A Chronology of Middle Missouri Plains Village Sites By Craig M. Johnson with contributions by Stanley A. Ahler, Herbert Haas, and Georges Bonani Smithsonian Institution Scholarly Press Smithsonian Institution Scholarly Press s m i t h s o n i a n c o n t r i b u t i o n s t o a n t h r o p o l o g y ? n u m b e r 4 9 Artifacts from the Craig Moun at Spir , Oklahoma April K. Sievert with J. Daniel Rogers Contribution by Javier Urcid SerieS PublicationS of the SmithSonian inStitution Emphasis upon publication as a means of ?diffusing knowledge? was expressed by the first Secretary of the Smithsonian. In his formal plan for the Institution, Joseph Henry outlined a program that included the following statement: ?It is proposed to publish a series of reports, giving an account of the new discoveries in science, and of the changes made from year to year in all branches of knowledge.? This theme of basic research has been adhered to through the years by thousands of titles issued in series publications under the Smithsonian imprint, com- mencing with Smithsonian Contributions to Knowledge in 1848 and continuing with the following active series: Smithsonian Contributions to Anthropology Smithsonian Contributions to Botany Smithsonian Contributions to History and Technology Smithsonian Contributions to the Marine Sciences Smithsonian Contributions to Museum Conservation Smithsonian Contributions to Paleobiology Smithsonian Contributions to Zoology In these series, the Institution publishes small papers and full-scale monographs that report on the research and collections of its various museums and bureaus. The Smithsonian Contributions Series are distributed via mailing lists to libraries, universities, and similar institu- tions throughout the world. Manuscripts submitted for series publication are received by the Smithsonian Institution Scholarly Press from authors with direct affilia- tion with the various Smithsonian museums or bureaus and are subject to peer review and review for compliance with manuscript preparation guidelines. General requirements for manuscript preparation are on the inside back cover of printed volumes. For detailed submissions require- ments and to review our manuscript preparation and style guides for authors, visit the Submissions page at www.scholarlypress.si.edu. S M I T H S O N I A N C O N T R I B U T I O N S T O A N T H R O P O L O G Y ? N U M B E R 49 washington, D.c. 2011 artifacts from the craig Mound at spiro, oklahoma April K. Sievert with J. Daniel Rogers Contribution by Javier Urcid aBstRact sievert, april K., with J. Daniel Rogers and contribution by Javier Urcid. Artifacts from the Craig Mound at Spiro, Okla- homa. smithsonian contributions to anthropology, number 49, xiv + 231 pages, 163 figures, 11 plates, 48 tables, 2011.?this monograph presents the historical context and detailed descriptions of a remarkable collection of more than 20,000 artifacts from the craig Mound at the spiro site in eastern oklahoma. spiro is one of the key sites known for the Mississippian Period (ad 900?1500) of the eastern United states. aside from the cultural importance of the site in regional history, the artifacts from spiro provide an almost unique glimpse into the ceremonial life and artistic innovations of a people who developed an important but poorly known cultural tradition. Between 1933 and 1936 the spiro site was looted, and artifacts were sold and traded to many collectors. subsequently, professional archaeological excavations were conducted, and those collections primarily reside at the sam noble oklahoma Museum of natural history. the smithsonian spiro collection is under the care of the Department of anthropology at the national Museum of natural history. the collection came to the Museum through 14 accessions between 1936 and 1986. the largest portion was acquired from harry M. trowbridge in 1958. of particular note in the collection are marine shells engraved with a wide variety of human and animal images. the collection also includes pigments, basketry, cloth- ing with dyed designs, pipes, weapons, ornaments, containers, and figurines made from several different materials. Many of the artifacts are made from raw materials that were acquired by the spiro people through an extensive trade network extending from the atlantic to the Pacific ocean and from the upper Midwest in the north to central Mexico in the south. cover image: Detail of Figure 2.1, craig Mound in 1913 prior to excavation. Published by sMithsonian institUtion schoLaRLY PREss P.o. Box 37012, MRc 957 washington, D.c. 20013-7012 www.scholarlypress.si.edu text and images in this publication may be protected by copyright and other restrictions or owned by individuals and entities other than, and in addition to, the smithsonian institution. Fair use of copyrighted material includes the use of protected ma- terials for personal, educational, or noncommercial purposes. Users must cite author and source of content, must not alter or modify content, and must comply with all other terms or restrictions that may be applicable. Users are responsible for securing permission from a rights holder for any other use. Library of Congress Cataloging-in-Publication Data sievert, april Kay, 1953? artifacts from the craig Mound at spiro, oklahoma / april K. sievert with J. Daniel Rogers ; contribution by Javier Urcid. p. cm. ? (smithsonian contributions to anthropology ; no. 49) includes bibliographical references and index. 1. spiro Mounds archaeological state Park (okla.)?antiquities. 2. indians of north america?oklahoma?spiro? antiquities. 3. indians of north america?Material culture?oklahoma?spiro. 4. Material culture?oklahoma? spiro?history?to 1500. 5. Mounds?oklahoma?spiro. 6. Excavations (archaeology)?oklahoma?spiro. 7. spiro (okla.)?antiquities. i. Rogers, J. Daniel. ii. Urcid, Javier. iii. title. E78.o45s6 2011 976.6?79?dc22 2010042354 issn: 0081-0223 (print) / 1943-6661 (online) the paper used in this publication meets the minimum requirements of the american national standard for Perma- nence of Paper for Printed Library Materials Z39.48?1992. List oF FigUREs vii List oF taBLEs xi PREFacE by April K. Sievert xiii 1 sPiRo anD thE DEVELoPMEnt oF a REgionaL sociaL sYstEM 1 J. Daniel Rogers a chronology of site organization 2 Evans Phase components 2 harlan Phase components 2 norman and spiro Phase components 7 Patterns of change 8 conclusions 11 2 histoRY oF sPiRo REsEaRch 13 April K. Sievert and J. Daniel Rogers Early and commercial Excavations at spiro 13 Professional Excavation at spiro 15 a new synthesis 17 3 sPiRo coLLEction at thE sMithsonian 19 April K. Sievert accession history 19 Early Years: 1936?1958 19 the trowbridge acquisition 21 the Meyer acquisition 22 Later accessions 23 Previous studies of the smithsonian Materials 23 conservation 23 Research 23 the Present study 24 historical sources 24 artifact analysis 24 contents i v ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y 4 containERs 27 April K. Sievert ceramics 27 style 27 Raw Material 29 Manufacture 30 Use 30 grog-, grit-, and Bone tempered types 30 shell-tempered wares 35 slipped wares 36 other ceramic Pieces 38 other wares 38 shell cups 39 condition and conservation 39 Raw Material 39 Residues 40 Manufacture 41 Use 42 taphonomy 43 Design styles 43 Unengraved shells 51 wooden Vessels 52 stone Bowls 53 other containers 54 5 FiBER aRtiFacts 55 April K. Sievert Raw Materials 57 cordage 58 Yarn 58 Braid 59 wrapped Basket Fiber 59 cloth 60 styles 60 spaced weft twining 61 wrapped warps 64 twined tapestry 65 simple weave 65 Fabric Decoration 66 Beads and textiles 66 Function and Use 67 Basketry 67 square shallow Baskets 68 coiled Baskets 69 twilled Matting or Baskets 69 open-work weave 70 twined Matting 70 Basketry impression 70 Fiber identification 71 N U M B E R 4 9 ? v 6 wEaPonRY anD EXtRactiVE tooLs 73 April K. Sievert chipped Lithic weaponry 73 style 73 Manufacture 74 hafting 74 Lithic Raw Materials 75 Use and curation 76 small Projectile Points 76 Large Projectile Points 89 Large Bifaces 91 haftable Bifaces 91 special chipped-stone weaponry 93 sword-Form Bifaces 93 Maces 95 celts 97 other weaponry components 100 Utilitarian artifacts 101 chipped stone 101 ground stone 103 7 oRnaMEnts anD DEcoRations 105 April K. Sievert Earspools 105 Pulley-shaped Earspools 105 other Earspools 112 Beads 113 shell Beads 113 stone Beads 121 copper Beads 122 Pendants 123 shell Pendants 123 Unusual shell Forms 125 stone Pendants 126 Bone Pendants 127 ceramic Pendant 127 gorgets 127 shell gorgets 127 Bone gorget 132 headdresses 133 Pins 133 headdress Plaques and Plates 135 cone ornaments 135 PLatEs 137 8 RitUaL EQUiPMEnt anD cEREMoniaL MatERiaLs 149 April K. Sievert Manufactured objects 149 Pipes 149 v i ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y Discoidals 159 copper 160 Figurines 162 Bone objects 166 wooden objects 168 Leather 170 other worked objects 170 Unworked Ritual Materials 171 Pigment 171 Minerals 171 other Materials 172 summary 173 9 hUMan REMains 175 Javier Urcid catalog number 423344 175 catalog number 423345 175 catalog number 423346 176 catalog number 423347 176 catalog number 423348 176 catalog number 423349 176 catalog number 423350 176 catalog number 423351 176 catalog number 448935 177 catalog number 448936 177 summary Remarks 178 10 concLUsion 179 April K. Sievert Representativeness of the collection 179 significance of the collection 180 the spiro collection and Mississippian Research 181 aPPEnDiX a: cataLog sUMMaRY 183 aPPEnDiX B: concoRDancE oF REFEREncEs 211 REFEREncEs 221 inDEX 227 List of Figures 1.1. contour map of spiro site showing the location of major features and recent excavations 3 1.2. grave lot periods and cultural phases 4 1.3. Excavations at house Mound 5 at the spiro site 5 1.4. ?high-status? artifacts per burial from selected spiro phase (ad 1250?1450) sites 10 2.1. craig Mound prior to excavation 14 2.2. north cone of the craig Mound during excavation by the wPa 16 4.1. ceramic attributes and key 28 4.2. ceramic vessel profiles 31 4.3. whole and partial vessels 32 4.4. crockett curvilinear incised bowl 33 4.5. Rollout of design from two incised vessels 33 4.6. Pennington Punctate rim sherds 34 4.7. carinated bowl 34 4.8. incised sherds 35 4.9. Effigy adornos 36 4.10. slipped and decorated rim sherds 37 4.11. non-caddoan southwestern sand- and grit-tempered sherds with black-on-white design 38 4.12. Busycon, or whelk shell, outer surface 40 4.13. Engraved Pleuroploca, or conch shell 40 4.14. Unengraved shell cups with exterior basket impressions 41 4.15. Residue on interior shell surface 41 4.16. well-preserved cup showing form of finished artifact 41 4.17. Detail of incised surfaces showing tool marks and overcutting 42 4.18. Drilled holes in shell apex perhaps indicating a repair 42 4.19. Braden a style entwined snake motif 44 4.20. Braden a style cup showing spiral designs similar to spiro Engraved ceramic styles 45 4.21. Braden B snake motif 45 v i i i ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y 4.22. Braden B bird motif with court-card symmetry 46 4.23. Mace, ogee with streamers, human head, and bilobed arrows on two sides of Braden B style shell 46 4.24. snake and warfare motifs in Braden c style 46 4.25. spiral patterns executed in Braden c style 47 4.26. Distal fragment of craig a style cup excised in low relief with incised decoration 48 4.27. craig a style carving showing birdman theme 48 4.28. Fragments of craig a style cup depicting bison 48 4.29. craig B style cup showing fabulous animal with complex forked-eye motif 49 4.30. Fabulous beast with serpentine, avian, and raccoon features 49 4.31. craig B cup depicting avian characteristics 49 4.32. craig B petaloid sun icons free floating across the field 50 4.33. craig B cup showing man and snake themes 50 4.34. craig c style cup showing snake and talon motif 50 4.35. craig c cup depicting two men facing a pole 51 4.36. apical fragment of craig c snake and talon cup 52 4.37. wood animal effigy bowl handle 53 4.38. wood bird head effigy perched on bowl rim 53 4.39. Limestone bowl with human effigy adornment 54 4.40. Limestone bowl with snake head adornment 54 5.1. one of trowbridge?s boxes of textile fragments 56 5.2. Yarns 58 5.3. cord-wrapped paddle 59 5.4. Braid 59 5.5. Braided rope 60 5.6. twined and resist-dyed cloth with ties 60 5.7. Bundles of spaced weft twining 61 5.8. Bundle, spaced weft twining 62 5.9. Bundle of cloth 62 5.10. spaced weft twining showing resist-dye pattern 63 5.11. spaced weft twining 63 5.12. spaced weft twining close-up 63 5.13. spaced weft twining 64 5.14. spaced weft twining 64 5.15. spaced weft twining tight weave 65 5.16. twined tapestry 65 5.17. twined tapestry 66 5.18. twined tapestry showing human face 66 5.19. simple weave cloth with resist-dye decoration 67 5.20. close-up of simple weave cloth with resist-dye decoration 67 5.21. square shallow basket 68 5.22. coiled baskets 69 5.23. Large mat or basket fragments 70 5.24. twill basketry 71 6.1. Flaked lithic artifact nomenclature and measurements 74 6.2. attributes of biface form 74 6.3. common forms of damage to projectile points 77 6.4. small projectile points 77 N U M B E R 4 9 ? ix 6.5. small projectile points 77 6.6. small projectile points 78 6.7. small projectile points 78 6.8. small projectile points 79 6.9. small projectile points 79 6.10. small projectile points 80 6.11. small projectile points 80 6.12. small projectile points 86 6.13. Large and unusual points 90 6.14. Large bifaces and drills 92 6.15. Large triangular preform or celt-form biface 93 6.16. Duck River bifaces 94 6.17. general sword-form bifaces 95 6.18. Maces 96 6.19. Depiction of mace on Braden B shell cup 97 6.20. celts: nomenclature and measurement 98 6.21. Elongate celts 99 6.22. cannel coal spatulate celts 99 6.23. spatulate celts 100 6.24. atlatl weights 101 6.25. other stone tools 102 6.26. abrading tools 103 7.1. Perforated pulley-shaped stone earspool fragments 106 7.2. Perforated pulley-shaped earspool with concentric circle decoration 108 7.3. Earspools with chevron decoration 109 7.4. Perforated pulley-shaped earspool 109 7.5. Divided flange earspools 111 7.6. composite wood and shell earspools 111 7.7. ceramic flanged ring earspool 112 7.8. canine wooden effigy earspools 112 7.9. Manufactured shell bead types 114 7.10. concave disc beads 116 7.11. spherical columella beads 116 7.12. cylindrical beads 117 7.13. concavo-cylindrical ?phalange? beads 117 7.14. Bulging beads 118 7.15. hexagonal and asymmetrical convex beads 118 7.16. shell mask bead 119 7.17. whole shell beads 120 7.18. Pearl beads 121 7.19. Phosphate beads 122 7.20. copper disc beads 122 7.21. small and medium copper beads 123 7.22. cat-man warrior wearing necklace of columella pendants and displaying scalps 124 7.23. sculpted columella pendant 124 7.24. Pendants 125 7.25. sculpted shell pendant 125 7.26. shell pendant 126 x ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y 7.27. shell hooks 126 7.28. cannel coal pendant 126 7.29. stone pendants 127 7.30. circular fenestrated gorget depicting sun symbol 128 7.31. craig a gorget fragments 129 7.32. gorget showing man facing forked pole 129 7.33. craig B style gorget depicting avian subject 130 7.34. noncircular fenestrated gorgets 130 7.35. gorgets with woodpecker head and Janus-faced figure 131 7.36. gorget made from Cassis shell 131 7.37. Masks 132 7.38. Bone gorget 133 7.39. copper pins 133 7.40. Bone pins 134 7.41. Bone awls, pins, or piercers 135 7.42. other ornamental objects 136 8.1. t-shaped pipes 151 8.2. cylindrical bowl t-shaped pipes 152 8.3. Decorated t-shaped pipes 153 8.4. stone pipes 153 8.5. ceramic and stone pipe fragments 154 8.6. Feline effigy pipe 155 8.7 Effigy pipe bases showing design detail 155 8.8. Feline effigy pipe 155 8.9. crouching human effigy pipe 156 8.10. human effigy pipe 157 8.11. human effigy pipe, pipe holder style 157 8.12. human effigy pipe, pipe holder style 158 8.13. two-headed human effigy pipe 158 8.14. human-bird effigy pipe 159 8.15. Discoidals 160 8.16. copper sheet artifacts 161 8.17. wooden figurines 163 8.18. wooden figurine, seated ?spiro Man? 163 8.19. galena figurine 164 8.20. shell figurine 165 8.21. shell cameo maskettes 166 8.22. Beaver leg bone bundle tied with cordage 167 8.23. Bone spatulas 167 8.24. wooden artifacts, blade-like forms 169 8.25. wooden spatulate artifact 170 List of tables 3.1. nMnh accessions of material from spiro 20 3.2. nMnh collection summary 25 4.1. ceramic types and sherd counts 29 4.2. characteristics of complete vessels 30 4.3. characteristics of shells cups 44 4.4. Engraved shell fragments not reported in Phillips and Brown (1978, 1984) 51 5.1. styles of textiles in the nMnh collection 57 5.2. characteristics of basketry in the nMnh collection 68 5.3. Fiber identifications for craig Mound textiles 72 6.1. small point styles and raw material summary 81 6.2. alba points measurement summary 81 6.3. hayes points measurement summary 81 6.4. agee points measurement summary 82 6.5. Massard points measurement summary 84 6.6. scallorn points measurement summary 85 6.7. Reed points measurement summary 86 6.8. Keota points measurement summary 86 6.9. Reed/Keota variety measurement summary 87 6.10. Presence of postmanufacturing heat alteration 89 6.11. Large point styles and raw materials 90 6.12. Large biface types and raw materials 93 6.13. Mace types, characteristics, and raw materials 97 6.14. Elongate celt characteristics and raw materials 98 6.15. spatulate celt characteristics and raw materials 100 6.16. atlatl weight characteristics 101 6.17. Frequency of utilitarian tools 101 7.1. counts of ear ornaments by style 106 7.2. characteristics of perforated pulley-shaped earspools 107 7.3. Earspools by raw material 108 7.4. Frequency and characteristics of shell beads 115 7.5. Variety and range of gastropod and bivalve shells 120 x i i ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y 7.6. Frequency and characteristics of stone beads 122 7.7. Frequency and characteristics of tubular copper beads 123 7.8. Frequency and characteristics of columella pendants 124 7.9. Frequency