tho Maintainability or Reliability'! Amwicm Northgatc sitr, ElPuso Colrrrtv. T~xns. .4ntiquity 51(4): 737-747. Texas Archaeological Salvage Project Research Report 5, University of Texas, Bohrer, Vorsila L. Austin. I994 Plant Remainsfrom the Wind Canyon Site in the Eagle Mountainsof Western Texas. Bachman, G. Appendix B. In Drrtrl Rt~c~ovryE.x-amwtion.r I974 Gcdogic Procrssrs and Cenozoic History Lzt thr Wind Canyon Sitr, 41M119,

REFEREN(:ES C'ITED 153 11dy/)~th(hung:, ’I’cxrrs,by M. 11. IIines, S. C’atnilli, Filecn A. Tomka. andK. W. Kibler. Reportsof 1988 Lithic Raw Material Selection andlJsc in Invcstigations 99. Prcwitt and Associates, the l1csu-t Basins of South-Central New Austin, Texas. Mexico. Thp Kivn 53(2):147- 16I.

Bohr, Vorsila L., and KarenR. Adam 19x9 ‘T’huOccupational Historyof Sites and the 1977 Ethnohotnniral Techniques and Apprr>ach>s Interpretation of Prehistoric Technological ut the Salmon Xlrin, NPW Me>A-ico.San Juan Systems: An l’.xamplc Iiom Cedar Mesa, Valley Archeological Project, ‘l’cchnical Utah. In ‘lime, I:’ncrg-y.and Stone Tools, Series 2. Eastern New Mexico University edited by Robin Torrance,pp, 17-26, Contributions in Anthropology X( 1). Cambridge University Press, Cambridge. thwman, K., 1. Montgomery, and D. Landis Carnilli, E., I .. Wandsnider, and J. I. Ebcn 1990 A~~c:ha~~ologir~alIn~vs/igation.sat LA 21 I77 I988 Distrihlrtionnl Sutwy and /Yycuvntion of’ nnll LA 27676, Ell& Collnty, Ntlw Mc>xico. Arrhncologicnl Lnndscups in thc Vicinitv Agency for C:onscrvation Archaeology, qf’E1 Paso, ‘lex-rrs. 1J.S. Department of the Keport CQ 86.3. Eastern New Mexico Interior, Bureau of I ,and Managemcnt, Las University, Portales. C:ruces.

Boyd, DouglasE., Jay Peck, and KarlW. Kibler Campbell, K. S. 1<>

154 REFERENCES CITED Chapman, Joseph A,, J. Gregory Hockman, and Degenhardt, William G., Charles W. Painter, and William K. Edwards Andrew H. Price I982 Wild Mammals of North America: Iliology, 1996 Amphibians nnd Rq>tilesqf'hicw Mtxico. Munugement, und Economicx, edited by University of New Mexico Press., Joscph A. Chapman and George A. Albuquerque. Feldhamer. Johns HopkinsUniversity Prcss, Haltimore. Dick-Pcddic, William A. 1993 New Mexico VegetntionI'rrst Present und Collins, Michael H, Futul-r. University of New Mexico Press, 1968 The Andrews Lake Locality: New Albuquerque. Archacological Data from the Southern Llano Estacado, Texas. MA.thesis, DiPcso, Charles C., John B. Rinaldo, and GloriaJ. University ol'Texas, Austin. Fenner I974 Cmrrs G'rund.s, A FdmTrading Center of' 1985 Geomorphology of the Lakewood Alluvium the Grun Chichimwa, vol. 6, Ccwrmics rrnd in the Brantley Area, Eddy County, New Shell. Anlerind Foundation SeriesNo. 9. Mexico. Appendix 5. In The Prehistory of Dragoon, Arizona. thc. Cdslmd Brrsin Southeasterm New Mmico, edited by S. Katz and P. Katz, pp. Dix, K. I.. A35-AI 18. Incarnate Word College, San 1962 A History ofBiolic.,mt-l Climntic- C:'hangcs Antonio, Texas. within the. North Americrrn Grusslmd. Contribution No, 33 1. Department of Plant 1991 'Thoughts on I'uture kxcavation of Burned Ecology, University of Saskatchewan, Rock Middens. In The Burned Rock Saskatoon. Middms of.Tc.x-rrs:An Archurologicd Symposium, edited by 'l'hotnas K. Hcstcr, Dodge, Wendell E. pp. 1-24. Studies in Archeology 13, 'I'exas I982 Porcupinr in WildMummals ?{North Archeological Research Laboratory, Amprica: Hiology, Mnnngement rrnd University of 'l'exas at Austin. Economics, edited by JosephA. Chapman and George A. Feldhamer. Johns Hopkins Cordell, Linda S., and Fred Plog University, Haltinlore, Maryland. 1979 Escaping the Confincs of Nonnativc 'l'hought: A Re-evaluation of t'uebloan I)ocblcy, John F. Prehistory. Anzcricrrn Antiquity 44:405-429. 19x4 "Seeds" of Wild Grasses: A MajorFood of Southwestern Indians.Economic Hotan)? 3X( 1):52-64.

Doleman, William 11. 1092 Analytical Approaches To Basin Floor Distributions. In Archnrological Distributions and Prehistoric Human &dogyVvol. 3, Lnnllscuppe Archeology in Cresson, John H. lhc Southern 1ulnrosa Rnsin,cditcd by 1977 Keply to: The Myth of Hipolar Flaking William 11. Doleman, Richard C.Chqman, Industries, by J. Solbcrger and L. Patterson. Ronald I ,. Stauber, and June-el Piper. Lithic 7kchnology6(3):26. Ccntcr for University of New Mexico, Office of Archaeological Research, Universityof Contract Archeology, Albuquerque. Tcxas at San Antonio. Doycl, D. Dane, Carle H., and George 0.Rachman 1974 Exrnvations intht. Esr:almlP Ruin Group. I965 Grnhgic Mcip of New Mr.xico. IJ.S. Southem Arizorlu. Arizona State Museum, Geological Survey, Washington,D.C University of Arizona, Tucson. Erdrmm, G. Pithouse Use. In Mogollon VmiihiIit,y, 1960 TI1 Acetolysis Method:A Revised edited by C'. Henson and S. IJpham, pp. 203- Ucscription. S1vnsk Botnnisk 7Xskrijt Ud 209. Occasional Papers No. IS Las Cruces: 54:561-564. Univcrsity Museum, New Mexico State Univcrsity, Las Cruces.

1991 Changing LandUse and Pit Stnlcture Seasonality alongthe San Simon Drainage in Southeastern Arizona. In Mogollon V, pp. 1 1-27. cOAS Publishing and Research, I ,as Cruces, New Mcxico.

Goodyear, Albert C. 1974 The Brand Sito: A 'l'echno-T:unctional Study of a Dalton Site in Not-theast Arkansas. Ar-kmms Archneologicd Rcsrardl S't'ricy 7:73-76.

Greer, JohnV. 1 965 A 'l'ypology of Midden Circlesand Mescal Pits. Soulhwcstrm Imr3 1 (3):41-S5. Colorado Archaeological Society, Boulder.

Findley, James S.,Arthur TI. IIarris, Don E. Wilson, 1067 Noles on Excavnietrl King Mlil'n'En Sitc?s. and Clyde Jones IY63-1Y6K.Bullelin of the Texas I975 M(1mrnals of NwMaico. University of Archaeological Socicty,vol. 38, pp.39-44, New Mcxico Press, Alhuquerquc. Austin.

Ford, Richard I. 1968 hkcnvntions rrt r~ Midrh Circle. Site) in El I077 Archeobotany of the Fort Bliss Maneuver Pnso County, '/kxcxns. Hullctin of thc 'I'exas Area 11, Texas. Appendix E. In Settienwnt Archaeological Society, vol. 39, pp. I I 1 -

Prltterns of.thr Errstrr-n Huc~coBolson, by 13 1~ Austin. Michael E. Whalen, pp. 199-206. ElPaso Centennial Museum, Publicationsin Gunnerson, DoloresA. Anthropology 4. University of Texas, El I956 'l'he Southern Athabascans: Their Arrivalin Paso. the Southwest. LII Palario h3( 1 1 - 12):346- 305. t'rizell, Jon 1985 Ceramics. In ..lr.c,hrrenlogicrrl Imwtitntions IIall, S. A. of 7hr-(.eSites Whitin The WlPP Core Arm. I977 Late Quaternary Sedimentation and Edirir:n Chambers Consultants and Planners, Bulletin 88:1593-1618. Albuquerque. 1985 Quaternary Pollen Analysis and Gallagher, J. Ci., and S. E. Bearden Vegetational History of the Southwcst. In 1980 Elwluntion qfCuIturn1 Resources ut Polken Rwords qf'theLatcJ-Quaternarv Smntley Rescr1y)ir. I:it!v Count-v, New North Amckwl Sc4iments, cdited by V. Maico. Archaeological Research Program, Bryant and K.Holloway, pp. 95-123. Llcpartmcnt of Anthropology, Southern American Associalion 01 Slratigrapliic Methodist University, Dallas. Palynologists, Ilallas, 'I'exas.

Gilman, PatriciaA. Hard, RobertJ. 1986 Seasonality and Mobility in San Simon 1983a A Model for Prehistoric Land Usc, Fort

IS6 REFERENCES CI'I'ED Bliss Texas. In American Sociev E~rrlvMrrn Site in Ermtern NLW Me.xica. Fort ,for Consewalion Archaeology Burgwin Research Center, Southern Procwdings I Y83, pp. 4 1-51. Methodist University. PublicationNo X. Fort American Society for Conservation Burgwin Research Cenler, Ranchos de Taos. Archaeology, New York. IIester, Thomas R., AlanD. Albee, and Cristy Willer I9X3b Excavations in the Custner Rungr 1977 Lithic Analysis ol'a Controlled Surface Archaeological DistrYct in El Puso Tmm. Collection from a Site in Western Montana. Publications in Anthropology no. 1 I. Plains Anthropologist 22: 239-244. Centennial Museum, Universityof Texas, El Paso. Hicks, Patricia 198 1 Mitigation of'b'our Arc:lzncological Sites on 1990 Agriadtrmrl Lkpmdmw in the Mountuin In thr Wusle Isoldon Pilot Projc.'1 : 1 new Pwspecti\ws on South westwn Prc>hislor:y, Ctrrlsbad, New Mtxico.fbr- WrJstinghouse,

edited by P. E. Minnis and C. L. Rcdman, h).Agency hrConservation Archaeology, pp. 135- 149. Westview Press, Boulder. Eastern New Mexico University,Portales.