and characteristics of shell gorgets 128 7.10. Fraudulent copper pins 134 8.1. characteristics of stone and ceramic pipes 150 8.2. characteristics of effigy pipes 154 8.3. Discoidal characteristics and raw materials 160 8.4. sheet copper artifact fragments 161 8.5. human effigy figurines and maskettes 162 8.6. Frequency and characteristics of bone artifacts 167 8.7. Frequency and characteristics of wooden objects 168 8.8. Burned mineral lump measurement summary 171 8.9. Unmodified galena nodule measurement summary 172 9.1. types of human teeth in the collection 176 Appendix A a.1. summary of artifacts in the spiro collection 186 Appendix B B.1. Reference summary of the nMnh spiro collection 212 Preface i n 1992, when i first glimpsed the small galena figurine reposing quietly in the drawer where it had crouched for 30 years, i frankly did not give it much thought. since then, that figurine has come to embody, for me, the in- tense specialness of the site of spiro. galena Man sits cross-legged, his limbs and facial features crudely modeled in a solid lump of the dull, gray mineral. while in the craig Mound, he must have been sitting on matting because he still carries basketry impressions and residues. as my work in wringing information from the spiro collection continued, i gained more and more appreciation for galena Man, wondering, just what does he represent? he has a kind of power? solidity and an abiding presence that transcends the mineral itself and perhaps even the culture that made him. he commands respect. there is no doubt that this collection is special, ceremonial, and unique. the legendary engraved shells provide the best glimpse we have into caddoan ico- nography. the well-preserved textiles testify to caddoan artistry and skill. the pieces at the national Museum of natural history (nMnh) represent a large and varied collection, and this monograph provides a glimpse into that collec- tion. Many of the artifacts have been pictured elsewhere over the last 70 years or so and are thus familiar. others, like galena Man, are relatively unknown. this work began as an attempt to document the large collection from the craig Mound at spiro for the Repatriation office of the nMnh. J. Daniel Rogers spearheaded this project and then hired me to do the bulk of the data collection. Little did i realize that the next year would be spent documenting over 20,000 artifacts and generating more than 2,000 pages of text. the project generated a catalog of spiro?s materials, and a summary report was submitted to the nMnh Repatriation office in 1992. since then, Dan Rogers and i have been working on this version, with Dan providing the overall context for spiro and me giving the basic artifact description. the chapters that follow are or- ganized topically, beginning with caddoan context, and followed by historical information about spiro and the collection that ultimately found its way to the smithsonian institution. chapters on containers, textiles, weapons and tools, ornaments, and ceremonial objects follow. human remains in the collection were analyzed by bioarchaeologist Javier Urcid with the assistance of osteologist April K. Sievert, Department of Anthropology, In- diana University, Student Building 130, 701 East Kirkwood Avenue, Bloomington, Indiana 47405, USA. J. Daniel Rogers, Department of Anthropol- ogy, National Museum of Natural History, Smith- sonian Institution, P.O. Box 37012, MRC 112, Washington, D.C. 20013-7012, USA. Correspon- dence: A. Sievert, asievert@indiana.edu. Manuscript originally accepted for publi- cation by Smithsonian Institution Press 27 May 2004; revised manuscript received by Smithson- ian Institution Scholarly Press 5 April 2010. SISP acknowledges that a primary factor contributing to the significant delay in publication of this work was the closure of the former Smithsonian Institu- tion Press in January 2005. x i v ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y Beverly Byrd, both formerly of the nMnh Repatriation office. these results are presented by Urcid as chapter 9. Finally, appendix a provides a list of the catalog numbers along with basic data and appendix B provides a concor- dance of catalog numbers with references to other pub- lished sources on many of the artifacts. help came from many quarters. the advice and opinions of James B. griffin and James a. Brown were invaluable. candace green, Karen Dohm, Dennis Peter- son, Robert Brooks, Don wyckoff, and timothy Baugh offered insights into various aspects of the collection. Bill Billeck provided an immense service by editing the origi- nal report to the Repatriation office. Jai alterman, sarah Flores, and Marcia Bakry contributed very substantially to the final editing of text and the preparation of images. sheree sievert gave me valuable advice for editing this version. Repatriation photographers Jerome Edwards and sarah Zabriskie, independent curator Pamela Keech, and James Brown assisted with the prodigious task of taking record photographs. the staffs of the national anthropological archives and the nMnh Registrar?s office were continually helpful. Registrar for the nMnh Repatriation office at the time of the study was Mar- jorie wilson. she not only tried to keep me organized, but she also spent huge amounts of time photocopying the resulting 2,500 pages of catalog supplement. since doing the analysis, discussions with Joe watkins, Karen wise, Lane Beck, and anne Pyburn gave me new per- spectives. Valerie cressler proved invaluable to the effort to correlate the artifacts with references and illustrations published elsewhere, and her work is evident in the con- cordance (appendix B). i am indebted to alex Barker, James Brown, and Bruce smith for reviewing the manu- script and suggesting improvements. as ever, Mike snow and sheree sievert put up with a lot. this study is offered as a testament to the important cultural traditions of a people whose descendants are still with us today and are still contributing their own unique perspectives to american diversity and cultural life. April K. Sievert Indiana University Bloomington, Indiana B eginning in the eighth and ninth centuries ad, a new cultural system took shape in the eastern woodlands of north america, a region ex- tending from the Midwest to the gulf of Mexico to the Mid- atlantic. the groups already living in this vast region either adopted or them- selves developed a way of life that accounted for some of the most complex so- cial achievements in prehistoric america, north of Mexico. this cultural ?style? is referred to as Mississippian and consisted of populations scattered across the region organized into chiefdoms, or what is sometimes referred to as ranked or middle- range societies (smith, 1990:1?8; Barker and Pauketat, 1992; Bense, 1994; Pauketat, 1994, 2007). as diverse as these societies were, they also shared many attributes, almost always including the cultivation of maize and other na- tive domesticates (smith, 1986), the construction of earthen mounds, and par- ticipation in a religious system known as the southeastern ceremonial complex (galloway, 1989; King, 2007). During the last two decades the quality and quantity of the archaeological data have improved, and we are now able to say much more about the nature of these societies. in fact, the eastern woodlands have become a major laboratory for studying the emergence of social complexity and variation within middle- range societies. archaeologists are now investigating settlement patterns, trade networks, evidence for social hierarchies, social memory and identity, the con- struction of adaptive systems, and many other factors associated with the rise of social complexity and regional similarities and differences (e.g., scarry, 1996; schroeder 2004; wilson 2010). throughout much of this research, there has been an emphasis on community and religious centers usually marked by the presence of one or more earthen mounds. Lately, this emphasis has given way to a more balanced consideration of all types of sites, from the grand mound centers like cahokia (Fowler, 1989; Brown and Kelly, 2000; Pauketat, 2009) to the lowly individual farmsteads (Rogers and smith, 1995). Even so, it has always been recognized that the mound centers offer great interpretive potential and must be included in any analysis of social change. the spiro site is a major mound center, and there can be little doubt that it played a significant role in the development of the prehistoric societies in the northern caddoan region, an area including the arkansas Basin and ozark 1 spiro and the Development of a Regional social system J. Daniel Rogers 2 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y highlands of oklahoma, arkansas, and Missouri. in fact, spiro was, by all available evidence, a ceremonial center of paramount importance (orr, 1946; Rogers, 1991a, 2009a; schambach, 1993; Brown, 1996). given the relevance of this site in both regional social dynamics and long- distance interactions, it is critical to understand the physical organi- zation of spiro itself. this chapter evaluates the changing layout of the spiro site in relation to local and regional de- velopments and assesses the extent to which changes in site organization can be used to evaluate the relationship be- tween regional interaction and the apparent development of increasingly centralized authority. this approach serves to contextualize the artifacts, from both a site and a regional perspective. the final section analyzes spiro?s external rela- tionships in the context of social developments of the cad- doan tradition in the arkansas Basin and ozark highlands. a chronoloGY of Site orGaniZa tion the spiro site (Figure 1.1) was utilized for a period of at least 550 years, beginning about ad 900 and ending at approximately ad 1450. Recently, the spiro chronology has been revised by James a. Brown on the basis of a de- tailed seriation of grave lots from craig Mound (Duffield 1973; Brown, 1996:153?167; see also Brown and Rogers, 1999; Rogers, 2006). the analysis prepared by Brown is a sophisticated study not only of the burial associations but also of marker attributes with regional and extraregional significance. the grave lots at spiro were ordered into four grave periods with several subperiods for a total of seven material and contextual clusters with chronological sig- nificance. Rather than the previous four phases, Brown now recognizes five cultural phases. these five phases are Evans, harlan, norman, spiro, and Fort coffee (Fig- ure 1.2). Each phase is discussed below in relation to the mounds and other features at the site. Evans PhasE ComPonEnts the earliest securely recognized components at spiro are representa tive of the Evans phase (ad 900?1050 or 1100). it should be noted, however, that some burials at the craig and ward Mounds are very similar to those of the earlier Fourche Maline phase. the term Fourche Maline refers to primarily woodland period hunting and gathering groups that lived across the caddoan region (schambach, 1998). the Fourche Maline phase as defined specifically for eastern oklahoma is based primarily on materials from the wister Valley and vicinity, 30?40 km south of spiro (galm, 1978, 1981, 1984; galm and Flynn, 1978; Bell, 1980). with the support of numerous radiocarbon dates, the Fourche Maline phase is defined as dating from circa 300?200 bc to ad 700?800 (galm, 1984:207). some pre- mound burials at craig and ward 1 Mounds and burials from ward Mound 2 (Rogers, 1980:111) exhibit Fourche Maline characteristics. if the burials in question can be attributed to the Fourche Maline phase, then they prob- ably date to the latest part. the artifacts associated with these burials are characteristic of the Fourche Maline phase but also occur in the later Evans and harlan phases (cf. galm, 1984:214?215). other premound burials at craig are somewhat more elaborate and include Le Flore Plain ceramics, associated with late Fourche Maline, Evans, and harlan phases, and larger quantities of ornate artifacts. these burials date to the Evans and the even later norman phases (Brown, 1996:166). whether or not it may eventu- ally be possible to isolate a Fourche Maline component, it is apparent that from its beginning the spiro site was a place of ceremonial importance. there is virtually no indication of a Fourche Maline habitation component on the site; only the earliest burials at the craig and ward Mounds indicate how the site may have been used in this phase. the Evans phase (ad 900?1050 or 1100) marks the beginning of recognizable Mississippian influences and the basic characteristics that were to become caddoan. at spiro the Evans phase is defined principally on the basis of a series of burials from the lower levels of craig Mound (Brown, 1996:137). a series of trade ceramics in these burials indicates the origin of a regional trade network and potentially other types of connections with the Plum Bayou culture in central arkansas (Rolingson, 1987). whereas williams Plain and Le Flore Plain ceramics, of local manufacture, continue to be present, other types also occur, including coles creek incised var. Plum Bayou and var. Lonoke and French Fork- like agee incised (Brown, 1996:162). other than the few burials at craig Mound, Evans phase assemblages are tentatively also recognized at house 4 and a midden area northwest of craig Mound (Rogers, 1980:115, 121; Brown, 1996:165). harlan PhasE ComPonEnts the first large- scale utilization of the site occurred in the harlan phase (ad 1050 or 1100?1250). this phase is named after the harlan site, another of the mound centers in the region, with its primary utilization concentrated in this time range (Bell, 1972; Rogers, 2009b). During this phase the site continued to expand, both physically and in terms of its role as a ceremonial center. Beginning with FI G U R E 1 .1 . c on to ur m ap o f s pi ro s ite s ho w in g lo ca tio n of m aj or fe at ur es a nd r ec en t e xc av at io ns . a da pt ed fr om R og er s (1 98 2: 4) w ith p er m is si on o f t he o kl ah om a a rc he ol og i- ca l s ur ve y. 4 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y the harlan phase, it is possible to talk about an overall organizational pattern that served to segregate activity areas for approximately the next 400 years. During the harlan phase additional burials were placed in the early construction phases of craig Mound and at ward Mound 1, a number of domestic buildings were constructed on the lowland portion of the site, and platform mounds and ?house mounds? came into use on the uplands. these new constructions produced a tripartite spatial arrangement at the site. Running from west to east, the structure can be characterized as areas relating to rituals of death, daily living activities, and disposal of the dead. the first of the major spatial divisions, encompassing the western portion of the site, may be inferred to be an area relating to rituals of death. the mounds and other features in this segment of the site signify ritual perfor- mances, some of which were apparently linked to mor- tuary or charnel house activities (explained below). this western area consists principally of a ring of mounds en- closing an area of approximately 2.75 ha, referred to as the ?Plaza? (Rogers et al., 1982). the mounds that make up the ring consist of house Mounds 1?4 and 6, Dollop Mound, copple Mound, and Brown Mound. also associ- ated with this area are house Mound 5 and Plaza Mounds a, B, and c. Five of these mounds (Dollop Mound, house Mound 5, and Plaza Mounds a, B, and c) were only iden- tified after the resumption of excavations at the site in 1979 (Rogers et al., 1982). the house mounds listed here are not substructure mounds as the term is used in Meso- american archaeology but are instead soil mantles built over the remains of presumable special-purpose buildings. house Mound 5 is a good example of the house mound pattern found at spiro and other mound centers in the region as well as in other portions of the caddoan area. although the works Progress administration (wPa) dug some house mounds, house Mound 5 is the only one excavated in recent years using fine- scale recovery techniques (Rogers, 1982). at house Mound 5 a series of three square buildings were discovered, each with four interior support posts and an extended entryway (Figure 1.3). these three buildings averaged roughly 9 ? 9 m in size and were built, used, and apparently intentionally de- stroyed in relatively rapid succession. these buildings, and by implication, the structures associated with other house mounds, are inferred to have served special purposes, perhaps functioning as combinations of charnel houses and the residences of elite functionaries. Bell (1984:229? 230) has argued that similar structures at the harlan site (36cK- 6) served a mortuary/charnel house function. the principle evidence for this interpretation comes from the unusual nature of the build ings and their contents. at house Mound 5 the exceptionally large size of the build- ings (compared to domestic structures of the same phase) and the possibility that the buildings were intentionally destroyed and subsequently rebuilt over almost exactly the same spot with the same directional orientation lend support to the identification of these structures as having a special purpose. Exactly what this purpose was is hard to say; however, the material remains found in the build- ings give some strong clues. of special note were a few Spiro IVB Spiro IV Spiro III Spiro II Spiro IB Spiro IA Fort Coffee Spiro Norman Harlan Evans 1650 1600 1500 1400 1300 1200 1100 1000 Spiro IVC Grave Periods Culture Phases FIGURE 1.2. grave lot periods and cultural phases. adapted from Brown (1996:161). N U M B E R 4 9 ? 5 fragments of human bone and three ovoid ?class a? bi- faces made of an imported chert (Rogers, 1980:168?169). the latter are similar to some bifaces recovered from buri- als at craig Mound. the general character of the artifact assemblage was also suggestive of a special function for these buildings. although the artifacts recovered were not what might be referred to as ?high- status goods? (that is, highly elaborate, decorated, or exotic), they did reflect a set of activities not typical of common domestic structures. in particular, tools associated with primary subsistence ac- tivities, such as hunting and horticulture, were virtually absent, whereas tools related to grinding and cutting ac- tivities were extremely common. it is not clear what was being ground or cut. the artifacts recovered indicate that a special set of activities was conducted at house Mound 5, but it is also probable that the structures were used for domestic purposes by some individuals. this inference is bolstered by the presence of a wide variety of potential animal and plant food remains. in addition to the platform mounds and house mounds associated with the western portion of the site, there was another type of mound identified as Dollop Mound and Mounds a, B, and c. these four mounds lie in an east? west linear arrangement extending from the western edge through the center of the Plaza (Rogers et al., 1982). the function of these mounds is unknown. the mounds FIGURE 1.3. Excavations at house Mound 5 at the spiro site. adapted from Rogers (1982:13) with permission of the okla- homa archeological survey. 6 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y contain a variety of artifactual debris, consisting primarily of common ceramics and lithics. although these mounds received only limited testing, there is presently no indica- tion of stratification or evidence of building remains. it may be that these mounds represent the piles of debris produced by a periodic cleaning of the Plaza or were the foundations for poles or other superstructures. the final type of mound in the Plaza ring consists of the two rectangular platform mounds: copple and Brown. Brown Mound is the larger of the two and was extensively damaged by pot hunters in the 1930s before excavation by the wPa. although the pot hunters caused a lot of dam- age, the wPa crews did identify the remains of a structure on the ground surface under the mound and several dis- crete mound construction episodes. one extremely inter- esting characteristic of Brown Mound, discovered by the wPa, is evidence for a subterranean pit that was dug from an upper level in the mound down through the mound and deep into the premound strata (Bauxar, 1953; Brown, 1996:105?112). whereas information on this feature is incomplete, it may have extended 6 m or more below the original surface. Because of groundwater and time con- straints on their lease, the wPa did not explore this fea- ture. in 1982 an attempt to learn more about this pit was initiated by the oklahoma archaeological survey (Rogers et al., 1989). an extremely high water table and limited funds forced curtailment of the project; however, this test did lend support to the existence of the pit as an aboriginal feature. copple is the second and smaller of the platform mounds, and like Brown Mound it was also partially ex- cavated by the wPa. in 1982 further testing at copple revealed a complex history of mound construction and use (Leonhardy, 1989; Peterson, 1989). the structure of the mound included at least three superimposed build- ings that had existed at early levels near the center of the mound. there was a considerable amount of ?midden? in and near these structures, and as part of the mound con- struction process, soil had been piled up against the sides of one or more of the buildings. at some point, however, each building had been dismantled, probably after being burned, and the space was filled in with soil. it is also ap- parent from the presence of ash and burned soil that a number of fires had existed at various times on different surfaces of the mound. a similar history of mound use can be identified at Parris Mound (34sQ- 12), located 30 km north of spiro (Muto, 1978). Leonhardy (1989) notes that mounds are stereotypically described as being built in stages. a layer of soil was piled up and used for some pur- pose for some time. this initial layer was then covered to make the pile larger and the pile was again used. the pile may have been enlarged several times, either as a platform for some ceremonial structure or as a place to inter the dead. always implicit in the ?construction stage? notion is that the piling of dirt was uniform around the axis of the mound and that each stage was a short- term, single- event phenomenon separated from the succeeding stage by some long- term period of use. Rather than discrete construction stages, it is more appropriate to regard copple Mound as composed of interior, middle, and periphery deposits. whereas mounds were an important focus of research in the western portion of the site, testing of intervening areas did take place. in general, habitation debris outside of the mounds was sparse, although some material was recognized by the wPa in the vicinity of house Mounds 4 and 6, and later testing revealed some midden areas west of house Mound 5, perhaps evidence of additional structures. the central portion of the site, including the slope down from Brown Mound and the flat area east to craig Mound, in the harlan phase, contained evidence of ac- tivities relating to everyday life. this area was tested by the wPa, and at least five square, four- center- post houses were excavated. notes from the testing in this area indi- cated that as many as 15 additional houses might be pres- ent. this number of buildings means that spiro has the densest concen tration of structures known for any harlan phase site in the arkansas Basin. the buildings are not es- pecially large, nor are there any associated house mounds. it appears that the houses are simple domestic structures. although it is possible that these were dwellings for elite, the scarcity of evidence for a sizable harlan phase popula- tion at any other site in the area makes it unlikely that a large (relatively speaking) group of religious or political functionaries resided at spiro in the harlan phase. the ceremonial specialists probably lived in the cyclically de- stroyed buildings buried under house mounds, whereas the buildings in the central portion of the site probably were ordinary domestic structures. the final site area relevant to the harlan phase is the eastern portion, including craig Mound and ward Mound 1. the principle use for this site area concerns final disposal of the dead, although three possible domes- tic structures were excavated in the area east of the ward Mounds. the burials in craig Mound are associated with very early levels, prior to the major mortuary deposits as- signed to the later norman and spiro phases. at ward Mound 1, disturbance by early excavations (thoburn in 1916?1917) and poor record keeping by the wPa limited the inferences that could be drawn, but this mound (and N U M B E R 4 9 ? 7 probably ward Mound 2) includes some of the charac- teristics of the house mounds on the western side of the site. there was probably a structure buried in the mound; additionally, there does not appear to be any stratigraphic differentiation within the mound fill, which might indicate a single construction episode or the continual and gradual addition of soil to the mound. only nine burials were ex- cavated under controlled conditions. some 28 additional burials were removed for which there is little information (see thoburn, 1930). only three burials were recovered from ward Mound 2, and none of these were associated with diagnostic artifacts. norman and sPiro PhasE ComPonEnts the detailed chronological analysis of spiro grave lots conducted by Brown (1996:133?167) made it possible to recognize a series of burials with distinctive material as- semblages at craig Mound that could be attributed to a time frame designated as the norman phase (ad 1250? 1350). in Brown?s (1996:137) grave lot sequence, the nor- man phase is based on grave period iii. at spiro only the craig Mound burials can be attributed to the norman phase; however, within the region there are several compo- nents linked with this time frame (Brown, 1996:163?165; Rogers et al., 2000). in terms of spatial organization, the tripartite ar- rangement that characterized the spiro site in the harlan phase was carried over, with certain important modifica- tions, into the norman and spiro phases (ad 1250?1450). During these periods an amazing growth occurred in the number and variety of exotic trade goods associated with the burials at craig Mound (Brown, 1996: vol. 2). these objects are described in detail and are the principal focus of this volume. the expansion in the trade good inven- tory is not matched, however, by an expansion of the site. in the western portion of the site there is only inferential evidence that the upper levels at Brown Mound can be attributed to the spiro phase plus a single intrusive burial in the flank of the mound. the house Mounds and other mounds associated with the Plaza show no indication of use in the spiro phase. similarly, the small cluster of build- ings that stood in the central portion of the site was aban- doned. the population moved from their close association with the ceremonial center to small clusters of houses scat- tered along the rich farming soils of the spiro vicinity (see Rohrbaugh, 1982, 1985a, 1985b). the major utilization of the spiro site during the spiro phase is reflected in the numerous and elaborate burials from craig Mound. During this phase the major portions of the four separate conical units of the mound are con- structed. Pot hunter digging destroyed much of the upper portions of the three lesser units, but wPa excavations did recover extensive information from undisturbed areas (see chapter 2). within the main cone of the mound the pot hunters tunneled into an open cavity they called the ?central chamber.? although later excavations revealed that this was not an intentionally constructed chamber, an unusual preservation context did exist within this hollow space, thus allowing survival of an extraordinary assort- ment of perishable artifacts (Brown, 1996:85?88). the construction of the main mound unit at craig Mound is characterized by a series of flat- topped stages, some or all of which are encircled by fences of cedar poles. Presumably, a building (probably a mortuary) stood on the summit of each mound level. a ramp providing access to the various levels may have extended to the north. of the several buildings that stood at different levels, there is only one, the great Mortuary, for which any appreciable amount of information exists. this is the mortuary that was revealed in the hollow cavity and partially destroyed by the pot hunters. the great Mortuary is the largest and the only mortuary that did not have its contents removed prior to dismantlement of the building and the addition of a new layer of soil to the mound. Recent research by James a. Brown (2010) suggests that the organization of the great Mortuary and other features at craig Mound potentially represents a spatial cosmology. the overall conical shape of craig Mound and the rounded deposit that caps the last of the platform surfaces in the main unit suggests a planned end to the use of this mound. there is, however, no evidence for the construc- tion of a new mound to begin anew the cycle of mor- tuary use, dismantlement, and final disposal of the dead at spiro or anywhere in the region. the closing of craig Mound seems to be associated with major changes in the arkansas Basin and coincides with the beginning of what is referred to as the Fort coffee phase (ad 1450?1660; Rogers, 2006). there is no verifiable evidence that the spiro site was used during the Fort coffee phase. the definition of the transition from the spiro to the Fort coffee phase is closely linked to the apparent decline of the complex status hierarchy represented in the burials at craig Mound and other regional mound centers (Brown, 1971a; Rogers, 1983, 1995). the decline of the status hierarchy extended to other aspects of spiroan political and religious organization, including a drastic decline in the construction and use of mounds and a corresponding decrease in exchange of exotic materials such as copper or marine shell (Rogers, 2006). 8 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y the truncation of the social hierarchy and the collapse of the eastward- oriented trade network in exotic goods are not reflected in obvious ways at the lower levels of society. the farmers that had lived in the sites surrounding the spiro ceremonial center continued to do so, and the houses they built were like those of the spiro phase (i.e., two center posts, rectangular, with an extended entryway). PattErns of ChangE Multiple patterns relating to cultural changes on a local and regional level are evident within the chronologi- cal ordering of site components described above. the in- terpretive position taken here is that a socially ?external? agent of change, such as environment, serves to provide a ?context of change? that must be considered in conjunc- tion with a variety of social and technological constraints. several factors?population fluctuations, the availability of appropriate cultigens, the presence of natural resources, individual and collective forms of agency, social con- straints on the production of surplus, or factors affecting the centralization of authority?play important roles in understanding the trajectory, extent, or even the general nature of social transformations (e.g., Beck, 2003). an- other consideration in understanding the cultural develop- ments in the study region is the potential significance of interaction between local polities. Most of these factors, but especially interaction and trade, are best considered by adopting a regional perspective that allows consideration of the relationship between sites. in the arkansas Basin and ozark highlands the development of an integrated regional perspective has been surprisingly slow to develop. only since the 1980s have significant efforts been made to integrate information from arkansas, Missouri, okla- homa, and texas (cf. Brown, 1984a; Kay et al., 1989; Pert- tula, 1989; Rogers, 1989, 1991a, 1991b). these studies have highlighted the cultural connections between regions, primarily during the Mississippian period (ad 800?1500), although connections extend far deeper in time (cf. ander- son and sassaman, 1996). From a regional perspective the events and pro- cesses responsible for the expansion of spiro during the Evans and harlan phases (ad 1000?1250) or the estab- lishment of other mound centers in the region are very poorly known. however, the late Fourche Maline and Evans phases represent the beginning of a ranked form of social organization associated with the Emergent Mis- sissippian (ad 800?1000) across the Midwest and south- east (cf. smith, 1990). the beginnings of the Mississippian are linked to a widespread phenomenon with no single point of origin but, rather, with the relatively synchro- nous development of similar social and material systems in a number of different regions (Phillips et al., 1951:451; smith, 1984:20). in the spiro area, changes are a part of the regional caddoan manifestations also associated with cultural developments in northwestern Louisiana, eastern texas, southwestern arkansas, and southern oklahoma. although clearly part of the caddoan tradition, the area of the arkansas Basin and ozark highlands presents local differences that form a distinctive variation on the overall caddoan theme. the chronology of cultural developments in the ar- kansas Basin and ozark highlands is outlined above in the context of the growth and change of the spiro site it- self. these developments have their origin in the transition from the Late woodland to the Mississippian. the Late woodland time frame (ca. ad 600?900) in the study area falls largely into what is locally referred to as the Fourche Maline phase (galm, 1984; Rogers, 1991a:224?225) and is characterized by hunting and gathering groups that used the bow and arrow and made simple forms of pottery. al- though there is some evidence for the presence of groups with Kansas city hopewellian affiliation, these peoples seem to have had little influence on later developments in the area. in contrast, there is a strong material continuity between the Fourche Maline and the Emergent Mississip- pian of the Evans phase. there is also some continuity in settlement location, but the transition to the Evans phase is characterized by a variety of social and technological changes that are far more significant than the areas of constancy with the earlier hunter- gatherer adaptations. around ad 900 several mound centers were established in the region, representing a sharp contrast to previous set- tlement patterns. the new settlement system is organized hierarchically, and sites tend to be located with access to soils suitable for horticulture. at several of these centers there is good early evidence for complex ceremonial activi- ties associated with the construction and use of mounds, mortuary activities, and special- purpose buildings, such as the structures mentioned earlier for spiro?s house Mound 5. the harlan phase is also characterized by increased evi- dence for horticulture, substantial architecture, expanded levels of long- distance exchange, and sharply different forms of social organization. the implications of the transition from the Fourche Maline to Evans phase for the organization of social con- trol are especially significant. in kinship- based societies, presumably characteristic of the Fourche Maline phase, authority is often structured and legitimated by ancestral sanctions (Meillassoux, 1960). Fourche Maline sites in N U M B E R 4 9 ? 