Hill, David V. 1991 Settlement Patterns and Ceramic Production in thc Paso dcl Nortc. In Actrrs dd Srcunrh Chngreso Hi,stor.ia Kq$onal Comparnda, edited by Ricardo Leon Garcia, pp.29-44. Universidad Autonomade Ciudad Juarez, Ilawley, J., G. Bachman, and K. Manlcy Mcxico. I976 Quaternary Stratigraphy in the Basin and Range and Great Plains Provinces. In 1996a Ceramics. InArc-htreological iresting ut Quatrmnry Strrrtigruphy ofNorth America, 109291, LA 109292, rrndLA 109294: Silcs editcd by W. C. Mahaney pp. 235-274. Along thc Potmh .Junction to I.M. C. #I Uowdcn Hutchinson and Ross, Stroudsburg, Eddy C'ormty, New Mexico, vol. I, Pennsylvania. Mest-~zleroPlrrins Archaeology, prepared by D. P. Staley, J. T. Abbott, K. A. Adams, I). IIaynes, G. V. Hill, R. G. Holloway, W.D. IIudspeth, 1977 Keply lo: ThM-vth qf'Bipolur Flrlking R. B. Roxlau, pp. 59-64.'l'echnical Report 1rrd~rstrie.s.J, Solherger andI Patterson. 1 1034-00IO, 'I'KC Mariah Associates Inc., Lithic Tcchnologv 6(3):40. Center l'or Albuquerque. Archaeological Kcsearch, Universityof 'l'exas at Sa11 Antonio.

Henderson, Mark 1074 An Ar,c:htrrologic.rrl Inventon, of.BruntIq Xcsrv-\wir. NPW Mexico. Archaeological Research I'rogram Contributions in Anthropology No. 18, Department of Anthropology, Southern Methodist University, Dallas.

Hendrickson, G., and R. Jones 1952 Gcology Lrnd Gr.ound- WderR~wurc.rs of' IIitchcock, Robert Ed& County. New Mexico.Ground-Watcr 1982 Patterns ol' Sedentisnl among the Eastern Report No. 3. New Mexico Bureau of Mines Hasanva of Botswana. In Politics rrnd and Mineral Resources, Socorro. Hislory in HmtJ Societies, edited by Eleanor 1,eacock and Richard Lee, pp. 223-267. Hester, J. Cambridge University Press, New York. 1972 Blucbvnter Lodity No. I: A Strrrtified

REF'ERENC'ES (.'ITED 1.57 IIol'lineister, Donald F., and Luis De La Tone 1960 A Revision of the Wood RatNeotomu stc?phcnsi..Journal of'Mmmolo~v 41(4):477.

Hokanson, JcffrcyH. 1996 Arr.hnlco1ogic.d SLIYVL'~dong North Loop Hold C'lrrlshad Lcldy C.'ounty, Ntw Maim. Report no. 3. Lone Mountain Katz, Susana R., and PaulKatz Archaeological Services, Inc. Albuquerque. A Summary ofKecent Archaeological Investigations in Guadalupe Mountains IIolloway, Richard G. National Park by the Deparlmenl of 198 1 Pwstwation (rnd Lxper~imentnlDingrncsis Anthropology, Texas 'l'ech University. oftho Pollen E-xinr. Ph.D. dissertation, Y'PX~S A~C~UPOZO~V19(2): I 1-1 5. Texas A & M University, College Station. A Status Reporton Research into the Hughes, Jack 'I,,, and P. Willcy Prehistory of Guadalupe Mountains 1978 Archwolog2v ui McKonzic Rescnwir. Office National Park, Texas.In Jornrrth Mogollon of thc State Archacologist. Survcy Report Archlnrd>g,:v:Prorx?cdings ofthc fi'irst 24. Texas Historic Commission, Austin. .Jornrrrlu Cbnfiwnce, edited by1'. H. Heckett and K. N.Wiscman, pp. 359-370.New Jelinek, Arthur J. Mexico State IIistoric Preservation Division, 1966 Form, Function and Stylein Lithic Analysis. Santa Fe. In Cul/urrrl C'hrrnge mt-l C.'ontinuity: I:'ssn,vs in Honor of'JumesBennPIl G'/*([fin,edited by 77w Prc4istory ofthe CdshLd Basin, Charles E. Cleveland, pp. 19-33. Academic Southcnstern Ncjw Mcx-ico. Kcport prepared Press, New York. by the Anthropological Research Facilityfi)r tlie Bureau of Reclatnation, Southwest I967 A Prphi.stor+icStylmm? in thr Middlc Pccos Regional Oflice, Amarillo, Texas. C'dIt~y, NPWMt7,~iro. Anthropological Papcr No. 3 1. Universityol"ichigar1 Museum of Prcns Put. thr? I+",history nnd Histoly of' Anthropology, Ann Arbor. thc Brunt1c.y I'rojjecf LocLlIiiy. Bureau of Reclatnation, Southwest Kegional Office, Jennings, Jesse I). Amarillo, Texas. I940 A Vwialion I!f',SoufhMIt~,stErY1Purblo Culturr. Teclmical Series BulletinNo. 19. An Archaeological C)vcrvicw of Laboratory of Anthropology, Museum of Southeastern New Mexico. Prepared for New Mexico, Santa Fe. New Mcxico IIistoric Preservation Division, Santa Fe [llraft]. .lohnson, Gregory A. 1989 Dynamics of Southwestern Prehistory: Far Kelley, Jane Holdcn Outside-1,ooking In. In Uynnmics of. 1984 The Ar.charolog,y of'thcSitwrr Rlrrncn Southwrsi Prehi,vfory,edited by L. S. Kegion clfSollthmstL.rl? NC'WMc.xic*o. Cordell and G. J. Gumerman, pp. 371-390. Anthropological Papers of the Museum of Smithsonian Institution Press, Washington, Anthropology No. 74. Ilniversity of D.C. Michigan, Ann Arbor.

Joyce, D. J., andD. G. Landis Kelly, V. 1986 Archarologicnl Yesting andL~duntion of 1971 Geologv of'the PPCOSC7014nfr:V, Southcrrstrtw Site IA322 76, Edl

Lehmer, Donald J. Imd, Kenneth .I., and WilliamE. Reynolds I948 The .Jornada Brunch ofthe Mogollon. Social 1985 Ar.chapolcgica1 Inlwtigations of M~PPSites Scicncc Hulletin 17. Univcrsity of Arizona, Within theWIPP Cow Area, I

Minnis, PaulE., and Alan J. Wormser 1984 Late Pithouse Period Occupation in the Detning Region: Preliminary Report 01' Excavations al the Florida Mountain Site (LA 18839). In Rrcwt Resewdl in Marshall, LarryG. Mogollon Arc+ncolog~l, edited by S. Upham, 1989 Bonc Modification and "'l'he Iaws of F. Plog, D. Batcho, and H. Kauffman, pp. Burial." In Bone Modificntion, edited by 229-249. Occasional PapersNo. IO. Ncw Robson Bolmichsen and MarcellaH. Sorg,, Mexico State University,Las Cruces. pp. 19-20. <.!enter for the Study of theFirst Americans, Orono, Maine.

Matthew, A. J., A. J. W(lods, and C:. Oliver 1991 Spots Beforethe Eyes: New Comparison Cham Lor Visual Age Eshnationin Archaeological Material. In Kccmt Devrlopmmts in Cernmic. Prtrology, edited by Andrew Middleton and Ian Freestone, pp. 2 1 1-264. British Museum (kcasional Murray, Harold D. Paper No. X 1. British Museum Research 19x5 Analysis of Molluskan Materials from the Laboratory, London. Brantley Project Area, Eddy County, New Mexico. In The Prehisfoty qf'lht>C(w1,shd Mauldin, Kaymond, Tim Graves, and Mark Hentley 13asin, Solrthcwstcv-n NPW Mcj.xica, by Susana 1008 Small Silcs in [he C'mlrrrl H~rt~:oRolson: A Katz and Paul Katz, pp. A-24-27. Prepared Find Rcport on I'rojwt 90-1 I. by the Anthropological Research Facility for Conservation Ilivision Ilirectorateof the Bureau of Reclamalion, Southwest Environment United Slates Amy Air Regional Office,Anmillo, 'l'exas. Defense Artillery Center, Fort Hliss. Nelson, Margaret C. Medellin-Leal, Fernando 1984 Research Design: Organization of 1982 The Chihuahuan Desert. In Rclfc;r(mv Seltlemenl and Technology. InLarlrlc,r- Hundbook of'the Dtwris qf'North America, Ranch Kcsc~rr~hPrqjcw: A Keport qf'thr? cditcd by Gordon Bender, pp. 32 1-38 I. Fimt Stwon, edited by Margaret C. Nelson, Grcenwood Press, Westport, Connecticut. pp. 9-14. Tcchnical SeriesNo. I. Maxwell Museum of Anthropology, Albuquerque. Mcra, H. P. 1943 An Outlinr of Cc.rmzic Developments in 199 1 The Study of Technological Organization. Southcvn rrnd Soutlwastern NPW Makw. In Archaerhgicul Method and Th~ory,Vol. Laboratory of Anthropology, 'I'echnical 3, cditcd by M. B.Schil'fcr, pp. 57-100. Series 1 1 Santa Fe. Univcrsity of Arizona Press, Tucson.

Miller. Myles A. Newcomer, M. H. 1994 Jornada Mogollon Residential Settlements I97 I Some Quanlilative Experimentation in in the Salt Flat Hasin and Delaware Handaxc Manufacture. World Archaeology Mountains of West Texas. Mogollon Vlim 3:x5-93 pp. 105-1 13. COAS Publishing and Research, I ,as C'ruces.

I60 REFERENCES CITED hisforicSubsistmce Allrlyllrfions un Whitc Smds Missile. Rangc. Laboratory 01' Anthropology Notes 277. Museutn of New Pcasc, D. s. Mexico, Santa Fe. 1967 Opal Phytoliths as indicators of'Paleosclls. Unpublished MA. thesis, New Mexico State 1982 Prehistoric G~~rheringSites new ilnckbercv University, Las Cruces. I,ake, Eddy Counfy,New Mexico. Laboratory of Anthropology Notes 305. Peckham, Stewart Museum of New Mexico, Santa E'e. 1990 From this Enrth: The AncientArt (?f~Pwelh Pottrrv. Museum ofNcw Mexico Press, 1985 An Assessment of Gathwing SitcsNcwr Santa Fe. H~rc.klxwyLaX-c, Edlv County, NPW Me.xic.0. I ,aboratory of Anthropology Notes No.4 15, Phelps, Alan I ,. Museum ol'Ncw Mcxico,Santa Fe. 196X A Recovery of Purslane Seeds in an Archaeological Context. The Artlfi1c.t6(4): 1- 1997 Site-Spccil'ic Data Rccovcry Quoslions. In 9. El Paso Archaeological Society. 'lksting Resultsand DatLr Recovery Plan,fiw the Cdsbcrd Rclief.Route,Eddy County, Ntw A4txic.u. Ol'ke of Archaeological Studies, Museum of New Mexico, Archaeology Notes 2 19, Santa Fe, Ol'fice of Archaeological Studics Staff' 1994 StnndaJnr-clLithic Art@ct Annlysis: Attributes Kcagan, Albcrt 13. ond Vur.iddc>C'odes Lists. Museum of New 1928 Plants Used by the White Mountain Apache Mexico, Office of Archaeological Studies, of' Arizona. Wisconsin Arr.htw1ogis.l Archaeology Notes 24C, Santa Fe. 8(4):143- 1 h 1.