9 many cases show long periods of continuity in their utili- zation and are also places for the burial of the dead. these sites embody a conceptual continuity between the activities of the living and the sources of authority provided by the ancestors. with the beginning of the harlan phase there is an abrupt departure from the settlement continuities of earlier times, probably associated with the redefinition of the basis for authority and patterns of organization at all levels of society (Rogers, 1996:56?57). Evidence from the Evans and harlan phases (Rogers, 1983) strongly indi- cates the presence of a ranked, chiefdom or middle- range type of social organization. in these types of societies au- thority is tied to positions recognized through hereditary linkages; however, social control tends to cut across strict lineage relationships to encompass larger segments of the population. the basis for authority is generally through direct access to the supernatural, rather than by appeal to the ancestors as intermediaries. the historic period in the southeast offers good comparative examples of chiefly authority being derived from supernatural sources, such as among the natchez in the Lower Mississippi Valley or the hasinai caddo of east texas (white et al., 1971; Rogers and sabo, 2004). the settlement pattern of the Evans and harlan phases, especially concerning mound centers as the loci of major ceremonial activities, is very different from prior patterns as seen in the Fourche Maline phase. this major differ- ence provides a basis for interpreting the Fourche Maline/ Evans transition. however, the social, technological, and environmental ?mechanisms? responsible for instigating the transition are, not surprisingly, poorly known; even so, some suggestions may be offered. in particular, increas- ing evidence for exchange and long- distance interaction is probably closely related to the general phenomenon of Mississippian emergence (Rogers, 1991a). on a local basis, control of exchange may have provided a means for individuals in positions of some authority (e.g., lineage heads) to gain wealth, prestige, and access to objects and information of ritual significance. attempts by individuals to consolidate authority could lead to a cycle of increas- ing control, ultimately contributing to the new form of social organization. attempts by individuals to consoli- date power are seen as a principal mechanism, although there are certainly other key variables in the equation. the availability of appropriate technologies and environmen- tal opportunities are especially important. in this regard, the arkansas Basin and ozark highlands were relatively sparsely inhabited and represented a region ripe for the ex- pansion of a horticultural adaptive strategy. the available data provides no basis for suggesting that environmental conditions in the ad 700?900 period would have signifi- cantly deterred the practice of horticulture in the region (henry, 1978; henry et al., 1979). with regard to the technology of maize horticulture, there is currently little evidence for the cultivation of tropical or indigenous cul- tigens from Fourche Maline sites. But the necessary tech- nology could have easily been made available at the Evans phase (ad 850?1250) transition through the expanding interaction networks. there is confirmation of maize in the southeast by around ad 400 (icehouse Bottom site; smith, 1986:44), plus there is evidence for maize (Maiz de ocho) of southwestern origin by ad 780 at the Davis site, an early caddoan site in texas (story, 1981:149). often a strong link has been made between sedentism and the adoption of agriculture, and it would seem that this is justifiable, to some extent, in the northern caddoan case. there are, however, extenuating circumstances, and as various researchers have pointed out, it may be inaccurate to link sedentism and agriculture too closely (hitchcock, 1982:224?225, 230; Powell, 1983; Rafferty, 1985:134). in the arkansas Basin and ozark highlands, maize or other cultigens are a relatively small component of paleoethnobo- tanical samples prior to ad 950 (wyckoff, 1980:419), and even after that date maize, in any caddoan site, is relatively scarce compared to sites in other areas of the southeast (schambach, 1982:191). this absence indicates that maize may not have played a strong role in the initial transition to sedentism in the study area. throughout the caddoan sequence maize was heavily supplemented by wild plants foods (Rogers, 2011). in terms of sedentism the northern caddoan area may again prove to be a contrast with adap- tive systems further to the east. an unexplored but still conceivable issue is that the continued emphasis on col- lecting and hunting wild resources was linked to periodic abandonment of settlements for purposes of conducting hunting and/or gathering expeditions. the data neces- sary to investigate this issue are, as yet, largely unavail- able. whether or not the northern caddoans did occupy their dwellings year- round, population densities were low and settlements were well dispersed across the landscape, which would have expedited access to not only a variety of natural resources but also soils suitable for agriculture (wyckoff, 1980). the relatively small population and dis- persed farmsteads of the caddoan settlement system have been viewed as a strategy for mitigating the effects of a region with unpredictable rainfall (schambach, 1982:192). with the establishment of mound centers such as spiro in the ad 800s and 900s the settlement system takes on a hierarchy of site complexity that correlates with the ranked forms of social organization associated with the 1 0 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y Mississippian period (Brown et al., 1978). During the Evans and harlan phases (ad 1000?1250) the harlan and spiro sites are roughly on the same level of complexity, although there are indications that even at this early date spiro incorporates some subtle differences that would seem to place it at the apex of the settlement system. namely, there is a wider variety of exotic and presumably high- status artifacts associated with the burials at spiro than at any other site in the region (Figure 1.4). this fact is con- sidered an indication of the level of control spiro exercised over the exchange of objects used as status markers and symbols of authority and ideological legitimation (Rogers, 1996). By the succeeding norman phase (ad 1250?1350) spiro accounts for the vast majority of exotic prestige goods. the variation in the presumed control of exotic trade items by spiro in the Evans and harlan phases does not necessarily mean that the northern caddoan region was operating as a single polity at this time. it is entirely conceivable that two or more systems were in operation at various times during these phases (Rogers, 1983:60), although as Berthoud and savelli (1979:748) point out, the exercise of political power is strengthened by privilege; simultaneously, privilege increases with the permanence of power. Prestige goods therefore become what Baudrillard calls a ?sign value?, embodying the ideology of domination. as such, spiro?s role in the region may have been enhanced by an early control of exchange ties with the East, closely linked to participation in the southeastern FIGURE 1.4. comparison of mean number of ?high- status? artifacts per burial from selected spiro phase (ad 1250?1450) sites. From Rogers (1983:69) with permission of the oklahoma archeological survey. N U M B E R 4 9 ? 11 ceremonial complex. By the norman phase the larg- est population concentration is in the immediate vicin- ity of spiro, further supporting the notion that spiro is in some way dominating, if not literally controlling, the ideological, social, political, and economic interactions in the region. the transition from the harlan to the norman phase incorporates a number of changes in addition to the cen- tralization of control at the spiro site. Domestic and cer- emonial architecture changes across the region from large (mean floor area of 46 m2) square, four- center- post build- ings to smaller (mean floor area of 37 m2) rectangular, two- center- post structures. such a change has strong impli- cations for the composition of the household plus changes in kinship organization on a larger scale (Rogers, 1995). there is also evidence for shifts in settlement distribution in the region (wyckoff, 1980:514) and for a number of changes in artifact styles. the changes in building type and settlement location are also linked to apparent changes in regional population levels. a review of information on buildings from the harlan and spiro phases (excluding those with special or presumed ceremonial functions) was made in order to compare the relative number of struc- tures and the mean floor area for each period. if these data are used as a rough indication of population trends, then the far greater number and larger size of domestic build- ings in the Evans and harlan phases in comparison to the norman and spiro phase suggest that the earlier periods had the greatest population. Population seems to decline in the norman and spiro phases, with the trend continu- ing into the following Fort coffee phase (ad 1450?1660) and ending with the eventual abandonment of the region in the protohistoric period. admittedly, the data for suggesting this population trend are suspect on the grounds of unsystematic sam- pling, yet if the inclination holds true under eventually closer scrutiny, then an inverse relationship would exist between population and social complexity. such a rela- tionship is contrary to the cultural evolutionary argument that generally acknowledges a link between population increase and augmented social complexity. it may be that the changes recognized as taking place between the har- lan and norman phases are not sufficiently sweeping to be considered as significant from a cultural evolutionary perspective. Perhaps this is true, but it is also relevant to consider the actual or potential role played by a process such as the centralization of authority, as far as popula- tion fluctuations are concerned. increased social complex- ity of this type does not inherently require an increased population, nor is it necessarily a response to social or environmental stress. it is not requisite to view increased complexity as a form of intensification. with increasingly centralized authority there will, however, be a leadership- based demand for increased production to supply the con- sumption needs of the elite and for their participation in long- distance exchange, but that increase could be met by shifting production rather than a long- term response such as increasing population to increase production. what seems to have been happening in the spiro case is a mode of social intensification at the elite level but not necessarily at the level of household production. it was noted some time ago (Plog, 1974:120) that in the southwest, social intensification and increasing agriculture correlated with a decline in population. whereas Johnson (1982) observed this on a cross- cultural basis, the correlation between pop- ulation and organizational complexity is relatively weak when comparisons are made using fine- grained popula- tion level estimates. these observations open the door to a more varied and realistic interpretation of the relationship between population levels and social change. concluSionS the spiro site was, throughout its history, an inte- gral and often dominating part of the trends in social and technological change characterizing the late prehistory of the arkansas Basin and ozark highlands. it was one of the first mound centers established in the region (perhaps sometime between ad 850 and 950), and it continued to grow in importance until the collapse of the chiefdom level of social organization around ad 1450. During the Missis- sippian period, mound centers in the caddoan area tended to be ?vacant? or sparsely inhabited. spiro certainly fits this pattern, especially during the norman and spiro phases. this was the time in which most of the elaborate burials were placed in craig Mound, but it was also a time of less extensive but more focused use of the site. it is likely that very few people lived at spiro during the spiro phase, perhaps only those individuals responsible for cer- tain ceremonial duties. activities at the site were concen- trated on the sequentially constructed buildings at craig Mound and perhaps the buildings that may have stood on the upper layers of Brown Mound. other than these struc- tures, probably no more than two at any one time, there is no other evidence for buildings at the site. this pattern of vacant ceremonial centers may be viewed as a physical reflection of the conceptual and cosmological separation of the elite, or at least the context of ceremonial activity, from ordinary people and domestic activities. 1 2 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y at about ad 1450 activities at spiro came to an end, probably associated with the collapse of the social system and the general decline of the Mississippian chiefdoms across the southeast beginning about this time. as with the beginning of ranked forms of social organization in the caddoan area, their decline also seems to have been linked with developments further to the east. whether environ- mental variability played a significant role in the demise of the spiro system, it does seem that change in the level of interaction with the East was a critical factor. a decline in the exchange of exotic goods from the East may have severely impacted the material sources of chiefly ideologi- cal legitimization and eroded one aspect of the basis for authority. complex chiefdoms are inherently unstable and prone to decentralization. if trade with the East was on the decline, for whatever reason, the upper levels of spiro society may have been unable to bridge the symbolic gap. although farming households continued to live in the re- gion, mound construction and other evidence for central- ized authority disappeared. s everal writers have told or retold the story of the excavations at spiro (clements, 1945:48?66; orr, 1946; hamilton, 1952:23?35; Brown, 1966a:23?35, 1996:43?62; hamilton et al., 1974:1?5; Phillips and Brown, 1978:1?8; Rogers, 1980:10?21; La Vere, 2007). the episodes in spiro?s excavation reflect the changes in the way archaeology has been con- ducted in north america during this century. First, the site was subject to severe damage as treasure hunters sought saleable artifacts. then, a massive wPa effort systematically removed huge amounts of fill, still with a goal of artifact recovery and only secondarily to understand the ways in which objects became part of the archaeological record. Finally, excavations in the 1970s and 1980s used a problem- based research approach. the chaotic digging by the early commercial relic hunters is particularly relevant here because the smithsonian collections derive from these early excavations. subsequent wPa excavations recovered materials that serve as the comparative collection to this one. Later excavations revealed site chronology, habitation, and site structure and provide the basis for interpreting spiro. Discussion of these episodes of archaeology at spiro and work done using these collections provide a framework for study of the materi- als from the site that are at the smithsonian institution. earlY anD commercial eXcaVa tionS at SPiro the first recorded excavations at the spiro site took place in 1916?1917 and were conducted by Joseph B. thoburn, a historian from the oklahoma his- torical society (thoburn, 1930, 1931:64?69). thoburn photographed the site prior to any excavations (Figure 2.1). thoburn excavated ward Mound 1 in an attempt to prove that the small ?prairie mounds? of eastern oklahoma were the remains of collapsed earth lodges, similar to those occupied by such groups as the Pawnee or arikara. thoburn?s excavations revealed the remains of a burned structure and several burials, which were, for him, reasonable confirmation of the human origins of the small mounds dotting the landscape. subsequent geo- logical and archaeological work illustrates that thoburn?s use of ward Mound 1 as a test for his earth lodge hypothesis was inappropriate. the so- called prairie 2 history of spiro Research April K. Sievert and J. Daniel Rogers 1 4 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y mounds, though still not completely understood, are geo- logical in origin, not made by people (Brown, 1996:10). thoburn?s contribution was that he identified the prehis- toric remains of eastern oklahoma, including spiro, as be- longing to people of caddoan origin (thoburn, 1931:60). caddoan is a term that refers both to a group of related languages and the cultural traditions that emerged in the four- state region of texas, Louisiana, arkansas, and okla- homa (Rogers and sabo, 2004). the concept of caddoan is a theme that still underlies most research on the late prehistoric phases of eastern oklahoma and has recently been expanded to include the ozark highlands of arkan- sas and a portion of Missouri (Brown, 1984a; Kay et al., 1989; Perttula, 1989; Rogers, 1989). Following thoburn?s work, the spiro site remained untouched until 1933 when the Vandagriff (also spelled Vandigriff and Vandergriff) brothers dug a trench into the center of Brown Mound. in the summer of 1933, Joseph w. Balloun began hunting for relics in the southern por- tion of the ?great temple Mound,? marking the earliest recorded digging into the mound now known as the craig Mound (Figure 2.1). the northern portion of the craig mound containing the primary cone belonged to a dif- ferent property owner than the southern portion. it had belonged for years to aunt Rachel Brown, who allowed no one on the property, thereby effectively protecting the site. at her death the property passed to william craig, who also protected the mound, and it is from his family that the name derives. when he died, the land passed to two grandchildren who were still minors. their guardian, george Evans, held stewardship of the property in 1933 (clements, 1945:52). FIGURE 2.1. craig Mound prior to excavation, as photographed by Joseph thoburn in 1913. Photograph courtesy of the Research Division of the oklahoma historical society (thoburn collection 196.2.1). N U M B E R 4 9 ? 