O'Connell, .lames F. Kccd, Lori Stcphcns, C!. Dean Wilson, and Kelley 1977 Aspects of Variation Central Australian Hays-Gilpin Lithic Assemblages. In Stonp Tuol,s LIS IO08 Raskettnaker Ceratnic 'l'cchnology: From Cultur.nl Mrrrkws, Change, Evolution,and Early Ceramic Horizon to the I)evelopmunt C;'ornylr?xi~,v,edited byK. V. S. Wright, pp. of Regional Traditions. Manuscript in 269-28 1. Australian Instituteof Aboriginal possession of author. Studies, Canberra. Reeves, C. 0'1 mghlin, 'lhomasC. 1972 'I'ertiary-Quatcrnary Stratigraphyand 1980 Thc, Keystone DrmSitc und Otlwr hchnic Geonlorphology of West 'l'exas and and Formntivr 5'itc.s in Northwest I:'l Pmo, Southem New Mexico. InGuidebook 23'" Tt.rtrs. Publications in Anthropology8. El k'ield C:'onfi.renw, hkw Mc.xico Geologic*Lrl Paso Ccntcnnial Museum, University of Socir1,v: Eusi-Cenird New ML.x.ic.o, edited Texas, El Paso. by V. Kelly and F. Tranger, pp. 1OX-117. New Mexico Geological Socicty. Parry, William .I.. and John I). Speth I084 Thr. C;mn,sqv ,@ring CLtmpsilp: Lrrf~ Kobbins, W. W., 1. P. Harrington, and B. Freire- Prehistoric. Oc.c.uprrtion in SouthensternNPW Marreco Mcxico. University of Michigan Museum of 19 16 l:'thnobotcrny of thr TPMWIndiirrns. Bureau of Anthropology, I'cchnical Reports No. 15. American Ethnobology Bulletin SS. Kesearch Keports in Archaeology Smithsonian Institution, Washington,D.C. Contribution IO. Ann Arbor.

Pearsall, D. M. 1988 Phytolith Analysis:An Ar.chrreo1ogic.d md

REFERENCES (.:ITEL> 161 PCL'OS Vdl~y,Muscum of Anthropology Soutlzwrstrrrr t'rcd1istoq~.edited by Paul E. .l.echnical Report4 No. 1 X, [Jnivcrsity of Minnis arid Charles I ,. Kcdman, pp. 103- 12 1. Michigan, Ann Arbor. Westview Press. Houldcr.

Rogers, Rose Mary 199 1 On Manos, Mctatcs and thc IIislory of Site 1972 Ceramic, Pecos Kivcr llrainagc Pccos and Occupations. Amuicorz Antiquity 56:460- Crocket Counties Texas. In Trunsrxclions of 473. thr. Srwnth Kqionnl Arc.har.olo@~d Symp7osium.fi)r S'r,uthrLatrm ncw Mcxic.o Schroeder, Albert H. (Assembler) mrl Wcstrrn Tcn~s,pp. 47-70. Lower Plains 19x3 The Pratt Cave Studies.7hc Artlfi1firc.t2 1 (1 - Archacological Socicty Prcss, Midland, 4), E1 Paso Archaeological Society. 'l'exas. Scott, Linda .I., and Mollie S. 'l'oll Koney, John 1987 Analysis of Pollen and Charred Macrofloral 1985 Prehistory of the Guadalupe Mountains. Remains with Identification of Charcoal Master's thesis, Dcpartmcnt of Samples lkom the NAHO-Santa Teresa Anthropology, Eastern New Mexico Excavation Projcct. In Prvhistoric hnrl University, l'ortales. Us[?in thr SollthrrnMcsillrr Bolson: Exc~rvnlior~son thr Nmqjo-llopi Imd

Rozcn, Kenneth (.I., and Alan P. Sullivan Ill Ex-c*hrmgclnom Srrntrr Twesn, Nt'w Mcriccl, 1989 The Nature of Lithic Reductionand Lithic by Beth 0'1,eary. Office of Contract Analysis: Stage Typologies Revisited. Archeology, University of New Mexico, Amcvknn Antiquity 54: 179- 1 84. Albuquerque.

Kunyon, John Sehastian, Lynn, and Signa Larraldc 1972 The LagunaPlata Site L.C.A.S. C- 10-c* LA- 1989 Living on tlrr I,nnd: I1,OOO Ycnrs ofHmwr1 5 148. In Trrrnsrrctions of.thL' Srl*mth Allrqktion in Southrralrrn N~rvMc..x.ico. Kr>,qionulArchueologicrrl Sympo,sirun.fiw Cultural Kesources Series no. 6, New Soulhenstrrn rim' Mcx-icw and W&ern Mexico State Office,Bureau of Imd Tcxas, pp. I0 1 - 104. Lower Plains Management, Santa Fe. Archaeological Society Press, Midland,

1987 Pottery Types ofthcSouthwest Federation of Archacological Societies (SWFAS)Area. Thr AI*I~/~~C.I25(4):23-59. Sellards, E. 1952 Eulv Mmin Arncv-icu. University of Texas, Schermer, Scott C. Austin. 1983 Testing rmlEwlurrtion of'AR-690. Agency for Conservation Archaeology, Eastern New Shcpard, Anna0. Mexico University, Portales. 196.5 CcrnrniL:s.fi,r tkc Arc.hrrcologist. Carnegie Institute of Washington, Publication 609. SchifYcr, Michael n. Washington, D.C. 1976 Bchviornl Archn~olo,q,v.Acadetnic Press, New Ynrk. Sheridan, '1'. I975 77w Ilittcr Kivcr: A I3ric

Spcth, John Stuart, Trace 1975 Miscellaneous Studiesin IIard-IIammer 1990 Exzuvation qf-u Sing(. Mcsilln Phlrsc Percussion Flaking: TheZSffccts of Obliquc 1’ithou.w in the Town of.Tortugas, DO~UAnrr Itnpact. Americm Antiquity 40:203-207. Cbunt,y, New Mexico. Batcho andKauffrnan Associates Cultural ResourcesReport No. 106. Las Cruces.

Sullivan, Alan P.,111 1995 Artifact Scatters and Suhsistcncc Organization. Journnl qJ!l‘FitddArrlt~u?oIog-y 22:49-64. Stallord, C. R. 1977 B@olar Tt.chdoD d the Escalnnte Ruin Sullivan, AlanP., 111, and Kenneth C. Rozen Group: Fkrkc. Production and litilizdon. 1985 Dchitagc Analysis and Archaeological llepartment of Anthropology, Arizona State Interpretation. American Antiquity University, Tempe. 50(4):755-779.

Stahle, David W., and JamesE. Dunn Sullivan, K. M. 1982 An Analysis and Applicationof the Size 1973 The Archaeology of Mangat and Some of Distribution of Waste Flakes from thc the Problems of Analyzing a Quartz Manufacture of Uifacial Stone‘l’ools. World Industry. Unpublished L3.A. Honors Thesis, Archarology 14:84-07. Department of Anthropology, Universityof Sydney. Stalcy, DavidP., Kathleen A. Adams, Timothy Doh, Jolm A. Evaskovich, DavidV. Hill, Richard Swank, GcorgcR. G. Holloway, William H. Hudspeth, and K. Blake 1032 The Ethnobotany of the Acoma and Laguna Koxlau Indians. Unpublishcd MA. thesis, 1096 Archa~wlogiml Invesfigdionsdong lhr University of New Mexico, Albuquerque. Potnsh .Junc*tinnto Cunninghrrrrr Strltioll 7kcmsnzission Line, Lrnd IxLz C’o1wtir.s. New Mcrirw. I’RC Mariah Associates, Inc., Albuquerque.

Stark, Miriarn T. 1995 The Early Ccrarnic Horizon and Tonto Hasin Prehistory. InThp Rooswell Toll, Molly S. Community O~lvdoprntwtStudy, vol. 2, 1983 I+’lotationand Mucrohotnnicnl MutPrials C’c.rmic C’hronology,Twhnology, and Rer.wrrrd.fi.om 7’wo Ckwmir Period Sites Economics, editedby J. M. IIeidkeand M. in lh~T~ulurosn Basin, New Mexico. T. Stark, pp. 349-296. Anthropological Prepared for Human Systems Research, Papers No. 14, Center Lor Deserl Tularosa, New Mexico. CastetterI ,ahoratory Archaeology, Tucson. for Ethnobotanical Sludies, TechnicalSeries 90. University of New Mexico, Albuquerque. Van Devender, T. 1980 Holocene Plant Remains fromKocky 1986 b’lotntion Analysis of‘TwoSites,from Whitr Arroyo and Last Chance Canyon,Eddy SwhMissile. liangc-.:HSR 8524-5 and County, New Mexico. Thp Soulh\vr?stem 8524-16. Prepared forHuman Systems Nuturnlist 25(3):361-372. Research, Tularosa, New Mexico. Chstetter l,aboratory for Ethnobotanical Studies, Vernon Ralph Technical Series 174. University New of 1976 Development of Prehistoric House Typesin Mexico, Albuqucrquc. the Jornada Branch.Awanyrr 3(4).

1993 Botanical Remains from a Jornada Wrtmica, J. Mogollon Campsite (LA 1644) near El 196 1 The Elida Sites, Evidenceof a Folsom Paso, ‘l’exas. Appendix 2. In Arrhewologicrrl Occupation in Rooscvell County, Eastern Ex-cwntion rrt the Cristo RqSilt: S1cnIrrnt-l New Mexico. Toxas A~~c.hncologicdS‘ociq? Pwk, Uon’rr Anu County, New Muicw, by lrldlctin 301209-215. Dorothy A. Zamora. Archaeology Notes63. Museum ofNew Mexico, Office 01’ Wetterstrom, WilmaT:. Archaeological Studies, Santa Fe. 1978 Plant Remains from Mesilla andEl Paso Phase Sites ofthe Hucco Botson: A 1997 Prelirnintyv Ovtwiw ?f‘Rotrrnicullindings Preliminary Report. AppendixF. In at LA 29.363. Ecrst of.Cu%vhrrd,NPW Mcxico. Srttlcment Pertterns c?f thr Wcstrrn Hueco Office of Archaeological Studies, Bolson, by Michael Whalcn.F1 Paso Ethnobotany Lab Technical SeriesNo. S I. Ccntcnnid MuseumPublications in Anthropology, vol. A. University of Texas Toll, Mollie S., and Pamela .I. McBride at El Paso. 1997 Botanical Notes from Field Surveyof ‘I‘hrec OAS Projects in the RoswclliCarlsbad Area: Whalen. Michaell?. Salt Creek #4 1.648 1, Ked t,ake Tank 1977 Sottlemt?nt Prrttcms qfthr 1:’nstrrn Hurco ft41.6461, Carlsbad Relief#41.64 I I. Ms. on Bolson. El Paso Centennial Museum, lile Museum olNew Mexico, Oilice of Anthropological Paper4. University of Archaeological Studies,Santa Fc. Texas at C1Paso.