15 Later in 1933, six men formed a group, adopted the name Pocola Mining company, and approached george Evans to acquire a two- year lease to excavate in the north- ern section of the mound. one excavator was John o. hobbs, who lived near the craig Mound and farmed near there with his native american wife for many years. other relic hunters in this group included william and Kimball a. MacKenzie, a father and son from Le Flore county (cle- ments, 1945:54). the remaining excavators were contem- poraries of the younger MacKenzie and included glenn cooper of Booneville, arkansas; James Vandargriff of Paeola, arkansas; and R. w. wall (hamilton, 1952:25). they began work late that year. For nearly two years these excavators concentrated on the lower (southernmost) three mounds, presumably because desirable relics could be found at relatively shallow depths. Eventually, anthropolo- gists at the University of oklahoma, including Forest cle- ments and David Baerreis, were alerted to the destruction at the site. in 1934, clements prepared the national Park service survey form (BaE manuscript 4845 in the national anthropological archives) for the official recording of spiro or, as it was then called, scullyville Mounds. this re- port states that the Pocola or ?Poteau? Mining company leased the property. anthropologists, who were appalled by the destruction of the craig Mound, were eventually able to gain the support from the public and state legis- lators needed to initiate legislation to restrict excavation of archaeological sites in oklahoma to qualified research- ers only, and this restriction was supposed to take effect in 1935. Unfortunately, they were not in time to halt the destruction of the mortuary features at craig Mound. the relic miners desisted digging in July of 1935, when the licensing requirement took effect. however, the legisla- tion proved difficult to enforce, and clements, its major proponent, left the area to conduct research in califor- nia. the excavators soon returned, knowing that they had but little time to continue digging before either the state forced them to desist or their lease expired in november of that year. at that time they had scarcely explored the largest and potentially richest portion of the northernmost mound. in the late summer of 1935 the Pocola Mining company tried a time- saving tactic: tunneling. the men of the company had been miners, so tunneling was a natural alternative strategy. tunneling proved to be remarkably lucky and effective when these men struck a large mor- tuary feature replete with funerary offerings, including copper, textiles, wood, and a large number of engraved Busycon cups. the men dug several tunnels that met in the center of the mound, where they reported discover- ing a central chamber virtually brimming with artifacts. word spread quickly, and artifact dealers arrived to pur- chase materials on the spot. Materials continued to sell throughout 1936 and later, perhaps aided by a newspaper article by MacDonald (1935) that appeared in the Kansas City Star proclaiming ?a ?King tut? tomb in the arkansas Valley.? the Pocola Mining company finally stopped digging when their lease expired on 27 november 1935, but not be- fore a substantial inner portion of the mound had been de- stroyed. Furthermore, at the cessation of their digging, the looters used dynamite to collapse their tunnels. Figure 2.2 shows the mound after the commercial excavations ceased. although they may have been aware that damaged artifacts would not bring as high a price, the commercial excavators worked in a style that was not conducive to preservation. it became clear after later excavation that entire burials were contained in baskets and that a large amount of perishable textiles was present. an inestimable amount of textiles, engraved shells, and wood was prob- ably destroyed during the tunneling procedure. clements (1945:56) describes the scene when he returned from cali- fornia later in 1935. sections of cedar poles lay scattered on the ground, fragments of feather and fur textiles lit- tered the whole area; it was impossible to take a single step in hundreds of square yards around the ruined structure without scuffing broken pieces of pottery, sections of engraved shell, and beads of shell, stone, and bone. the site was abandoned; the diggers had completed their work. ProfeSSional eXcaVa tion at SPiro in 1936, the oklahoma historical society obtained the lease for the craig property at auction. clements began directing systematic excavation of the remainder of the damaged craig Mound with the aid of the wPa (Fig- ure 2.2). in two seasons, until november of 1937, Uni- versity of oklahoma and wPa personnel removed nearly the entire area of the mound that fell on craig property. the number of workers employed in the excavations was large, and the experience of the supervisors was limited. one result was confusing and scanty provenience and stratigraphic information on the numerous burials recov- ered from the mound. however, clements filed a wPa report (1935?1938, naa Report 4844) that contains in- formation regarding original archaeological context. the craig Mound was nearly leveled during these excavations. 1 6 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y wPa crews worked between 1936 and 1941 to ex- cavate the vestiges of mound remaining on the adjacent Brewer property as well as other mounds and house floors around the site. in 1941, world war ii brought an end to the wPa work before any significant site report could be published. a fair amount of publicity was accorded to the excavations at spiro due in part to the spectacular na- ture of the artifacts, which was enhanced by the excellent preservation. For example, a photo feature in The Daily Oklahoman (sunday, 13 December 1942) entitled ?soon- est sooners? focused on the wPa excavation and on cle- ments? and Baerreis? work at spiro. Publications on the craig Mound during the post- war period ensued and included Burnett?s (1945) study of materials that the Museum of the american indian, heye Foundation, had received and, in the same volume, cle- ments? ?historical sketch of the spiro Mound.? two arti- cles appeared in American Antiquity and would therefore have received a relatively wide circulation: orr?s (1946) treatment of the general archaeology and Bell?s (1947) paper on trade. hamilton (1952) produced a major pub- lication on the looted objects from spiro in a special issue of Missouri Archaeologist, to which James B. griffin also contributed. the wPa project turned up a large amount of ma- terial similar in style to and, in some cases, refitting to that recovered by the commercial relic hunters. From these excavations, it was clear that the craig Mound had con- tained a mortuary deposit unique in size and complexity. some burials along with considerable funerary offerings had been placed on litters made of cedar poles. other burials were made in baskets. although the wPa project resulted in the recovery of a large amount of material, in- cluding intact burials, the mining operation into the center of the mound clearly struck the densest deposit of mate- rial. near the end of the excavations an interested student named Robert E. Bell took a number of photographs of objects and the work of the relic hunters. Robert Bell was FIGURE 2.2. the north cone of the craig Mound during excavation by the wPa. Photograph taken by h. M. trowbridge on 6 september 1936 (trowbridge catalog in waldo wedel correspondence file, national anthropological archives, national Museum of natural history). N U M B E R 4 9 ? 17 later to become professor of anthropology at the Univer- sity of oklahoma and the principal scholar studying the prehistory of the state (Merriam and Merriam, 2004). ad- ditional details of craig Mound are described in chapter 1. after world war ii, the site slipped into relative ob- scurity, and no further excavation was attempted for 30 years. with the construction of the arkansas River navi- gation system in the late 1960s, spiro once more began to receive attention. Much of the area occupied by the site was scheduled for use as a borrow pit associated with the construction of a nearby lock and dam on the arkansas River. if not for the forethought of a few individuals, much of the site would probably have been lost. the state leased the property and began development as an archaeological park, which included reconstructing the craig and ward Mounds in 1970 and opening a visitor?s center in 1978 (Rogers, 1980:21). in 1970 the area of the craig Mound was excavated with the goal of correlating wPa maps with extant topography (wyckoff, 1970). Remnants of the craig Mound were located during this effort, and the fortuitous discovery of the remains of a cedar pole that had been recorded in wPa notes allowed for correlation with the old wPa grid. although a report on the engraved shells compiled by Duffield appeared in 1964, a comprehensive report of the craig Mound excavations conducted by the wPa was not published until the 1960s and 1970s when James a. Brown undertook that task of analyzing the artifacts and records from the site. his analysis was published by the stovall Museum of science and history and the University of oklahoma Research institute as a series of reports des- ignated as spiro studies (Brown, 1966a, 1966b, 1971b, 1976; Brown and Bell, 1964). Between 1979 and 1982 systematic research was once again undertaken at spiro (Rogers, 1980, 1982, 1983; Rogers et al., 1982, 1989). the principle objective of this work was to understand spiro?s organizational character- istics during the four chronological periods represented at the site and to relate these patterns to larger- scale trends in the arkansas Basin and ozark highlands. to fulfill this objective, test excavations were conducted in the ?village area? on the lowland portion of the site (Rogers, 1980). on the upland area, five previously unidentified mounds were tested, one of which was excavated extensively (Rog- ers, 1982). test excavations and a magnetometer survey were conducted in the so- called Plaza area (Bennett and weymouth, 1982; Rogers et al., 1982). subsequent to these excavations, additional nondestructive remote sens- ing research was conducted in 1994 (goodman et al., 1995) and 2006. a neW SYntheSiS in 1996, Brown combined and updated his original analyses into a major new two- volume study, The Spiro Ceremonial Center, bringing the early excavations at spiro into focus again, fully 60 years after the relic- hunting epi- sode that first brought fame to the craig Mound. Brown reworked the craig Mound chronology and synthesized a new picture of the great Mortuary, the burial feature at the center of the craig Mound, from which the smithson- ian collections originate. the only other person to publish specific informa- tion about the central mortuary feature was hamilton (1952). when hamilton attempted to recover informa- tion regarding the commercial excavations, he talked to members of the original excavation team. he published a description of the craig Mound deposit as John hobbs recalled it (hamilton, 1952:29?31). the interior of craig Mound was alleged to contain a hollow structure, and a diagram was published by hamilton (1952: pl. 5) and later by Brown (1996: fig. 1- 28). at the time of his first spiro studies in the 1960s, Brown largely dismissed the reports that there had been a hollow tomb at the center of the craig Mound. tombs would be uncharacteristic of any north american mortuary treatment. however, Brown (1996:85?94) reanalyzed the reports of an air- space at the center of craig Mound, and he now thinks that an airspace did exist, mostly as an artifact of mound construction and settling. the huge piles of artifacts could have settled beneath a system of cedar posts that leaned inward. the general lack of crushing of many objects sug- gests they were protected in a hollow void, although not an intentionally created void. Furthermore, the work of the relic hunters served to enlarge any airspace that may have been there, thereby making it seem larger and more discrete. Brown refers to the central portion of the craig Mound, with its well- preserved caches, as the great Mor- tuary. he has also been able to characterize the burials in the great Mortuary deposit, identifying litter burial heaped with shell cups, textiles, and shell beads. careful study of the grave lots associated with the great Mortuary and with other levels and contexts at spiro has enabled Brown to seriate the grave lots into a chronological se- quence. the great Mortuary burials constitute a period identified as spiro iVB, which dates to around ad 1400 (Brown, 1996). the way that the spiro iVB period fits into the overall sequence at spiro in terms of Brown?s (1996:133?134,161?164) chronology is summarized in Figure 1.2, with revisions from Rogers (2006). 1 8 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y the grave lot sequence will be referred to elsewhere in this report because it provides a way of associating the material from the commercial excavation with the mate- rial from the wPa excavations. some artifact types have more limited distribution, while others occur throughout the sequence. Brown discussed associations and distribu- tions in terms of his grave lot seriation. For the sake of comparability, the present text will also. acceSSion hiStorY the members of the Pocola Mining company, the relic hunters who first dug into the main mound at spiro, marketed their finds to collectors and relic deal- ers, many of whom visited the site. hamilton (1952) wrote that there were three dealers who purchased most of the artifacts directly from the miners: Joe Balloun, the first relic hunter at craig Mound; g. E. Pilquist of Dardanelle, arkansas; and J. g. Braecklein. these people, in turn, resold material to other dealers or col- lectors. word of the finds at spiro Mounds reached the smithsonian institution quite early. Jack Reed of Fayetteville, arkansas, traveled east with a number of artifacts that he showed to scholars. he apparently appeared at the smithsonian with Mr. McRoy and Mrs. Lake of Fortune and Time magazines on 18 May 1936 to exhibit 12 shell cups, shell gorgets, a small wooden mask with shell eyes and ear plugs, a string of pearls, textiles, and matting. Reed said, ?shells had been thrown out of trenches or burial chamber,? a space he described as having textile- covered walls and roof timbers (national Museum of natural history, 1960). Much of the material found in the commercial excavation made its way to serious collectors who amassed large collections from this operation. it was through these collectors, who eventually sold or donated their collections to the museum, that the national Museum of natural history (nMnh) accessioned materials from spiro (see table 3.1). there were 14 donations or sales of mate- rial from spiro to the smithsonian during the 50 years between 1936 and 1986. only Braecklein and trowbridge provide information regarding the original ar- chaeological context for some artifacts. information regarding collectors and their donations to the smithsonian is derived from the museum?s accession files housed primarily in the office of the Registrar of the nMnh along with supple- mental documentation from correspondence or other materials in the national anthropological archives. Early yEars: 1936?1958 Dealers purchased directly from the miners then sold the artifacts they acquired to relic collectors and other dealers. george Braecklein, one of the collectors to 3 spiro collection at the smithsonian April K. Sievert 2 0 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y witness the commercial excavation, was an architect with Braecklein, Besecke & swanson of Kansas city, Missouri. Braecklein is credited with gathering objects personally and later distributing them to other collectors. he appar- ently purchased some objects from John hobbs, an original member of the Pocola Mining company team. Braecklein was the first to bring materials from craig Mound to the attention of the smithsonian institution. in 1936, not long after the cessation of commercial excavation, Braecklein wrote to the smithsonian and then loaned a crown- type mace chipped from jasper (as it was then identified) and a catlinite pipe to the smithsonian so that the nMnh could make casts for the museum collection. the chipped mace was recorded as having been ?found in hematite bed, one end next to a cremation? (national Museum of natural history, 1936). Later in 1936 these pieces were returned to Braecklein, along with complimentary casts made at the nMnh. Ultimately, Braecklein transferred these and other artifacts to another collector, harry trowbridge, who later sold them to the nMnh. therefore, the catlinite pipe (423152) and the mace (423197) ultimately returned to the smithsonian. Many of the materials at the nMnh passed through the hands of george Braecklein. Braecklein also indicates that he personally picked up or found objects that came out of the mound, so in some cases Braecklein was the original collector. apparently, during December of 1935 and January of 1936, Braecklein continued to locate artifacts that he later distributed to collectors. several of the pieces in the collection bear Braeck- lein?s initials in pen and ink as well as brief descriptions of provenience. the early acquisition of plaster casts was followed soon thereafter by a larger donation of artifacts from Braecklein in 1937, when he sent over 90 items, comprising sherds, beads, and lithic artifacts. in 1938, Braecklein contacted Dr. setzler at the smithsonian and made another donation, this time including bone artifacts, beads, and pipe fragments. Between 1937 and 1958, the smithsonian acquired small lots of artifacts from spiro. these included a textile received in 1937 from D. i. Bushnell, Jr. in 1943, henry hamilton donated nine lots of textiles, including one piece of twilled matting with copper adhering, shredded fibers, colored cloth fragments, and other fragments of feathers, hair, and cloth. he had purchased these from various deal- ers. in 1950, Edna graham, the widow of collector Judge william graham, donated his relic collection. this collec- tion contained material from across the United states, in- cluding a few pieces from spiro?shell beads that graham had purchased from h. t. Daniel and copper pins that he received from g. E. Pilquist. accession number 206664 from the University of oklahoma in 1955 consists solely of a plaster cast of a shell recovered during the wPa excavation. the Lightner accession shown in table 3.1 was not an outright donation, but a loan, and these artifacts are no lon- ger at the smithsonian. however, circumstances surround- ing the loan are indicative of the way in which objects from spiro spread about the country. in 1956, clifford Evans, then an associate curator in the Division of archaeology at the smithsonian institution, happened into the Lightner Museum of hobbies in st. augustine, Florida. Evans no- ticed seven engraved shells on the bottom shelf in a case of native american exhibits. he recognized that these could well be from spiro, or at least they closely resembled the engraved shell cups found in abundance at spiro. Evans immediately inquired about the objects, their provenience or any catalog data that may have existed. cecil Zirkan, general Manager of the Lightner Museum, replied on 13 January 1956 that ?unfortunately, Mr. Lightner had died before the catalog of the collections were completed, and no data of any sort existed? (Zirkan, 1956). Zirkan suggested that the Lightner Museum send the shells to the nMnh for examination. this procedure was carried out, and the artifacts were repaired, refit, photographed, and judged to be excellent examples of shell engraving from spiro. these artifacts had not been reported or listed in any publications that discussed the shell engraving from spiro (such as hamilton, 1952). the nMnh and Lightner Museum agreed on arranging an TABLE 3.1. national Museum of natural history (nMnh) ac- cessions of material from spiro. NMNH accession number Year Donor 137937 1936 anthropological Laboratory, nMnh 145540 1937 J. g. Braecklein 145859 1937 D. i. Bushnell, Jr. 150758 1939 J. g. Braecklein 166197 1943 henry w. hamilton 187689 1950 william J. graham via M. Edna graham 206664 1955 University Museum, University of oklahoma 210052 1956 Lightner Museum of hobbies 217907 1958 Richard g. slattery 228126 1960 harry M. trowbridge 249362 1963 smithsonian institution 272249 1966 Richard K. Meyer 350434 1986 R. King harris 358176 1986 J. allen Eichenberger N U M B E R 4 9 ? 21 ?indefinite loan? of the materials to the smithsonian so that they could be included in an exhibit on north amer- ica. this meant that the nMnh could keep the shells indefinitely but that ownership was still vested with the Lightner Museum of hobbies and that at any time, the Lightner Museum could request that they be returned. in 1973, the Lightner Museum did request return of the shells to place them on display in a totally renovated museum. therefore, in 1974, four of the shell fragments (411904?411907) were returned to the Lightner Museum and are no longer in the collection at the nMnh. how- ever, three fragments (411901?411903) were on display at the smithsonian at the time and were not returned to Florida until the late 1980s, when the artifacts were re- moved from display. in 1958, Richard gates slattery of Rockville, Mary- land, donated a few spiroan objects as part of a larger collection of artifacts collected from Maryland, the Dis- trict of columbia, and 22 other states (accession 217907). slattery was described at the time as a ?knowledgeable amateur? and an officer in the Maryland archaeological society (unpublished, national anthropological archives Reference guide, p. 419). accession 249362 represents the accession of a cast made by the smithsonian of a blade originally collected by Braecklein. thE trowbridgE aCquisition one of the major collectors of artifacts from spiro was harry M. trowbridge of Bethel, Kansas. his col- lection consisted of 3,191 specimens, including some ex- tremely well preserved textiles. trowbridge amassed the collection between 1935 and 1952 from dealers and col- lectors who had, in turn, gotten them from the mound?s looters during the 1933?1935 commercial excavation. trowbridge?s collection was already legendary, and many of his artifacts had been illustrated in hamilton (1952) and Burnett (1945). trowbridge was keenly interested in the scientific value of his collection and was especially intrigued by the tex- tiles. he had made contact with the smithsonian through waldo wedel, then an assistant curator, as early as 1936, when he sent wedel a bit of yarn from a spiro textile. By 1937 trowbridge had sent textile samples to be analyzed for composition (whitford, 1941; willoughby, 1952). in 1938 he published a short report on an ?analysis of spiro Mound textiles? in American Antiquity (trowbridge, 1938:51?52 ). he also wrote a brief report of the results of fiber testing done at the national Bureau of standards in 1937 and published it in the Journal of the Illinois Ar- chaeological Society (trowbridge, 1944). according to a 1942 letter from trowbridge to wedel, trowbridge relays the information that much of his spiro material was at that time on display at the Kansas city Museum, Kan- sas city, Missouri (a fact also noted by hamilton, 1952). ?the museum offered the best room in the building, and built in addition, a fire- proof room for the spiro textiles, out of waste space opening off of the other room . . . in six large cases in the main room are most of the relics? (national Museum of natural history, 1960). wedel and trowbridge established a cordial and fa- miliar correspondence. when trowbridge?s collection was accessioned, wedel was the chair of the Department of anthropology, and he handled much of the administration of this large accession. on 14 May 1958, trowbridge wrote a letter to Rem- ington Kellogg, director of the U.s. national Museum, offering his collection for sale for $20,000. By 20 May a purchase order, number 38446, had been prepared, indicating how important this collection was deemed by the nMnh. however, it was not until early in 1960 that the collection was received and completely cataloged into the collection of the nMnh. although it is not pre- cisely clear from trowbridge?s correspondence, it seems as though his actions may have been prompted partly by a desire to see his collection in the hands of people who would appreciate and use it and perhaps partly from fi- nancial need. at the time of the transaction, trowbridge wrote the following passage in the preface to his catalog on 27 June 1958, and it describes well his attitude toward the collection. For twenty- two years he has felt the joy and the anxiety of a trustee, delegated by the fates to pre- serve these evidences of a people skilled in many arts. the mound material has been virtually a member of the family, the textiles having had in recent years their own light- proof and fireproof room, and the other specimens similar space ad- joining. he gives them up as one would an old friend, but tempered with the knowledge they will be under the expert and enduring custody of the basic research and education center of our country, ? the smithsonian institution. Unlike most of the other donors of artifacts from spiro, trowbridge kept records of his collection and maintained a catalog. he had assigned each artifact a number and recorded from whom he received the objects as well as any other information regarding the original 2 2 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y circumstances under which the artifacts were obtained. his records provide the only extant link to the original archaeological context for any materials in the nMnh?s collection derived from commercial excavations. in many cases, he recorded which of the original mem- bers of the Pocola Mining company excavated particular pieces, and he recorded from whom he purchased vari- ous pieces. his suppliers included Joseph w. Balloun, J. g. Braecklein, h. t. Daniel, albert h. hansen, John hobbs, Kenneth g. Mcwade (the only dealer not specifically men- tioned by hamilton, 1952), g. E. Pilquist, h. i. Player, and Lyle a. stephenson. in 1958, trowbridge prepared an abridgement of his own catalog and sent it to wedel. a series of letters written between trowbridge and wedel in 1965 describe drawings that trowbridge had loaned the museum but which ultimately had been returned to trowbridge. harry trowbridge died following a massive stroke in 1969 at the age of 81. he had been a serious amateur archaeologist throughout most of his life. he was born in 1888 in Blue Mound, Kansas, served in army intelligence during world war i, and graduated from the University of Kansas (Kansas City Star, 1 July 1969). trowbridge served as president of the Kansas city chapter of the Mis- souri archaeological society. after he retired, he devoted most of his time to establishing and acting as curator for the wyandotte county Museum in Kansas city, Kansas. this museum still retains trowbridge?s papers and, pre- sumably, trowbridge?s original documentation regarding the spiro collection (J. Brown, northwestern University, personal communication, 1992). it is apparent from reading his notes and letters that he treated the materials from spiro with great respect. Many of the textiles in his collection had been bundled originally, and trowbridge himself had painstakingly unrolled several of these. trowbridge then housed them under glass in wooden frames. although in hindsight this may not have been the best way to treat the objects, he was clearly motivated by a desire to save them from per- ishing. when it came time to ship them to the nMnh, he contacted wedel for clearance to ship the textiles standing up on end, rather than lying flat. trowbridge felt that the potential damage from slippage in the frames (although they appear to have been well constrained within their frames) was lower than potential damage should the glass break during transportation. trowbridge?s collection comprises roughly one half of the collection in the nMnh. the other major acquisition came from Richard K. Meyer of Peoria, illinois, in the late 1960s. thE mEyEr aCquisition Richard K. Meyer of Peoria, illinois, began collecting in the 1930s and by the 1960s had amassed an impres- sive collection that included artifacts of pottery, stone, shell, wood, bone, and copper. Meyer appreciated the aesthetic qualities of materials from spiro, and his collec- tion contained some of the more remarkable artistry from the craig Mound. For example, Meyer had collected rare wooden figurines, which were well preserved and remark- able for their detail. Engraved shell is particularly plentiful in his collection. Meyer did not catalog his collection as trowbridge had done, so there is no information pertaining to who his suppliers were or where in relation to the craig Mound certain artifacts might have been recovered. hamilton (1952) reports that Meyer was a student at the time of the excavation and that he traveled directly to the site to acquire materials. Meyer had at one time sent his collection to the na- tional textile Museum in washington for study. the act- ing head of anthropology for the national Museum of natural history, Richard B. woodbury, apparently heard of the collection through alan sawyer of the national textile Museum. woodbury?s first correspondence with Meyer was on 26 June 1965, when he wrote Meyer inquir- ing about the collection and expressing nMnh?s interest in acquiring it. Meyer was interested in donating the col- lection because it would amount to a significant charitable contribution with substantial tax advantages. that many items were at the national textile Museum proved to be extremely convenient because when the materials were do- nated to the smithsonian, many of them were already in washington. Meyer personally delivered other items on 23 May 1966. at that time Meyer was directing the Lakeview center Museum in Peoria, illinois. Meyer?s accession was rather complicated in that not all materials were listed as being donated in the same year. Meyer wanted to maximize the tax advantages of the large donation, so he requested an accession process that would allow approximately one- third of the materials to be for- mally accessioned in each of three years, 1966, 1967, and 1969. steven williams at the Peabody Museum, harvard University, appraised the materials in this collection for tax purposes. after this appraisal was rejected by Meyer as being too low, the collection was reappraised by andre Emmerich of the andre Emmerich gallery, new York city. george Metcalf inventoried the objects upon arrival at the nMnh, and Beth gibson implemented conserva- tion of some items. N U M B E R 4 9 ? 23 latEr aCCEssions More than a decade passed before the next donation of materials from spiro. in 1980 the large collection of Plains artifacts belonging to R. King harris of Dallas, texas, was sold to the smithsonian. although the mu- seum actually received the materials in 1981, they were not formally accessioned until 1986. when harris died in 1980, his collection and personal papers were also sent to the smithsonian. the national anthropological archives houses his extensive document collection, which includes both a biography of the man and a bibliography of his work. harris was a dedicated amateur born in 1912 who studied engineering at southern Methodist University until the advent of the great Depression cut short his col- lege career. he became deeply interested in texas archae- ology and conducted excavations across the state. harris was also a codeveloper of the caddoan conference. his biography indicates that materials were kept first in a china cabinet in his dining room and then in a museum in the back yard when the collection exceeded the size con- straints of the dining room. the materials that harris collected are the only frag- ments that are not from the commercial mining operation. harris had visited spiro Mound, probably during the late 1950s, and collected material from the surface. therefore, although it is certain these materials are from spiro, it is less certain where they came from at the site. all of these artifacts are weathered, supporting the information that they were indeed found on the surface. they comprise lithic tools (steeply retouched tools and points), potsherds, and engraved shell fragments. Presumably, the shell frag- ments derived from excavation back dirt, although this is not certain. these materials cannot be clearly associated with a mortuary context. the final acquisition is that of a set of casts made of artifacts from north america. Referred to as the Eichen- berger cast collection, it did contain examples of spiroan materials. PreViouS StuDieS of the SmithSonian materialS ConsErvation During the 1970s conservation scientist Joan s. gard- ner worked extensively with the collection. she was par- ticularly interested in textiles and worked with many of the pieces in the collection. she prepared cards containing photocopies of the original catalog cards and kept her own notes on conservation measures taken, identifications that she elicited, and stylistic and descriptive observa- tions. these cards were of inestimable aid in producing this report. in many cases, gardner?s records also include Polaroid photos of the objects that can be used for iden- tification and study. these records are maintained in the anthropology conservation Laboratory as part of the collections management records of the Department of anthropology. rEsEarCh although orr (1946) wrote a general treatment of spiro?s archaeology, hamilton?s report in 1952 provided a more extensive overview of archaeological materials from the site. hamilton was a serious avocational archaeologist who surveyed materials from all of the major collectors of spiroan artifacts around the Plains and Midwest. ham- ilton consulted with collectors and photographed pieces from their collections. Much of what he pictured and dis- cussed includes materials now at the smithsonian. clements published a report in 1945 of the history of the wPa excavation and submitted a summary report to the wPa. however, an in- depth report of the findings from the wPa work was not completed at that time. it was not until Brown tackled the materials in the 1960s that a comprehensive interpretive analysis of spiro came about. Brown?s work does refer to materials held in col- lections other than those from the wPa excavations, and in this context some of the materials in the smithsonian were referenced. studies of iconography have in some cases utilized de- sign elements on spiroan materials at nMnh. Fundaburk and Foreman (1957) compiled illustrations on Mississip- pian southern cult iconography, and their work also pro- vided a venue for discussing some of the materials in the collections of the smithsonian. shell in particular has been well documented. Duffield (1964) wrote an encyclopedia of designs on shell and uti- lized materials from the smithsonian. in the 1970s Philip Phillips of harvard University and James a. Brown under- took a massive project to correlate all engraved shell frag- ments by making rubbings, grouping designs by shape, and refitting portions of single shells by matching the tracings. in 1974 Joan gardner at the smithsonian joined this project and provided tracings of shells that had not been previously surveyed. this project culminated in a six volume work, later condensed into two tomes, which include basic descriptions of shell engraving technology, 2 4 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y illustrations of all shell fragments known, and interpre- tation of manufacturing and iconographic styles (Phillips and Brown, 1978, 1984). textiles were also of considerable interest. trow- bridge (1938) published early on the textiles, and his study was followed by a discussion by whitford (1941) regard- ing vegetal textile identification. King and gardner (1981) published a more intensive study of the textiles. Kuttruff (1988) produced a doctoral dissertation on Mississippian textiles and took samples from the nMnh?s collection for analysis. hamilton et al. (1974) analyzed copper materials from spiro, and