Torrance, Robin 1980 ‘L’hc Pithoust. Period of South-Central New I983 ‘l’imc Hudgeting and Hunter-Ciathcrcr Mexico. In An Ar.c.haroIo~ical,Syntkcsis o/‘ Technology. In HMntpr-Grrtltprt.,. Emmm,y South-Ctwtrrrl rrnd Southwcstrm Nrw in Prehistory: AEuropcwn Pcxspc~ctivr, Mrxicn, cditcd by S. LeBlanc and M. edited by G. Bailey, pp. 11-12.C’ambridge Whalen, pp. 3 18-386. Prepared for the University, Cambridge. USDI hrcau of Land Management. Onfile at the Laboratory of Anthropology, Sam Tuan, Yi-Fu, C. E. Everard, .I. Ci. Widison, and 1. Fe. Bennett 1973 Clirnute qf’NwMc.xico. Rcvised Edition. 198 I The Origin and Ilvolution of Ccramics in New Mexico State Planning Office, Santa Western Texas.Bllllefin of ‘the ‘lkxns Fc. Archneologicrrl Societ?; 52:215-229.

United States Department of Commerce, Weather 1994a Turquoise Ridge unll Lute Prt.hi,storic Bureau Residenriul Mohilitl)in the Desert Mogollon 1965 C:limnlt>Summrrry qf‘thr UnitrdStatm Rqion, University of [Jtah Anthropological Supplement for I9S I through 1060: New Papers No. I 18, University of Utah Press, Mexico. Climatographyof the United States Salt Lake City. No. 86-25. Washington, D.C. I Y94b Moving outof the Archaicon the Edge of the Southwest. American Antiyuity 59:622- 638.

164 KEFEKENCES CITED While, J. P. Witnbcrly, Mark, and Peter Eidenbach 1968 Fabricators, Outils, Ecailles or Scalar Cores? 198 1 Preliminary Analysis of Faunal Remains Mmkind 6( 12):658-666. From Fresnal Sheller, New Mexico: Evidencc of Differential Butchering Whitficld, C. J., and H. L. Andurson Practices during the Archaic Period.In 193% Secondary Succession of the Desert Plains At.r:h~r~cll~~~~ir:alEssnys in Honor r?f.Murlr Grassland. I:'c:ology 1926-37. Wirnherly, edited byM. Faster, pp. 20-40. Thc, Artifirct 19(3-4). Whiltlsey, Stephanie, Richard Ciolek-Torello, and William I,. Dcaver Wiseman, KeggeN. 1994 Resurrecting the Oolam: The Early Playas Tnciscd, Sierra Blanca Variety:A Formative Pcriodin Arizona. Paper New Pottery 'l'ypein the Jornada Mogollon. presented at the 7thMogollon Conference, In Tmnsactions qf'thP16Ih Kegionnl Las Cruces, New Mexico. Archucologirnl Svnzposiumjbr Southerrstern Ntw Mpxico rrnd Wpstern Texas. Published Wilding, L. P, N. E. Smeck, andL. R. Drees by the Midland Archaeological Society, 1077 Silica in soils: Quartz, Cristobalitc, Midland, 'l'exas. Tridyrnite and Opal. InMinwrrls in Soil Erndronmcwts, pp. 47 1-552. Soil Science Bison, Fish, and Sedentary Occupation: Society, Madison. Startling Data From Rocky Arroyo(I .A 25277), Chaves County,New Mexico. In Wills, W. 11. Vimw of'th.Jornu(lu Mogollon, edited by 1988 hrly Prchistor.ic Agriculturt, in the Colleen M. Heck, pp. 30-32. Eastern New Amwicrrn Southwst. School of American Mexico University Contributionsin Research Press, Santa Fe. Anthropology, vol. 12, Portales.

Wills, W. H., and 'I'homas C. Windes Thp Lnnd In Between: Archaic rind 1989 Evidence ol' Population Aggregation and Formativr Occupations along the UpprXio Dispcrsal During thcBaskcmaker 111 Period /Iondo ofSouthenstern NPW Mexico. Office in Chaco Canyon, New Mexico.Amcrirnn ol'hrchaeological Studies, Archaeology Antiquity 54(2):347-369. Notcs 125, Museum of New Mexico, Santa Fe. Wilmsen, Edwin N. 1970 Lithic Anulysisand C'ulrurd Infircwce, II U.S. 285 Seven RiversProject: Plan for Prrleo-Indinn Cmc. Anthropological Papers Data Kecovery at Four Archaeological Sites ofthe University of Arizona No. 16. along South Seven Rivers, Central Eddy Tucson. County, Ncw Mexico. Officeof Archaeological Studies, Archaeology Notes Wilson, C. Dean, and Eric Elinman 100, Museum of'Ncw Mexico, Santa Fe. 199 1 Early Anasazi Ceramics andthe Hasketmaker 'l'ransition. Papcr prcscntedat Cultural Setting. In Tc>stinyRrsults and the Anasazi Symposium 189 I, Mesa Verdc Drlfrl Rwowty Plmzfbr the Curlshnrl R&f' National Park, Colorado. Rolltr, Edr(v Counl,y, NFW Mexico. Office of Archaeological Studies, Archaeology Notes Wilson, C?. Dean, Eric Hlinman, .lames M. Skibo, and 2 19, Museum of New Mexico, Santa Fe. Michael Brian Schifi'er 1996 Designing SouthwesternPottcry: A introduction to Selected Potlery Typesof Technological and ExperimentalApproach. Central and Southeastern Ncw Mcxico. In htcrprrting. %uthl.twstern Diversity; 1)ocutnent on file at the Ol'lice of Undwlving Principles and Overarrhing Archaeological Studies, Museumof New Puttrrns, ediied by P. R. Fish and .I. .I, Keid, Mexico, Santa Fe. pp. 249-256. Arizona State University

Anlhropological Research Papers No.28, York, J.C.? and A. W. Dick-Peddie Tempe. I969 Vegetation Changesin Southern New Mcxico during the Past IIundred Latnora, Dorothy A. Years, In Arid 1,undy in 1997 Testing Rr?,smltsand 1111tl1 Xc~~cyyPlun For Pcrspectil,c, cditcd by W. G. The Cwlshntl Kc1iL.f'Koutr, Eddy County, McCjinnis and H. I. Cioldman, pp. NPWMuicw. Office of Archacological 157- 166, University of Arizona Studies, Archaeology Notes 2 19, Muscum Press, Tucson. ofNew Mexico, Santa Fe.

166 REFERENCES CITED APPENDIX 1. ANALYSIS CODING FORMS

Ceramic Analysis Attribute and Variable Codes Carlshad Relieflioute Project

Sitc Number possible schist Site number is recorded on the upper right 13 Large gray feldspars corner of analysis form. Ceramic Type P.S. Number Jornadn Mogollon Brown Wares Point Provenience Number 2101 .lornada Plain Brown Body 2 102 Jornada Plain BrownKim Lot Number 2103 Very thin polished unpainted, similar Consecutive numbers arc assigned to lines to sherds described having derived within an F.S. group. This numberlinks data hmEl Paso Polychrome line with ceramics. 2 104 Brown Ware with smudged interior Temper 2202 Red Slipped Brown Identification based on microscopic 2207 Three River R/T examination. 2302 Chupadero B/W(solid designs and side lines) Not examined 2303 C'hupadero BIW (hatchurcd Indeterminate 2304 Chupardero solid and None hatchurcd Medium small to rclalivcly large 2305 Chupadero HIW angular white to light gray and Tndcterrninate design occasional clear angular igncous 23 10 Unpainted probably from quartzite and feldspar fragments Chupadcro B/W (lcucocratic igncous or granite) 3 103 Indeterminate Non-local Textured along with rare small black (possibly plains, possibly Corona lkagmenls within or outside lighter 4 IO2 Corona Corrugated fragments. Very small sugary white temperthat Wiseman Brown Ware Categories may be clcar, dull while, or black, 1 Jornada and larger fragments tendto bc very 2 El Paso crystalline or sugary in appearance. 3 South Pecos h Donlinated by relatively 4 Jornada South Pecos large, angular while 5 McKenxie fragments along with largc round 6 El Paso Polycllrome clear fragments, 7 Medium sub-angularto sub-rounded Pastc Color (Recorded for interior and exterior whitc, gray, and reddish liagments surfaces) of about the same size. May still bc 1 Dark gray to black cross sectionbut variation of leucocratic rock. oxidized on one surface Dominated by medium sized clear 2 Dark gray to black cross section x sand rounded grains along with 3 Brown or reddish throughout angular whitcfi-agmenls. 4 Oxidized red or brown on outside 9 nark angular, sherd or rock dark gray or black on outside (Chupadero Black-on-white temper) 5 Red and gray streaks 10 Light sherd 6 White to light gray 11 Schist 7 White and dark streaks 12 Large angular white and Pigment (Recorded lorinterior and exterior surfaces) higher degrec of polishing or paint 1 None on the interior surfaces. 2 Mineral black 4 Jar body 3 Mineral red 5 CookingiStorage jar rim 5 Organic 6 C.’ooking/Storagejar neck 6 White clay IO Indeterminate 7 Mineral and white clay 11 Seed Jar Rim 8 lndetcrminatc 12 Indctcrminate due to missing surface Manipulation (Recorded l’ur interior and exterior 13 Indeteminate handle surfaces) Modification I Plain unpolished 1 None 2 Plain slightly polished or 2 Indcterminate smoothed 3 Ilrilled hole 3 Plain heavily polished 4 Reshaped vessel basedon 4 Plain unpolishcd with modified rim striations 10 Sooted both sides 5 Surface missing 11 Sooted interior 6 Surface worn 12 Sooted exterior 7 Indoteminate textured 13 Naturally worn and shaped 8 I ndentcd C.’orrugatcd 14 Hoth sides sootcd

Slip (Rccordcd for interior and exterior surfaces) Shrrd Count 1 None 2 Red slip Sherd Weight 3 Stnudged 4 White slip Vcssel Number

Form 1 Indeterminate body polished both sides 2 Bowl rim Commcnts 3 Bowl bodyBS determined bya much