their report gives a nice overview of the various patterns and styles present on embossed copper sheets. some of the materials in the nMnh collection are featured in that report. a concordance that references prior discussion or il- lustration of the materials in the collection of the nMnh is included in appendix B of this volume. this concor- dance should facilitate further research using the materials and enable the study of the iconography by referencing illustrations in Phillips and Brown (1978, 1984). the PreSent StuDY Materials are presented and discussed in a similar for- mat to that utilized by Brown (1976, 1996) in his analysis of spiro collections from the wPa excavation and other sources. this should maximize comparability of the data presented here, thus facilitating any attempts to correlate information from the two primary collections. however, different nomenclature is used in some instances. Brown employs the term ?sociotechnic? to refer to those items that are in some way grand versions of what would otherwise be everyday kinds of artifacts. i sometimes refer to such unusual, exaggerated, or ritual objects as being ?special- elite,? a word that subsumes both special social function and ceremonial meaning. otherwise, my categories for discussion are similar, although i do not make a clear dichotomy between utilitarian and special- elite artifacts based just on form. this distinction may be turbid because of the clearly ceremonial nature of this mortuary deposit. Even if objects appear to represent utilitarian items, they still may be imbued with ritual as- pects and therefore may be accorded special treatment or deposition. Phillips and Brown (1978:18) distinguish among five general functional contexts: (1) utilitarian equipment, (2) weaponry, (3) ornaments for the hair, head, or neck, (4) ritual objects such as pipes, shell cups, and figurines and (5) perishable containers and clothing. Brown?s 1976 report and his 1996 revision concerning the artifacts also reflect this general division as well as heeding broad raw material or manufacture categories. My approach also uses a combination of functional and raw material catego- ries. the first step in organizing this collection for study in- volved assigning each catalog entry to a raw material type. the seven types include (1) lithic, (2) ceramic, (3) textile, (4) bone, (5) shell, (6) metal, and (7) other artifacts. in many cases, however, the artifacts are composite and pos- sess elements from more than one material type. Material types also cut across functional categories. with a collection of the complexity of that from spiro, working with material type only as the primary grouping is unsatisfactory. Rather, i created a grouping based on a combination of function, raw material, and manufac- ture. artifacts are grouped analytically into (1) vessels of ceramics, shell, and stone, (2) fiber artifacts that include woven and braided objects along with basketry, (3) weap- onry and extractive tools, many of which are stone, (4) or- naments that include beads, pendants, gorgets, and objects of various raw materials, and (5) ritual and special objects, including pipes and other artifacts again of various materi- als. an overview of the collection is shown in table 3.2. in analyzing the collection, information was drawn from two sources. historiCal sourCEs Descriptive, conservation, and historical information regarding these artifacts derives from catalog and conser- vation records. included herein are any accession files and extra notes regarding provenience that might have been located. information from old descriptions or evaluations for the artifacts documents the history of interpretation of this collection. when trowbridge sold his collection to the smithsonian, he prepared a condensation of his own records as a catalog that he gave to waldo wedel. this catalog contains information regarding dealers from whom trowbridge bought artifacts. this document is available in the national anthropological archives in the waldo wedel personal correspondence file under ?trowbridge.? artifaCt analysis artifact analysis included examining, describing, mea- suring, and photographing the artifacts. the types of data included in the study comprise the following: N U M B E R 4 9 ? 25 1. Volumetric data with sizes and weights of artifacts allow comparison with materials described by Brown and others. 2. Detailed measurements may contribute to a better resolution in the chronology represented by mate- rial from craig Mound. some attributes may have chronological relevance. information such as sherd thickness can be used to differentiate some chrono- logical ceramic types. 3. stylistic information includes components of de- sign and iconography, which provide information about symbolic systems and can aid in both an assessment of meaning and in linking the objects from spiro to other archaeological contexts. 4. Raw materials can point toward trade relation- ships and intergroup interaction. 5. Manufacturing indicators suggest how craft pro- duction at spiro may have been organized. spiroan artifacts represent a high degree of workmanship and skill, and such craft working is still not com- pletely understood. it has been suggested that much of the material may have been traded in, as evi- denced by the raw materials utilized. Manufacturing details are helpful in sorting out some of the organi- zational and situational aspects of production. 6. Use and wear indicators contribute to interpreting how artifacts function within the contexts that en- gendered their use. there is still much we do not know about the specific nature of artifact usage in ceremonial situations, and often, a designation that an artifact was for ?social display? lacks depth. 7. Residues can indicate contexts for use. in other cases residues represent the sole vestige of the orig- inal archaeological depositional environment. there are three additional considerations to make in an analysis of this particular collection, and these stem from the nature of excavation, distribution, and acquisi- tion of artifacts by collectors. the first is the possibility of fraudulent artifacts being included in the collection. hamilton (1952) believed that the proportion of fake ar- tifacts in the collection would be small. First, there were relatively few people handling the materials at the site, and they were distributed very quickly. second, even the fine art from spiro did not bring high prices because the massive amounts of relics distributed from spiro increased the supply to the extent that they flooded the market for antiquities (hamilton, 1952). suspected fakes included copper pins and beads. hamilton also mentions some double- platform pipes being faked. the second consideration is the inclusion of artifacts that are not from spiro at all but are otherwise prehis- toric artifacts. collectors of relics were not always experts in north american archaeology, although many were. therefore, it would not have been unlikely for dealers to market non- spiroan artifacts as reputedly deriving from spiro. For example, there are some potsherds from the southwest in the collection (see chapter 4) that may not have come from spiro. a final consideration concerns artifacts from the craig Mound whose integrity has been compromised. in some cases, dealers may have sought to increase the chance for sale by reconstructing broken artifacts, thus rendering im- perfect objects ?whole? to consumers who put a premium on the perfect. Fillers, pigments, and adhesives may all have been put to this task. in addition, damaged portions of artifacts may have been rebuilt or even removed and then reconstructed. TABLE 3.2. summary of the spiro collection in the nMnh. Artifact count Artifact group and subgroup (whole and partial) containers ceramic 322 shell 907 other 4 textiles cloth bundles or pieces 107 Baskets 34 a cordage 260 weaponry and extractive tools small projectile points 166 Large projectile points 12 Large bifaces 15 sword-form bifaces 10 Maces 4 celts 15 other lithic tools 26 ornaments Earspools 211 shell beads 22,028 stone and copper beads 59 Pendants 112 gorgets 54 Ritual and special objects Pipes 33 copper 382 human effigies 9 wood 92 Bone 43 Total count 24,905 a includes 2 lots. 2 6 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y these three factors pose a challenge to analysis. since this report is designed to give a complete description of the collection, i paid considerable attention to recording details that may help to assess the effect of these factors. there are only a few human remains in the collection. these are not from discrete burials but represent bones picked up at random at the site. Bioarcheologist Javier Urcid, formerly of the Department of anthropology Re- patriation office in smithsonian?s national Museum of natural history, presents the analysis of these remains in chapter 9. throughout the discussion, specific objects are re- ferred to by catalog number. For this collection it became necessary to assign subnumbers to refer to specific items within larger lots. although each object could not be con- sidered separately (there are nearly 25,000 individual ob- jects, including beads), artifacts were divided into subunits that would be analytically meaningful. therefore, catalog numbers include my subnumbers, prefaced by ?s? in cases where there is more than one object or entry per catalog number. t he smithsonian collections from spiro include containers and con- tainer fragments of ceramic, shell, stone, wood, and basketry. Baskets and basket fragments were found throughout the craig Mound and will be included with the chapter 5 treatment of textiles. ceramic ves- sels include a few whole pots and numerous sherds. in far greater numbers are containers made from the shells of large whelks of the genus Busycon, a univalve that inhabits the gulf of Mexico. ceramicS stylE ceramics of the harlan and spiro phases have been well described (Bell, 1953; suhm et al., 1954; Brown, 1971b, 1996; Rogers, 1980 Bell, 1984; Fisch- beck et al., 1989). Descriptions of ceramics must account for the variability inherent in typologies, the descriptive criteria used, and the differing opinions of analysts. the typology used here follows Brown (1971b, 1996) and Rogers (1980). James a. Brown (northwestern University), James B. griffin (national Museum of natural history), and J. Daniel Rogers were also consulted regard- ing ceramic classification. Because ceramics from the wPa excavations at spiro had been classified by Brown as part of his work, an attempt was made to place sherds from the nMnh collection into the same classificatory scheme to main- tain comparability between these two collections. Variables shown below serve to characterize the regional ceramic styles. nomenclature and criteria used are summarized in Figure 4.1. Variables noted include the following: 1. Vessel form is the form of the vessel in respects to three different shapes, namely, bowl, jar, or bottle. subcategories describe the kind of bowl or jar, e.g., carinated. the presence of slip is an indicator that was useful in deciding vessel form on the basis of body sherds. according to J. a. Brown (personal communi- cation, 1992) jars were not slipped, whereas bowls were. therefore, for slipped body sherds, a designation of bowl is reasonable, especially if the slip occurs on 4 containers April K. Sievert 2 8 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y both faces. Bottles can have slip only on the outer surface, and bottle sherds will not be smooth on the inner surface. 2. Rim cross section is the shape of a cross section of the rim. 3. Lip form is the method of finishing off the lip, either flat or round, with variations shown in Figure 4.1. 4. Rim orientation indicates rims can be standing, in- verted, or everted. 5. Bases can be either flat or round. Flat bases may have margins that are rounded, defined, or stilted. Round bases may be quite spherical, in which there is no appre- ciable change in arc from the body to the base, or they FIGURE 4.1. ceramic attributes and key. after Brown (1971b:fig. 2; 1996:333). N U M B E R 4 9 ? 29 may be truncated, in which case there is a change in direc- tion for the body as it meets the base. 6. surface treatment possibilities include slipped, smoothed, burnished, and polished, with decorations that are incised, engraved, and painted. 7. color can be useful for distinguishing pastes and for analyzing firing. 8. Paste composition can be characterized according to texture: fine, medium fine, medium, medium coarse, and coarse. 9. the temper composition is really apparent temper because petrographic analysis has not been done. 10. wall thickness can be useful for distinguishing be- tween ceramic styles or vessel forms when the pastes are similar. 11. size is a useful indicator of vessel function and is relevant to an understanding of pottery manufacture. table 4.1 shows the number of sherds, weights, and mean wall thickness recorded for the various ceramic types. raw matErial Raw material is divided into two analytical compo- nents, paste and temper. in most cases the paste is fine- grained grayish clay. temper can include grog, grit, bone, and shell with inclusions of charcoal, sand, mica, and shale. Brown (personal communication, 1992) sees ceram- ics with sand in the paste as exotic. there were several sherds with sand in the paste, some of which are prob- ably truly from the spiro mounds. there are others with TABLE 4.1. sherd counts, measurements, and stylistic references for the various ceramic types. a dash (-) indicates that data are not available. Mean wall Total Number thickness weight Style and subgroup sherds (cm) (g) Stylistic reference(s) grog tempered williams Plain 23 1.14 825.2 Brown (1996:343); Rogers (1980:150) Leflore Plain 19 0.62 426.6 Brown (1996:346); Rogers (1980:153) smithport Plain 1 0.56 809.9 Brown (1996:348) Undifferentiated plain 34 0.71 865.4 Brown (1996) Undifferentiated incised 3 0.61 31.1 Brown (1996) crockett curvilinear incised 6 0.58 582.4 Brown (1996:358?359) Pennington Punctate 2 0.62 29.0 Brown (1996:360) Friendship Engraved 1 0.66 42.9 Brown (1996:376) white Engraved 1 0.74 - haley Engraved 1 0.56 74.3 Brown (1996:377) hatchel Engraved 1 0.64 109.3 slipped Poteau Plain 34 0.80 1,802.9 Brown (1996:405) sanders Engraved 6 0.77 431.7 Brown (1996:403) sanders Plain 109 0.77 2,121.9 Brown (1996:401?403); Rogers (1980:154) sanders Plain/old town Red 26 0.69 1,249.1 Brown (1971b, 1996:404) sanders Plain/clement Redware 2 0.61 196.5 sanders: Brown (1996); clement: Brown (pers. comm., 25 Mar 1992) Undesignated bowl 1 0.72 637.2 Undesignated slipped 4 0.78 36.0 shell tempered woodward Plain 20 0.80 770.4 Brown (1996:389); Rogers (1980:157) Effigy 2 0.65 92.0 Undifferentiated plain 4 0.56 41.4 Brown (1996) Undifferentiated incised 1 0.55 13.8 Brown (1996) other southwestern 6 0.56 225.0 Polychrome 1 0.78 33.0 Unclassified 5 0.55 266.7 glazed 1 0.72 21.0 Total sherds 314 3 0 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y extremely sandy pastes that appear to be from pots of the Puebloan southwest and therefore are unlikely to be from spiro at all. manufaCturE with the exception of a tiny pinch pot (448930), the ceramic vessels are built by coiling. walls are generally well smoothed, and most are burnished or even polished to a high sheen. slipped vessel surfaces are burnished as well. some of the pots are evidently quite large, as sug- gested by the low arc of curvature for some large body sherds. this is especially true of sanders Plain. handles and adornos are modeled. usE some pottery in the collection exhibits signs of use, indicated by residue on the inner surfaces of the sherds. the whole pots do not appear to have been heavily uti- lized. one bottle has a worn patch on one side, and a base contains green pigment. Utilitarian pots used intensively will tend to break preferentially over wares that are less intensively used. therefore, whole pots probably represent the less utilitarian among the ceramic wares. the collection contains six complete vessels and three more partial vessels. Vessels complete enough for vessel dimensions to be measured are listed in table 4.2. Ves- sel cross sections are shown in Figure 4.2. three of the complete vessels, the carinated bowl designated white Engraved, the gourd- form bowl, and the smithport Plain bottle are unique in that there are no sherds from these styles, just the complete vessels. there are numerous frag- ments from red oval bowls. (see the specific sections for sanders Plain and old town Red for a discussion of clas- sification of these oval bowls.) grog- , grit- , and bonE- tEmPErEd tyPEs these types were the most numerous and the most varied in style and decoration. they include plain, incised, and slipped wares. Williams Plain this is a relatively thick, grog- tempered ware. in three cases, the temper also contains grit, bone, and small amounts of shell. For the four sherds for which the rim orientation could be estimated, three are everted and one is standing. the lips are flat or rounded and are either rolled, thinned, or direct in cross section. color for williams Plain is highly variable, with hues ranging from light yellowish brown to very dark gray and black. included are reddish browns and various shades of brown and gray. the colors cluster around a hue of 7.5 YR in the Munsell soil color chart. one partial vessel is probably williams Plain (448648 and 448643). this vessel is a globular jar having a flat base and everted rim. it has the form of a restricted jar (see table 4.2) with an everted rim. the bowl is heavily coated with residue that appears to have been heated. along a broken edge are crystals of unknown composition. the temper in some cases appears to contain a small amount of shell or shell- tempered grog. TABLE 4.2. characteristics of complete vessels in the national Museum of natural history (nMnh) collection. a dash (-) indicates not applicable. Length Diameter Height Style Form (cm) (cm) (cm) crockett curvilinear incised cylindrical jar - 11.53 9.58 Undesignated spouted bowl 21.5 12.38 16.80 sanders Plain/old town Red oval globular bowl 16.8 11.87 7.69 sanders Plain/old town Red oval globular bowl 18.0 11.80 8.00 sanders Plain oval globular bowl 16.7 11.10 7.50 smithport Plain Bottle - 16.40 23.90 williams Plain Restricted jar - 8.76 8.37 white Engraved carinated bowl - 24.50 14.20 Undesignated Pinch pot - 2.33 1.83 N U M B E R 4 9 ? 31 Leflore Plain similar to williams Plain in paste, this ware is gener- ally thinner and the colors lean toward grays and black. surfaces tend to be burnished as opposed to merely smoothed. Vessel forms include bowls and jars. the five rim sherds demonstrate that the lip cross section is direct, and the orientation of the rim is either standing or everted. Lips can be either flat or rounded. temper contains a bit more grit, bone, and sand than does the williams Plain. Differentiation between williams and Leflore Plain wares were made on the basis of wall thickness and surface treat- ment. table 4.1 indicates that the mean thickness for Le- flore Plain is nearly half that recorded for williams Plain. Smithport Plain one complete bottle (423144) is present (see table 4.2). this bottle has a flat base, well- rounded spherical body, and a long neck that is missing the rim (Figure 4.3a). on one side there is a pitted patch approximately 7 cm in diameter, apparently caused by the vessel being knocked or rested repeatedly against a hard surface. the apparent temper is grit with some bone. Firing clouds are present on the body, which varies in shades of brown and gray (hue 10YR on the Munsell soil color chart). the inner sur- face is dark gray. the outer surface is smoothed but not burnished. James a. Brown (personal communication, 25 March 1992) suggested that this vessel is possibly an im- port from arkansas.1 Crockett Curvilinear Incised one complete vessel (Figure 4.4), a base and rim sherd from another vessel, two refitted rim sherds, and a base represent this type. these vessels are low- carinated cylin- drical bowls. Decoration includes incised and punctated design on burnished surfaces. the single complete vessel (423147) has a motif representing the apical end of a shell, and this motif is repeated four times (Figure 4.5a). these FIGURE 4.2. ceramic vessel profiles: a, williams Plain (448948); b?c, williams Plain, base and carinated body sherds from two different vessels; d, sanders Plain oval bowl (448641s03); e, white Engraved bowl (423148); f, coles creek gourd- form bowl (423145); g, crockett curvilinear incised bowl (423147); h, smithport Plain bottle (423144). 3 2 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y incised circular motifs are not connected but are separated by a biconcave shape thickly punctated within. the knobs that would naturally occur on the apex of a whelk shell are simulated with punctation. the incising is shallow, but the punctations are deep and done using a reed tool with a maximum diameter of 3 mm. the design was punctated as the maker held the bowl mouth up and inserted the tool obliquely (see rollout diagram Figure 4.5a). the rim is circled with horizontal incised lines. the jar appears to have been mended and reconstructed, perhaps by the dealer who sold it. tenacious black sooty residue is pres- ent on the inner surface. a different well- blended black residue covers mends and was likely applied by the dealer/ reconstructor. one fingerprint is evident in the clay. other examples of crockett curvilinear incised ceram- ics include two refitting pieces from a single bowl: rim sherd 448942s18 and base sherd 448644s01. this vessel is deco- rated with three horizontal lines around the rim, a curvilinear FIGURE 4.3. whole and partial vessels: a, smithport Plain bottle (423144); b, coles creek redware bowl with hatched decoration (423145); c, sanders Plain oval bowl (448641s03 or 423146); d, sanders Plain/old town Red oval bowl (448641s01). N U M B E R 4 9 ? 33 design, and punctations done with a stick. the surface is burnished. this vessel is notable because it contains a thick deposit of what appears to be glauconite (a greenish mineral) in the base. the green residue underlies a buff- colored silt- like sediment. a base sherd from a separate cylindrical bowl (448642s02) exhibits a similar greenish mineral deposit. the presence of mineral deposits such as glauconite, which are clearly nonedible materials, indicates that these cylindrical bowls were probably for something other than food pro- cessing. a ritual or craft use seems likely. Perhaps these were used for preparing pigments used in painting. Pennington Punctate this ware is represented by two refitted rim sherds from bowl 448642s17 (Figure 4.6). the type is indistin- guishable from crockett curvilinear in paste, but there is some difference in incised design, with Pennington incis- ing conforming to geometric and especially triangular ele- ments as opposed to the circular and spiral patterns seen on crockett. the example here carries design elements of both types. Diagonal elements laid out in a horizon- tal band around the rim are characteristic of Pennington Punctate. the concentric rings under this diagonal element are more characteristic of crockett. Punctations are made using a reed that is set nearly perpendicularly to the sur- face, leaving a well- rounded small circle. Friendship Engraved this ware is distinguished by having small raised ridges on the neck of the vessel and is represented by only one sherd (448642s19). one of these ridges is clearly visible on the right side of this body sherd from a bottle. the grit- tempered ware is black on the outer surface and pinkish on the inner surface. there is also a brown stain on the inner surface. tool and smoothing marks are visible on the inner surface. across the little diagnostic vertical ridge there are two fine hatches that form part of the overall design. the FIGURE 4.5. Rollout of design from two incised vessels: a, 423147, crockett curvilinear incised; b, 423148, possibly white Engraved. FIGURE 4.4. crockett curvilinear incised bowl (423147). 3 4 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y variety derives from southwest arkansas, probably the ouachita highland area (suhm et al., 1954:274). White Engraved this designation was given to a large simple- carinated bowl (423148) with a band of motifs formed around a cross or ?X? pattern along the upper outer edge (Figures 4.5b, 4.7). the paste is gray clay with a medium- fine texture, and it is tempered with grit, charcoal, and very fine shell. the bowl has been reconstructed using some form of glue or filler mixed with coarse grog that is plastered over breaks. the joins have become dry and brittle. the surface is burnished, incised, and punctated. the horizontal block design consists of large X?s in squares. the spaces between the X design and the square are filled either with hatching or tiny puncta- tions done with a stick. incised lines define each square, and there is an extra vertical incision between each motif. Faint rippled horizontal polishing toolmarks are present on the inside. this type is not listed by Brown (1971b, 1996) as coming from the wPa excavations at spiro. this type desig- nation is based on a suggestion from James a. Brown. there are no other sherds of this type. this bowl in particular may be relatively late and may resemble styles employed by na- tive americans after the arrival of Europeans. Hatchel Engraved hatchel Engraved is represented by one sherd from the body of a bottle (448642s15). it has heavy, probably culturally relevant, residue on the inner surface. the outer surface is engraved and burnished, whereas the inner sur- face displays the rough surface and toolmarks typically seen on bottles. the design is without the parallel lines that suhm et al. (1954:290) record as usually dividing the circular panels found on this type. Four parallel lines ring the neck, and one lower down defines the top of a decorated panel (Figure 4.8a). the area between circular motifs is filled with smaller circles, and the bands between motifs are filled with cross- hatching. temper is grit and grog, and the color is a very dark brown (10YR3/2). Haley Engraved one large body sherd has the geometric pattern of bands of narrow parallel lines characteristic of haley En- graved (suhm et al., 1954:284). the design has a verti- cal orientation overall. incised grooves are U- shaped and were probably executed using a finely pointed stick tool (Figure 4.8b). tiny spurs extend out from the outermost lines of the L- shaped notched bands, making these lines appear thorny. the temper is grog, and the sherd is very dark gray. this relatively late style shows similarity to the Braden shell engravings that include ticked concentric ovals. Brown (1971b:129) writes that haley is known from a post- great Mortuary burial as well as a great Mortuary burial and the general collection. the sherd shown in Figure 4.8b closely resembles the cylindrical bottle shown in Brown (1971b: fig. 18f; 1996: fig. 2?31f). FIGURE 4.6. Pennington Punctate rim sherds (448642). FIGURE 4.7. carinated bowl, possibly white Engraved (423148). N U M B E R 4 9 ? 35 Undifferentiated Grog- Tempered Plain this category includes sherds that were ambiguous or that did not fit into established types (n = 34). there is considerable variation in temper and paste within this group. two refitting sherds (448642s03), for example, contain a considerable amount of charcoal in the paste along with grog. some of the undifferentiated sherds dis- play residues. sherd 448642s09 is from a base and has a tenacious residue on the outer surface. also, 448642s11 is coated on the inner surface with a thick dark reddish brown material. although it could well be postdeposi- tional, this material could just as well be associated with use or treatment. within this undifferentiated group, pastes vary from medium coarse to medium fine in texture. two sherds (448642s02 and 448642s09) contain some sand. Most of the sherds in this group have surfaces that have been smoothed or, in the case of nine sherds, burnished. Undifferentiated Grog- Tempered Incised Undifferentiated incised sherds are less common (n = 3). there are two sherds with curvilinear incised lines (448646s03), and they appear to be from the same pot. another sherd (448642s20) has a dark gray paste of medium- fine texture and grog temper. the engraved de- sign consists of four parallel horizontal lines. this rim sherd from a jar exhibits a thinned and rounded rim with an inverted orientation. it has some shell residue on its badly eroded exterior surface. shEll- tEmPErEd warEs Woodward Plain woodward Plain is the most common type of shell- tempered ware in eastern oklahoma. it is a plain, heavily shell- tempered ware without slip. the surface is usually smoothed and occasionally burnished (but burnished to a lesser degree than are some of the other types). colors range from light brown to dark gray, with the prepon- derance of pieces in the gray category. Brown (1996:390) refers to woodward Plain as a regional variant of Missis- sippian Plain, the ubiquitous shell- tempered ceramics of the southern Mississippi Valley. the most common shapes are jars and bowls. the smithsonian collection includes 20 woodward Plain sherds, comprising 16 body sherds, 2 base sherds, and 2 rims. of the two rims, both are di- rect and standing. the lips are in one case rounded and the other case flat. Four of the sherds (515780s01) were recovered from the surface at spiro and are not directly associated with the commercial digging in the 1930s. Effigy two adornos or rim ornaments from pots are also present (Figure 4.9). one effigy is in the form of a human head with simple molded features. the face measures 6.6 cm long by 5.0 cm wide. the sherd was heavily tempered with shell that has since been badly leached. the other is an effigy handle from a negative painted vessel. this resembles a duck?s bill and is decorated in black pigment on a buff ware. the background is painted black, leaving a motif of unpainted dots. Perhaps this style associates with the style neelyville negative Painted, listed FIGURE 4.8. incised sherds: a, 448642, hatchel Engraved; b, 448642, haley Engraved. 3 6 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y by Brown (1971b) as occurring at spiro. this sherd is shell tempered, measures 8.25 cm long by 7.62 cm wide, and is 3.74 cm high. several styles of negative painted ceram- ics are recorded for Mississippian contexts in the lower ohio valley (hilgeman, 1985). hilgeman finds that nega- tive painted wares were produced between ad 1250 and 1450 and that a black- on- red slipped ware predates black on buff. the example from the spiro collection is black on buff. it was likely to have been imported because this type of decoration was used at relatively few areas, including sites on the ohio River such as angel and Kincaid, in the cairo lowlands, and near nashville (hilgeman, 1985:195). Undifferentiated Shell Tempered two contexts include unidentified plain shell- tempered ware (448642s12 and 448642s13). these are undesig- nated primarily because they are both badly burned, and charring may obscure details. another, 448642s20, is an incised and punctated sherd that is tempered with shell. sherd 448646s05 shows toolmarks on the inner sur- face and was perhaps from a bottle. the outer surface is smooth and black. sliPPEd warEs Poteau Plain Poteau paste is similar to that of woodward Plain, but the vessels are slipped. a number of sherds were slipped with deep red clay; others received a thin slip much lighter in color. For those with a light- colored slip, it was some- times difficult to identify the slip as being distinct from the paste. Undesignated Bowl however, one bowl (423145) is a vessel that mimics the shape of a gourd or possibly a shell vessel (Figures 4.2f, 4.3b). horizontal incising around the rim is reminiscent of the decoration often seen on coles creek incised ceram- ics (thorne and Broyles, 1968:110). the temper is grog, bone, and shell. the unusual characteristic here is a red slip that was applied to both the inner and outer surfaces. al- though coles creek, a shell- tempered ware, is not custom- arily slipped, Brown (1971b) lists one coles creek vessel as bearing a red slip. the shape of the pot is that of a spouted vessel, with an incurved rim opposite the spout. a similar, though not identical, form can be seen in Brown (1996: fig. 2?20a), who describes it as coles creek Polished Plain. the piece, however, could be considered to be of undesig- nated style. Both griffin (1952) and hamilton (1952:41) point to an analog in a vessel pictured by Moorehead in his work on cahokia (Moorehead, 2000: pl. xvii, fig. 6). the pot he shows from the sawmill Mound site is more globu- lar, and the spout has a clear break in the form at the point at which the spout extends outward. Sanders Plain sanders Plain is a relatively fine ware with a medium finely grained paste and grog temper. Many pieces are heavily slipped in red. the designation was given to all nondecorated red- slipped, grog- tempered body sherds in the smithsonian collection. sanders Plain is well made, has medium- fine texture, and is tempered with grog and grit that can include a smaller percentage of shell. in most of the sherds represented, the core is black. the color is often quite brilliant at 10R4/8 on the Munsell soil color chart. sanders Plain sherds make up the bulk of the ceramic col- lection. neck sherds indicate that at least two bottle forms are represented (448641s19 and 448645s01). Sanders Plain/Old Town Red thorne and Broyles (1968:78) list old town Red as a shell- tempered ware. Brown (1996:401) describes the ware as having either grog or shell temper. the ves- sel form represented is that of the oval globular bowl (see Brown, 1996: fig. 2?36b, for an identical vessel). Brown FIGURE 4.9. Effigy adornos: a, human head effigy adorno (448649); b, negative painted vessel handle (448643s08). N U M B E R 4 9 ? 37 differentiates between old town Red bowls of oval form and similar sanders Plain bowls depending on the mor- phology of the base (Brown, 1971b:180; 1996:404), with old town Red bowls having a rounded base and sand- ers Plain always having a flat base. the examples of oval bowls in the nMnh collection are not absolutely flat, nor are they truly rounded, a fact that emphasizes the problems inherent in creating and using ceramic typologies. James B. griffin, national Museum of natural history (personal communication, 1992) stipulated that old town Red has limestone in the temper?these bowls do not. there are at least five such vessels represented. three of them are complete enough to offer overall pot dimen- sions (table 4.2). there is some variation in the relative roundness of the base, and several bases could sit with- out rocking. there is one complete (reconstructed) vessel (423146). a dull red material has been applied over the original pot surface, and it is difficult to tell how much of the actual pot remains. the whole pot is from trow- bridge?s collection, whereas the fragmentary pots are from Meyer?s. one base was refitted from four separate sherds in Meyer?s collection, and the numerous other sherds might be refitted into more complete pots. this vessel type is the most numerous in this collection. the incidence of this type and many similar bowls is unlikely to be coincidental, and it is reasonable to speculate that they were part of a cache. although it is debatable whether they are best designated old town Red or sanders Plain, that they fall into such a narrowly defined pattern in form is not. the bowls are all very similar. some contain residues, notably patchy black films that can appear on inner or outer surfaces. Sanders Engraved the paste is identical to sanders Plain. Decoration most often consists of hatching along the rim. there are four rim sherds and two body sherds with portions of the incised rim intact (Figure 4.10a,c). Four sherds are engraved on the inner surfaces of the rim. these derive from vessels that have quite deep profiles (Figure 4.10a). the vessels have a very shallow curvature but still prob- ably fall