168 APPENDIX 1. ANALYSIS CODING FORMS LITHIC ANALYSIS ATTRIBUTES AND VARIABLE CODES CARLSBAD RELIEF ROUTE PROJECT

Material Typc 632 mica 633 gypsum 000 unknown 634 calcite 00 1 chert, undiff. 635 sclcnitc 002 Pedemal chert 640 galena 003 Tccolote chert 64 1 henlatite 004 Washington Pass chert 650 coal 005 Brushy Basin chert 65 1 jet 0 0 h Baldy IIill chert A60 fossil, undiff 007 Alibatcs chert h6 I crinoid stem 008 Tecovas chert 670 concretion, undifl'. 0 00 clastic chert 67 1 concretion, sandstone 010 San Andres chert 672 concretion, hematite 011 Santa Fe chert 999 other 012 IIueco chert 0x0 chalcedony, undiff. Material Texture and Quality 100 silicified wood, undiff. 101 Zuni wood glassy 2 0 0 obsidian glassy and flawed 20 1 Jemez, generic finc-grained 202 Polvadera Peak fine-grained and flawed 203 Grants Ridge medium-grained 2 04 Red Hill medium-grained and flawed 205 Antelope Wells coarse-grained 206 Mule Creek coarsc-grained and fla~cd 300 igneous, undiff. 3 10 nonvesicular basalt Artifact Morphology Catcgorirs 320 vesicular basalt 330 granite 00 indeterminate 340 rhyolite 0I angular debris 342 andesite 02 core flake 400 sedimentary, undiff. 0 3 biface flake 410 limestone 04 resllarpening llake 420 sandstone 05 notching flakc 430 siltstone 06 bipolar flake 440 mudstone 07 blade 450 shale ox lwnmerstone flake 500 metamorphic, undiff. 09 channel flake s 1 0 quartzite, undiff. 10 potlid 530 quartzitic sandstone 11 strike-a-light flake 540 schist 2 0 tested cobble 54 1 serpentine 21 core, undiff. 542 micaceous schist 22 unidirectional core A00 mineral, undiff. 23 bidirectional core 610 Lurquoisc 24 multidirectional core 020 azurite 25 pyramidal core 62 I malachite 30 cobble tool, undiff. 630 quartz crystal 31 cobble tooi, unidirectional 63 1 massive quartz 32 cobble tool, bidirectional

APPENDIXI. ANALYSIS CODING FORMS 169 3 0 unifacc, undil'l: 20 1 unidentified large projectile point 41 unifacc, early stagc 102 unidentified smallpro,jectile point 42 uniface, middle stage 703 unidcnlilied stemmed pro.jcctilc point 43 unilace, late stage 204 unidentified large sternnled projectilcpoint 50 biface, undiff. 205 unidentified small stemmed projectile point 51 biface, early stage 206 unidentified corner-notched projectile point 53 bike, middle stage 207 unidentified large comer-notched projectile 53 biface, late stagc point 90 reworked tool, undiff. 208 unidentified small corner-notched pro.jcctile 0 I reworked early-stage tmil'ace point 9 2 reworked tniddlc-stage unifacc 200 unidentilied side-notched pro.jcctilc poinl $1 3 reworked late-stage unifacc 2 I 0 unidentified large side-notched pro.jcctilc 94 reworked early-stage biface point 95 rcworked middle-stage biface 21 1 unidentified small side-notched projectile 06 reworked late-stage bit:acc point 100 unworked cobble 2 12 unidentified tluted point 120 unidentified Paleoindian projectilepoint Artifact Functional Catcgurics 22 1 Clovis point 222 Folsom point 00 1 utilized debitage 223 Midland point 002 retouched dobitage 224 Plainview point 003 utilixedretouched debitagc 225 Firstview point 004 llake tool 226 Jimmy Allen point 0 10 hammerstone 227 Meserve point 01 1 chopper 228 Agato Basin point 012 plane 229 Hell Gap point 013 axe 230 Belen point 23 I Milncsand point 0 14 pecking stone 0 1 5 hoe 232 Eden point 016 maul 233 ScottsbluL1'1 point 017 tcharnahia 234 Scottshluff 11 point 050 drill 235 Frederick point 05 1 graver 300 unidentified Archaic pro.jcctile poinl 052 spokeshave (notch) 30 I Say point 053 denticulate 302 Hajada point 075 core-chopper 303 San Jose point 076 scraper-graver 304 Armijo point 077 chopper-hammerstone 305 En Medio(Haukctmakcr IT) point 078 core-hamrnerstonc 306 Chiricahua point 080 strike-a-light fling 307 San Pedro point 08 1 gun flint 308 Augustin point 090 unutilizcd angular debris 309 Pelona point 09 I unutilized flake 4 (:I 0 Basketmaker 111 point 092 unutilized core 40 I Pueblo side-notched point 093 unutilizcd cobblo tool 402 Mogollon side-notched point I00 uniface, undifferentiatcd 500 Athabaskan side-notched point 101 end scraper ti00 eccentric IO2 side scraper 60 1 thrcc sidenoches 103 endlside scraper 900 lire-cracked rock 104 thumbnail scraper I so biface, undifferentiatcd Dorsal Cortex 151 knife 152 Cody hili Estimate in 10 intervals 200 unidentified pro.jcctile point

170 APPENDIX I. ANAI.YSIS COIIINC; FORMS 07 rounding and02 Flake Platform 'I'ypr 08 rounding and03 09 rounding and 04 00 not applicable 10 noncultural edge darnage 01 cortical 11 birdirccticmal retouch and abrasion 02 cortical and abraded 12 battering 03 singlc facet 13 rotary 04 single facet and abraded 14 unidirectional retouch and wear 05 mullil'acet IS unidirectional retouch and battering Oh rnultifacet and abraded 16 bidirectional retouch and battering 07 retouched 17 bidircctional retouch and wear 0 retouched and abradcd IX abraded for platfonn preparation 09x abraded 19 abrasion 10 collapsed 2 0 unidirectional wear and working edgc abradcd I1 crushed 21 serrated dcnticulate 12 absent 22 unidirectional retouch and bidirectional wear 23 bidirectional wear and workingedgc abradcd Portion 24 bidircctional retouch and unidirectional wear 25 unidirectional retouch and abrasion 0 indeterminate fragment 50 hiface edge 1 whole 51 uniface cdgc 2. proximal 52 projectile point blade edgc 3 medial 53 projoctilc point base edge 4 distal 54 drill shaft edge 5 lateral 55 bil'ace edge, abraded for platform prcparation 6 collapsed platfonll 56 hoppcr edgc 57 strike-a-light flint edge 5x gunllint edge length by width by thickness, weight, or both Thermal Alteratiou

Cortex Type 00 not applicable 01 potlids, ventral surface 0 not applicable 02 potlids, dorsal surface 1 waterworn 03 potlirls, other surface 2 nonwaterwom 04 crazed 3 indeterminate 0 s crazed and0 1 0 k crazcd and 02 Platform Lipping 07 crazed and03 OX crazed and 04 0 not applicable 09 potlids, ventral and dorsal surfaces I prcscnt 10 potlids 2 none present 11 color changc

Distal Termination Wear Patterns 0 not applicable 00 not applicable I feather 0 1 unidirectional utilization 2 hinge 02 bidirectional utilization 3 step 03 unidirectional retouch 4 absent, snapfracture 04 bidirectional retouch 5 obscured, cultural alteration OS rounding 6 obscured, lloncultural alteration Oh rounding and 0 1 7 broken in manufacture

APPENDIX I. ANAL YSlS c'O13TNG FOR" I71 x outrepasse concave 9 axial convcx 10 plunging concave-convex straight serrated projections irregular sernircgular Edge Sllapc regular

00 not applicable GROUND STONEANALYSIS ATTRIBUTES AND VARIABLE CODES CARISBAD RELIEFROUTE PROJECT

Material Type Prrform Morphology

000 unknown indeterminate 00 1 chert rounded cobble I00 silicificd wood core 200 obsidian chunky and angular 300 igneous undifferentiated flattened cobble 310 nonvcsicular basalt slab, not further specified 320 vesicular basalt slab, thick (I 0 cm.+) 330 granite slab, thin (5-10ctn) 340 rhyolite slab, very thin (4cm) 400 sedimentary undifferentiated 410 limcstonc Production Input 420 sandstone 430 siltstone 0 indeterminate 440 mudstone 1 none (natural form) 450 shale 2 slightly modified (cy 50pcrcenl area) SO0 rnctamorphic undifferonliated 3 mostly modified (SO-99pcrccnt) 510 quartzite undifferentiated 4 fully shaped 530 quartzitic sandstone 540 schist Shaping 54 1 serpentine ti00 mineral undifferentiated 0 indeterminate 610 turquoise 1 none ti20 azurite 2 grinding 62 1 malachite 3 flaking 630 quartz crystal 4 pecking ti3 1 massivc quartz 5 grinding and flaking 632 mica 6 grinding , llaking, pecking 633 gypsum 7 tlaking and pecking 634 calcite x grinding and pecking 635 selenite 9 other 640 galcna 10 incised 64 1 hematite ti50 coal Weight or Dimensions 65 1 Jet 660 fossil undifferentiated 66 1 crinoid stcrn 670 concretion undifferentiated 0 no ti7 1 concretion, sandstone 1 yes 672 concretion, hematite 9 9 9 other Ground Surface Measurement

Material Texture and Quality Mano Cross Section Form

1 glassy 0 indeterminate 2 tine-grainod I biconvex 3 medium-grained 2 convex-concave 4 large-grained 3 dome, oneflat side 4 subroctangular 5 square or rectangular indeteminate 6 wedge Whole 7 loaf end fragment 8 irregular, one tlat side medial fragment 9 irregular, one concave side edge fragment 10 irregular, one convex side internal fragment II diamond corncr 12 triangle com(s) only missing 13 airfoil 14 trapezoidal

Metnte Depth indetcnninate fragment indeteminate Plan View OutlineForm polishing stone, pottery polishing stone,plaster 0 indeterminate abrading stone 1 circular shaft straightener 2 oval shapcd slab 3 subrectangular jar cover 4 irregular anvil 5 square pounding stone h rectangular palette lapidary slono 7 subtriangular x wedge mortar 9 othor regular form coma1 hamrncrstonc PhotolTllustrntu nlano one-hand mano two-hand tnano 0 110 I ycs two-hand mano, trougll two-hand mano, slab Flaked Surface or Margirl Present two-hand mano, loaf-shaped melale 0 no metate, basin 1 yes metate, trough trough, cnds opun Heat trough, ends closed trough, one end open 0 indeterminate metate, slab I none maul 2 reddened maul, notched 3 fractured ~naul,groovcd 4 crazed mining too1 5 2, 3 axe h 2,4 axe, notched,1 7 3,4 axe, notched, 2 x 2,3,4 axe, notchc, 3 9 other axe, chipped, notched 10 blackcncct axe, grooved axe, %1 grooved Use Number axe, grooved,full axe, grooved,spiral Portion hoe hoe, nolched

174 APPENDIX 1. ANALYSIS CODING FORMS ox2 hoe, grooved 8 grooving OX3 Ichamahia 9 grinding 090 cylindrical tool 10 other 09 1 pestle 092 paint stone Sccondary Wcar 100 ornalnent (Codes follow primarywear) 101 pendant I02 bead Alterations 103 pipe 111 cloudblower 0 indeterrninatc 120 concretions, crystals, and lnincrals 1 none 2 dri Iled Ground Surface Cross Section 3 incised 4 grooved 0 indetenninate 5 notched 1 flat I, 2,3 2 concave 7 2,3,4 3 CollveX 8 2,3,4, s 4 grooved 9 2,4 5 irregular 10 2,4, 5 6 faceted 11 2, 5 7 other 12 3,4 13 3,4, 5 Ground Snrf;lce Sharpening 14 3, 5 1s 4, 5 0 no I yes Adhesions

Grorml Surface Texture 0 indetenninate 1 now 0 not applicable 2 pigment stains 1 indeterminate 3 inorganic, other 2 coarse 4 gum or pitch 3 rnodcrate 5 organic, other 4 fine 5 polished Striation

Primary Wear 0 none 1 length 0 indelerminate 2 width 1 nunc 3 circular 2 striations 4 random 3 pitling S I, 2 4 edge damage 6 indeterminate 5 bevelling 7 transverse (axes) 6 battering 7 polishing FAUNAL ANALYSIS ATTKISU'L'ES AND VARJARLE CODES CARLSBAD RELIEF ROUTE PROJECT

Certainty (record onlyif uncertain) 1 lairly certain 2 less certain

Taxon Unknowns 1 unknown small 2 unknown 3 small mammalimed - lrg bird 5 unknown mammal 10 small mammal 1s small to medium mammal I8 medium m

Mammalia Mammals 35 very large mamrnal (largerthan deer)

40 Marsupialia (marsupials) 41 r1iddphi.s ~tirgirtitrrla(Virginia opossum) 50 Insectivora (insectivores) 100 51 Soricidae (shrew family) 1 0 I 52 Sorer cinercws (masked shrew) 102 53 Sorrx \qyans (vagrant shrew) 103 54 Sowx n(rn14,s(dwarf shrew) 104 55 Sowx rntwimni (Merriam's shrew) 10s 56 Sorm pdustris (water shrew) 57 Notiosocx uwwfiwrti (desert shrew) 106

60 Chiroptcria (bats) I07 61 Phyllostomatidae 1 ox 62 Choeron,ycteris mrxicnnu (long tonguedbat) 63 Lq~ton.vc.trrissnnhorwi (Sanborn's long- 110 nosed bat) 64 Vespertilionidae I11 65 Myoh~vI[/&r (cave bat) 112 66 Myofis yumtmwsis(Yuma myotis) 67 A4yoti.s lur!jfirgu.s(little brown 113 niyotis) 68 M,votis nuricu1lr.s (southwestern 1 I4 myotis) 69 Mymtis wotis (long-eared myotis) 1 IS 70 Myotis th,vsmode,s(fringed rnyotis) 116 71 is 1.olnns (long-leggcd rnyotis) 72 Myotis c.alifiwnic.us(Calif. myotis) I17 73 Myotis leihii (small-footed myotis) I18 74 I,asior~,yctrrisnorfhngans (silver haired bat) 7s Pipisltx~llushespcrus (western pipistrelle) 119 76 E~~te.sic.u,s.filc~l,s(big brown bat)

176 APPENDIX 1. ANALYSIS CO1)INCi FOKMS 120 grasshopper mouse) 121 159 On.yc.honl,ys torridus (southern grasshopper 122 mouse) I23 160 S’igmodon hispidus (hispid colton rat) 124 161 Sigwzodot7,fidh’iventrr(tawny-bcllicd colton 125 rat) I26 162 Sigrnodon ochmgnarhu.s (yellow-nosed cotton rat) 127 163 Npotoma .sp. woodrats I28 I64 Neotomrr micropus (southern plains woodrat) 129 I65 Ntwtoma nlhigulr (white-throated woodrat) 130 I hh Ncotoma stqhensi (Steven’s woodrat) 131 167 Nwtomn mexicann (Mexican woodrat) 1 A8 Ntwtorna cinrrw (bushy-tailed woodrat) 132 I69 C’lethrionomys gnpperi (southern red- 133 backed vole) 134 I70 Phcnncornys intermrrlius(heather volej 171 Micro1i.s ,spy. 135 I72 Microtis prrln.sylvunic.us (meadow vole) 136 173 hlicrotis montnnus (montane vole) I75 Micwti,s rnt,xic.:nnu.r(Mexican vole) 137 176 Microtis /ongicnudu.s(Long-tailed vole) I77 Mi(-rotis ochroguster- (prairie vole) 138 178 Ondufru zibpthicus (muskrat) 179 Muridrrrc (old world rats and mice) 139 180 Rrlttus nrflus (black rat) 181 Xnttus nonvgicns (Nolway rat) 140 I x2 Mus wruscul.7 (housc mouse) 141 I X3 Znyu,sprinccps (western jumping mouse) I 84 Ewfhizon dorsrrtum (porcupine) 142 I85 Mvocastor cqvp1.s (nutria -not nativeto North America) 143 1 X7 small rodent 144 1xx medium to large rodent 14.5 I46 190 Lagomorphs (lagomorphs) 191 Ochotona pritwcys (pika) 147 200 Leporidae (haresand rabbits) 20 1 ,~yhdugu.ssp. (cottontails) 148 202 1q1u.ssp. (jack rabbits) 149 203 SyMIugus,floridunu.s (eastcrn cottontail) 150 204 ,Sylvilr~gusnuttdlii (Nuttall’s cottontail) 205 Syl~il~~guxauduhonii (desert coltontail) 151 206 Lqus rlmericanus (snowshoe hare) 207 Lqus townscndii (white-tailed jack rabbit) 152 20% I,epu,s cnlifbrrricus (black-tailed jack rabbit) 200 /quscal1oti.s (white-sided jack rabbit) 153 154 3 00 Carnivora (carnivores) 155 301 small carnivore (weasel to skunk) 156 301 medium carnivore (badger to canid) 157 303 large carnivore (wolfto bear) 15x 310 Canidae 11 1 Ctrnis sp. 470 Ulis airps (domestic sheep) 312 Cmis lutr.rrns (coyotc) 47 1 O~is/Cqwrr(domcstic sheepor goal) 313 ~~~r~li,s.~~r~lili(rri.s(dog) 472 C:upr.n hirclis (domcstic goat) 314 C:'rrnl.s lupus (gray wolf) 475 Sus .scrojii (domestic pig) 315 Vu/pe.s wdpw (red hx) 480 EYUUS SIJ. 3 1 h Vulps 1~1ox(swift fox) 48 I Eyulrs t~inrrs(burro) 317 V/c& mncrotis (kit hx) 482 Eyuus cvrhrrllus (horse) 318 Urwyon e,iner.rorrl,5.cntc.ll,s (gray fox) 319 small canid (fox) Avrs (birds) 320 Ursidae (bears) 487 small bird 32 1 Urdus (mt>r.ic:mus(black bear) 488 medium bird 322 U1w.s nrc'tos (grizzly bear) 489 large bird 328 Procyonidae (racoon. ringtail. coati) 490 medium to large bird 329 R~mrarise~usastutus (ringtail) 491 very largc bird 330 Procyon lotor (racoon) 500 Aves 33 1 Nmun n(rsm (coati) 333 A4urks Lrrnericana (marton) Caviiformus (loons) 334 A41rstrlrr ~rrnincw(ennine) 50 I Gaviidae (loons) 335 Mustr1cr.frmtrla (long-tailed weasel) 502 Gwin immw (common loon) 336 Mustrirr nigriprs (black-footed ferret) 337 M~otcltrvison (mink) Podicipediformes (grebes) 339 Gulo gulo (wolvcrinc) 503 Podicipedidae (grebes) 340 Tru-irka taxus (badger) 504 Podicqws SJ?. (common grebes) 350 skunk sp. 35 1 Spilogrrlc grrwi1i.r (western) orputoriL1.s Pelicaniformes (pelicansand allies) (eastern spotted skunk) SO9 Pr1ic.rrnu.v nr))thl.or.~l.v!vnol.ho.s(white pelican) 352 Mephitis rnc~Jd1iti.s(striped skunk) 5 10 Phnlncl.ot.ronr\. nuritrrs (double-crested 353 Mqhitis rnacwIurrr (hooded skunk) cormorant) 3 54 I"oncp~rtu,smwdeucus (hog-nosed skunk) 3 00 1,ulr.o cmatlonsis (river otter) Ciconiiforrnes (herons, storks, and allies) 3 70 Felidae 5 1 1 Ardeidae (herons and bitterns) 371 Fcdis cancdor. (mountain lion) 5 1 2 Ardw hwodirrs (great blue heron) 372 Fc>lisr~&s (bobcat) 5 13 Butorictrs ~i~~cwn.s(grccn heron) 373 /.'ells onctr (jaguar) 5 14 Crww-orlius cr1bu.s (common egret) 3 74 Fdis domcsticus (domestic cat) 5 15 Lrucopho.vx thh(snowy cgrct) 400 Artiodactyla (artiodactyla) 5 1 h Nyctirorrm nyc:ticornx (black-crowned night 40 1 small artiodactyl (domcstic sheepto white- heron) tailed dccr size) 5 17 Ixohrychus milis(lcast bittern) 407 small-medium artiodactyl (deer, pronghorn, 5 18 Botaurus 1entiXinosu.s (American bittern) mountain sheep size) 5 19 Ciconiidae (storks and wood ibises) 403 large artiodactyl(clk or bison size) 404 Dicot?/lr?str(jrrrr4 (collared peccary) Anseriforrnes (swans, geese and ducks) 405 snlall to medium artiodactyl 522 Illor .sp. (swans) 408 Cervidae 524 Bruntu cnnrrrlrmis (Canada goose) 400 C'mweiuphus (elk) 52s Anscr dbjfions (white-fronted goose) 410 Orlocoileus SJI. 526 Chew hyprrhrrtr (snow) or caerulcsccns 41 1 Odocoillws hemionus (mule deer) (blue goose) 412 Cldocoilrw I*irg-iniunu.s(white-tailed deer) 527 ducks 420 Anlilocqm anwricrrnn (pronghorn) 528 Ilr?ndt.ocygntr bicolor. (fulvous tree duck) 438 Bovidac 529 Anatinae (surlhce-feeding ducks) 439 130,s tarus (cattle) 530 Anus ,sp. 440 C1vi.s canu~ltwsis(mountain shccp) 53 1 Anm plntyrh,ynchos (mallard) 450 Bison bi,son (bison) 532 Anus dirrzi (Mexican duck) 45 1 HodlJison 533 Anns strc'pw (gadwall)

178 APPENDIX I. ANALYSIS wrmC;FORMS 534 Anrrs ucuta (pintail) 5x4 Typ~muchu.spullidic~int:.t~u(lesser prairie 535 Anrls crrm1inmsi.s (green-winged teal) chicken) 536 Ann.s discors (blue-winged teal) 585 Pcdiocwtcs phasianrllus (sharp-tailed 537 Anus cyanopiem (cinnamon teal) grouse) 538 Mur-ccn rrmcv+:ann (American widgeon) 586 Cr~ntrnc~rrcus/l~,(~phrrsirrnus (sage grouse) 539 Spatulu alypeutn (shoveler) 587 t'hasianidae (quail, partridges, and 540 Ahspnnsn (wood duck) pheasants) 54 1 Aytl1-vinm. (diving ducks) 588 Chlinus virginianus(bobwhite) 542 A,vth,y~rlmwicnnrr (redhead) 5x9 Cnlllip~plnsquwnrrtrr (scaled quail) 543 Ayth-ya callrrris (ring-neckcd duck) 590 Luphort,vx gumhdii (Ciambel's quail) 544 A-vthya lwlisiner-irr (canvasback) 59 1 Cvron.y.x- muntc!zumLw (harlequin quail) 545 Ayhyll marilu (greater scaup) 502 Plza,sinnu,s colchicus (ring-necked pheasant) 546 Aytlyn uffinis (lesser scaup) 593 A Irctor.i.s gruecu (chukar partridge) 547 Bucq7hulrr crrlnngdn (goldeneye) 594 Meleagris gallopovo (turkey) 548 Illlcephulrr nlheol~i(bufflehead) 540 Mdunitta Ilt&mdi (white-winged scooter) Gruiformes (cranes,rails and allirs) 550 Oxvur~ljnrnaicc,nsis(ruddy duck) 595 Gr1l.s umwicmu (whooping crane) 55 1 Lophodytrs c:ucullntu.s (hooded duck) 596 Gr-us cunrrdensis (sandhill crane) 552 MW~USrnerXatzser (common merganser) 507 Rallidae (rails, gallinulw, ccwts) 553 Mcrgus srwutor (rcd-brcastcd rncrganscr) CMliformes 59X Gallifoms b'alconiformcs (vulturcs,hawks, falcons) 554 Cnthrrr-tcs rrlll-rr (turkey vulture) Charadriiformes (shorebirds,gulls, and allics) 555 Accipitridac (kitcs, hawks, and caglcs) 602 Charadriidae (plovers) 556 Arcipito. .vp. (short-winged hawks) 609 Scolopacidac (sandpipers, snipes, curlews, 557 Accipiter gcniiliAs(goshawk) godwits etc.) 558 Accipiter .str-irrtlls(sharp-shinncd hawk) 63 1 Recurvirostridae (avocets and stilts) 55') Accipiier cooptv+ii(Cooper's hawk) 634 Phalaropodidac (phalaropes) 560 Bulw sp. (broad-winged hawks) 637 I aridae (gullsand terns) 561 Butco.jrrmnieensis (rod-tailcd hawk) 562 Jluteo plu,ypterus (broad-winged hawk) Columhiformes (pigeons, doves and allies) 563 Butw swninsoni (Swainson's hawk) 646 Columbidae (pigeons and doves) 564 Buko ulhonotutus (zone-tailed hawk) 647 C'nlumhia,fi-l.sc.irrtn(hand-tailed pigeon) 565 Butw krgopus (rough-lcggcd hawk) 648 Col~rmhliw (rock dove, dotnestic pigeon) 566 lluteo rqylis (ferruginous hawk) 649 Zmrrirlu rrsinticn (white-winged dove) 567 Rutru nitidr~s(gray hawk) 650 Zcwaidur-n mncoul-n (mourning dove) 568 Parrrhutro ~micinctu.~(Harris hawk) 65 I Ectopistrs migrutorius (passenger pigeon) 569 Il~ll(~g~l1~~Lrnthrucinus (black hawk) 652 I:'ol~rmhignllina pnssel-inu(ground dove) 570 hyuila chrysaelos (golden eagle) 653 Sr:nrdufdlo incn (Inca dove) 57 1 Hn1iuctu.s Iclrrocr~~lza1u.s(bald eagle) 572 Circus cyuneus (marsh hawk) Psittaciforimes (parrots, macaws and allies) 573 Pundion halitrtcus (osprey) 654 Psittacidac (parrots) 5 74 Falconidae (caracara and falcons) 655 Rh,vn(-wpsittopnclzyr-nchn (thick-hilled 575 Crrrucmn chrr-iwny (caracara) parrot) 576 I.'dc.o rnc~xicwzus(prairie falcon) 056 Arrr sp. (macaws) 577 I.'ldco pwegrintus(peregrine falcon) 657 Arrr militrrris (military macaw) 578 Frllco,f&momlis(aplomado falcon) 658 Arrr mncun (scarlet macaw) 579 Frrlco co1umhnriu.s (pigcon hawk) 580 Fulcu Aspr~nvl-iLsu(sparrow hawk) Galliformes (grouse, pheasants, turkeys and allies) Crrcoliformes (cuckoos, roadrunners and allics) 58 1 Tecraonidae (grouse and ptarmigan) 650 C'uculidae (cuckoos, roadrunncrs and ani) 582 Uc~ndmgrrp~sohscur-us (hluc grousc) 660 Coc~cyzusamr~ricmus (yellow-billed 583 I,ngopw Ic~uc~~~r~~s(white-tailed ptarmigan) cuckoo) 66 1 Gt~oc~oc~y.x~IifOrnianus (roadrunner) 748 Paridae (chickadees and allies) 662 Crotophqu .sulcirostris (groovo-billcd ani) 749 Sittidae (nuhatches) 752 Troglodytidae (wrens) Strigiformes (owls) 755 Mirnidac (mocking birds andthrashers) 663 Stigidae and Tytonidae (owls) 763 Turdidae (thrushes,solitares and bluebirds) 664 Tvto nlhrr (barn owl) 764 Turrlis mzgm~r~rius(robin) 645 Otr4.s rrsio (screech owl) 766 Sylviidae (gnarcatchers and kinglets) 666 Otus flnmnwolus(tlammulated owl) 7A7 Motacillidac (pipits) 067 Kuho virgir.rionL1.s (great horned owl) 76% Homhycillidac (waxwings) 668 C;lmciu’iurn gnomrr (pygmy owl) 769 Stunlidae (starlings) 669 Mir:tnthtw, whitntyi (ell’owl) 770 Vireonidae (vireos) 670 ,S+o@to c:.rrniculrrrin(burrowing owl) 77 1 Parulidac (wood warblers) 67 1 Strix- occidmtrris (spotted owl) 772 Ploceidae Passer domesticus (housc 672 Axio otis (Inng-eared owl) sparrow) 673 Asio gl~rmmrra(short-eared owl) 773 Icteridae (blackbirds, orioles, and 674 At&ius acudicrls (saw-whet owl) meadowlarks) 774 ‘l’hraupidae (tangers) Caprimulgiformes (whip-poor-willand allies) 775 Fringillidae (grosbeaks, finches, towees and 675 Caprimulgidae (goat-suckcrs) sparrows) 676 Crq~rimu1gu.r\wzfiw.s (whip-poor-will) 116 Passcrifurmcs 677 Phalamoptilr~s nuttnllii(poor-will) 678 Chor.rJc.il~sminor (common nighthawk) Domestic Bird 679 Chordrilc~s acxtipennis(Icssser nighthawk) 777 C;nllus gallr~s(domestic chicken) 778 chicken-like Apudiformcs (swifts and hummingbirds) 680 Apoidae (swirts) Reptiles 683 Trochilidac (hummingbirds) 780 Hcptilcs 684 Trogonidae (wagons) 7X 1 Testudinata (turtles andtortoises) 782 Chclydridac (snapping, muck,. andrnud Coraciiformes (kingfishers, hornhills andallies) turtles) 685 Alcedinidae (kingfishers) 783 C:hclydru scrptwlinrr (snapping turtle) 784 Kinosternon sp. (musk andrnud turtles) Piciformes (woodpeckers, toucanss, andallies) 7x5 k’ini,sternon~flnvc.sccn.s(yelllow mud turtle) 688 Picidac (woodpcckcrs) 786 Kinostcrnon sonorirnsc. (Sonora mud turtle) 689 Colaptes auratus (yellow or red-shafted 7x7 Emydinae (box andwater turtles) flicker) 788 Clemmys mrrrmorrrtrr (western pond turtle) 789 C,’h~v.rnv,v.ssp. (painted turtles) Passcrifurmes (small perching birds) 79 1 Chrywm.vs pic~lttr(painted turtle) 703 792 Pseur/om.v.s sp. 725 794 P,swdornys .scripta (pond slider) 726 795 Pstwdon~.vsconcinnrr (cootcr) 727 796 Torrrrpcvtc ,sp. (box turtles) 736 798 Twr.rqJmr mrnrrtu (western box turtle) 737 799 ‘I’rionychidac (softshcll turtlcs) 738 x00 Tri0n.w ,sp. (softshell turtles) 739 x0 1 Trion.yx spinzfhs (spiny sollshell) 740 802 liion-yx rmticlrs (smooth softshcll) 74 1 810 Squarnata (lizardsand snakes) 742 x1 1 Sauria (lizards) 743 820 Ophidia (snakes) 744 825 Colubridae (nonvenomous snakes) 745 x30 Iktrrodorr nnsicus (western hognosc snakc) 746 835 C,Wubcr .sp. (racers) 747 840 Lnwpvprltis .st) (kingsnake)

180 APPENDIX 1. ANALYSIS CODING FORMS 91 horn corc 92 horn sheath 100 cranium 160 mandible 170 indeterminate tooth (maxillary or mandible?) 1x0 hyoid Amphibians 200 vertebra 860 Amphibian 20 1 atlas X6 1 Urodela (salamanders) 202 axis 862 Ambystomrr tigrinnm (tiger salamander) 203 cervical vertebra 870 Salicnta (frogs and toads) 204 c3 X71 Scaphiopus (spade-foottoads) 205 c4 880 Buhnidae (true loads) 206 c5 890 Ranidae (true frogs) 207 e6 208 c7 Fish 210 thoracic vertebra 900 Ostcichthyes (lishj 21 1-224 Tl-Tl2 225 lumbar verlebra 00 1 Salmonidac (salmon, trout,etc.) 910 Catostonlidae (suckers) 226-232 Ll-L7 91 1 Ijiol7u.s huhalusr (small mouth buffalofish) 233 sacral vertebra 920 Cyprinidae (minnows) 235 caudal vcrtcbra 930 Iclaluridae (catfish) 236 sacrum 040 Centrarchidac (sunfish) 240 sternum 950 Percidae (perch) 24 1 manubrium 960 Scrranidac (bass) 243 rib 244 ossified cartilage Mussel and Shell 245 clavicle 997 snail 746 scapula 098 rnussel or shell 247 innominate 24 x pelvis Articulation 300 humerus 1 yes 30 1 radius 2 probably 301 ulna 3 same individual 303 radio-ulna 4 picccs of thcsame bone 3 04 carpal 305 scaphoid 306 lunar 307 scapholunar 308 pisiform lndeterminate fragments 309 cllneifom7 1 indetertninate element 310 unciform 2 indeterminate fragment 311 trapezoid-magnum 3 long bone fragment 315 metacarpal 4 llat bone I'ragment 316 metacarpal 1 5 cancellous tissue 317 metacarpal 2 318 metacarpal 3 Skeletons 319 mctacarpal4 50 complete skeleton 320 metacarpal 5 SI most of a skclcton 322 manus -first phalanx 52 partial skeleton 323 manus -second phalanx 324 manus -thirdphalanx Mamrnals and general 330 femur 90 antler 33 1 patella 332 tibia 71 1 hsed lumbar 333 fibula 717 fuscd thoracic 334 tibioiibula 720 carapace -33 5 tarsal 730 plastron 336 astragalus (tibial tarsus) 73 1 epiplastron 337 calcaneum (fibular tarsus) 732 cntoplaston 338 cuneifonn 733 hyoplastron 3.79 navicular 734 xiphiplastron 340 cuboid 790 Bivalur 34 1 nitviculocuboid 800 Gsh axial 342 lateral malleous X63 Pectoral tin 34s metatarsal 866 Abdominal vcrtcbra 346 metatarsal 1 X67 Caudal vcrtcbra 347 metatarsal 2 878 dorsal rib 348 metatarsal 3 x79 Ventral rib 349 metatarsal 4 883 Ilorsal fin 350 metatarsal 5 884 Anal fin 152 pes -first phalanx 885 Caudal fin 353 pes - second phalanx 886 Ultimate vertebra 354 pcs -third phalanx 887 Penultimate vertebra 355 sesamoid X88 Epural 356 baculum 357 carpal or tarsal Elcmcnt Sidc 359 vestigial phalangeiass. hoof core 0 Not applicable 3 60 nxtapodial 1 indetcrnhatc 36 1 vestigial metapodial 2 axial 362 pllalam 3 right 363 first phalanx 4 lel't 304 second phalanx 365 third phalanx Chmplotoncss 0 not applicable Specializcd Klernents: 1 indctcrrninatc 600 coracoid 2 complete 60 1 lurculum 3 more than 75 percent complete 602 carpornetacarpus 4 50-75 pcrccntcomplutu 003 pollex 5 25-50 percent complete 606 digit 11. phalanx I 6 less than 25 percent complete ti07 Digit ill 608 Digit 11. phalanx II Fragmentation 609 Specializcd wingphalanx 0 complete (analytically) 610 sysacrum 1 li-agment 61 I tibiotarsus 2 shaft fragment 612 tarsometatarsus 3 end fragment 613 muscle splint 4 lateral 615 phalanx ( 1) 5 medial 6 anterior 616 phalanx (2 or 3) 6 I 9 ungual phalanx. claw 7 postcrior 620 urostyle 8 superior 62 I sygostylc 9 inferior 650 fused calcaneundastragalus 10 proccss 660 suprascapula 13 cancelous 700 basipterygiurn 70 1 modified vertebra Cranium 702 trunk vertebra I6 antler base or junction

I82 APPENDIX 1. ANrlLYSIS CODING FORMS 20 enamal 124 arch and body fragment 21 root 125 body fragment 50 case fragment 126 arch fragment 51 vault 127 spinous process 52 anterior vault 128 transverse process 53 posterior vault 129 articular facet 54 occipital region 130 epiphysis 55 basioccipital 56 €ace/frontal Tnnominatr 57 parietal I40 iliutn 60 sphenoid 141 ilium fragment 70 temporal region I42 ischium 71 petrosa 143 ischium fragment 72 auditory bulla 144 acetabulum 73 aditory bulla fragment 145 acetabulum fragment 75 zygomat 146 pubis X0 liontal region I47 ilium and acetabulum XI orbital rcgion 148 ilium, acetabulum and ischium x2 rnaxilla 149 acetabulum and ischiurn 83 maxillary fragmcnt I so acetabulurn and pubis 84 maxillary incisor IS1 ischium and pubis 8.5 maxillary canine 151 acetabulum, ischium and pubis X6 maxillary premolar 153 pubic symphysis 87 maxillary molar IS4 1/2 ilium and % ischium 8X maxillary molar/ premolar fragment 1 s5 ‘/z ilium, ‘/z ischium and 54 pubis $9 anterior lnaxilla (hrousc pad.) IS6 ilium fragment and acctabulum I‘ragment 90 palateipalatine 157 ischium and acetabuliurn fragmcnt I.) 5 nasal IS8 pubis fraglncnt Q h tnax and case frag, I60 glenoid I61 bladc fragment Mandiblc 162 neck 100 ascending ramus 165 glenoid fragmcrlt 1 0 1 coronoid process 102 rnandibular condyle Long Bonrs 103 horizontal ramus 220 proximal 104 mandibular dentition 22 1 proximal and 113 shaft 1os rnmdibular incisor 222 proximal and 2/3 shaft 106 mandibular canine 223 proximal and shaft 107 mandibular premolar 224 proximal shaft fragment 108 mandibular molar 225 proximal fragment I09 mandibular molar premolar fragment 226 proximal fragment -lateral or anterior 110 mandibular symphysis 227 proximal fragment -medialor posterior 111 body fragment 228 proximal epiphysis 112 ramus and tnost of body 229 proximal epiphysis fragment 113 symphysis fragment 230 distal 114 ramus fragmcnt 23 1 distal and 1/3 shaft 115 body fragment tooth sockets present 232 distal and 213 shaft 116 body and ramus fragment 233 distal and shaft 234 distal shaft fragment Vertebrae 235 distal fragment 120 body 236 distal fragment - lateral 121 arch 23’7 distal fragment -medial 122 process 238 distal epiphysis 123 body and arch fragment 230 shaft (2/3+)

APPI:’NIlIX 1. ANALYSIS CODING FORMS 183 240 shal't split lengthwise X08 Supraoccipital 24 1 shaft split -proximal and mcdial X09 Plerolic 242 shaft split -proximal and lateral x 1 0 Opisthotic 243 shaft split -proximal and anterior XI I Exoccipital 244 shaft split -proximal and postcrior 811 Basioccipital 245 shaft split-distal and medial 813 Parasphcnoid 246 shaft split -distal and lateral X14 Basisphenoid 247 shaft splil -distal and anterior 815 Prootic 248 shaft split -distal and posterior X16 Pterosphenoid 249 shaft split -medial portion 818 Symplectic 250 shaft split -lateral portion 8 I9 Quadrate 25 1 shaft split -anterior portion 820 Eclopterygoid 252 shaft split -posterior portion x2 1 Palatinc 270 epiphysis fragment 822 Entoptergoid X23 Melaplerygoid Specializrd Klcmcnts 824 Prcopcrculum Cranium x25 Operculum lbird] 826 Suboperculum 300 jugal 827 interoperculum - 30 1 quadrate 828 Articular 302 quadratojugal X29 Angular 305 supraoccipital 830 Dcntary 306 beak 83 1 Maxilla 307 dentary 832 Premaxilla 308 sclerotic ring X4 1 Basibrachial 842 Hypohrachial Carapace 843 Ceratobrachial 600 nuchal sculc X44 Epibrachial 60 1 nuchal bone 845 Suspensory pharyngobrachial 602 neural X46 Gill raker 603 marginal X47 Sagitta (otolith) 604 pygal 848 SO1 (Lachrymal) 605 suprapygal 849 so2 (Jugal) 606 costal 850 SO3 (Suborbital stay) 607 pleural 85 1 SO4 609 unidentified carapace 852 so5 6 IO platron x53 SO6 X54 Nasal Piastrot1 X55 Tabular 6 1 I epiplaslrodhumeral X56 Poslternporal x57 Supraclcithrum 6 12 cntoplastron 6 13 hypoplastron 858 Postcleithrum 6 IS xiphiplastron X59 Cleitlmrm 860 Hypcrcoracoid Fish 86 I Hypocoracoid

Cranium Mandiblc 80 1 Vomer 8 17 I-Iyornandihular 801 Ethmoid X33 Interhyal X03 Pre-frontal X34 Epihyal X04 Frontal 835 Ceratohyal X05 Sphenotic 836 Basihyal X06 Parictal 837 Hypohyal X07 Epiotic X38 Glossohyal

184 APPENDIX 1. ANAI.YSIS CODlNG FORMS 839 Urohyal I light x40 Hrachiostegal 2 medium 3 heavy Axial 862 Actinost Animal Alteration 864 Basipterygium 0 not applicable 865 Pclvic fin 1 indetenninate 868 Centnun 2 absent 869 Neuropophyscs 3 carnivore gnawing x70 Neural spine 4 carnivore tooth punchlre X7 1 Neural arch 5 3 and 4 X 72 Haemopophyses h scatological x73 Haemal spine 7 scat? x74 Hacmal arch (and haemal canal) 8 rodent gnawing x75 Prezygapophyses 9 possibly agenl‘? 876 Postzygapophyscs 877 Parapophyscs Animal alteration location 880 Pterygiophore 0 not applicable XX 1 1,epidotrichium 1 indctcrminatc 882 hctionotrichium 2 proximal X89 Hypural plate5 3 dista1iinfcric)r 890 Urostylc 4 endsledge 5 shaft 900 egg shell 6 projections 7 boay 10 cntirc specimen 11 arch 0 not applicable 1 indctorminate young 12 posterior 2 fetallneonate 3 imnalure Burning type 4 juvcnilc (2/3+ size) 0 not applicable 1 indctcrtninate 5 mature 1 11011e Age Criteria 3 light scorchcd (tanibrown) 0 not applicable 4 heavy charred (black,gray) 1 epiphysial fusion 5 calcine (while) 2 porous A browniblack 3 cornpact 7 blacWcalcine 4 size 5 deciduous dcntitiodtooth eruption Burning location 6 permanent dentitioldtooth eruption 0 not applicable 1 indeterminate Environmental Alteration 2 even, entire specimen 0 no( applicable 3 discontinuous, ontirc specimen 1 indeterminate 4 proximal 2 none 5 distal 3 pittingicorrosion ti end( s) 4 sun bleachcd 7 shaft 5 checked/exfoliated x projectioninmgin 6 root etched 9 intcrior or cxtcrior only 7 polishedrounded Boiling/cooking brown Environmental alteration drgrcr 1 present (color and round) 0 not applicable 2 possible (color no rounding) 3 rounding 7 midshaft R proccss S waxy C) surface (eg. cranial fragment) Processing 10 proximal shaft 0 not applicable 11 distal shaft 1 indeterminate 12 body 2 none 15 neck 10 cuts, unspecikd 20 anterior 11 transverse cuts 21 superior 12 longitudinal cuts 22 posterior 13 oblique cuts 23 interior 14 random cuts 24 inferior 19 shallow scooped out area 25 arch 20 proccss rcmovcd 30 diagonal shaft CUI 011’ 30 grooved 31 horn 31 groovedlcut off 32 base 40 impact fracture 4 0 acetabulum 41 spiral fracturc 9 9 indeterminate 42 snap break 5 0 sawn Modification 51 portion cut OK (:) not applicablu 52 chop 1 indeterminate 53 chop and cuts 7 manufacturing debris -split 55 abrasion 3 manufacturing dcbris-cuts or grooves 58 cut and snap 4 nlanufacturing debris-polishistriations 59 diagonal break 5 drilled 60 longitudinal break h tool -piercing hl transversc break 7 tool -bead or tubc 65 13 and 40 8 tool -ornament 66 split 9 tool -use deiined 70 scrapes- long parallel scratches 10 tool -scraper/spatulatc 75 hone flake 11 tool -other X0 peel 12 tool fragmenl 90 end hacked off IS non-tool 25 edge wear1 flaked Processing Location 40 pigment 0 not applicable 09 possiblc mx~ification I proximal 2 distal Cornmerits 3 medial 1 yes 4 lateral 2 biology 5 ends 3 flotation 6 margins

186 APPENDIX 1. ANALYSIS CODING FORMS APPENDlX 2. LEGAL DESCRTPTION

Lcral Description:

Elevation: 990.60 m (3,250 ft). Elevation: 989.07 m (3,245 ft).

UTM Coordinates: . UI'M Coordinates:

NMSH'TD Station NMSIITD Station: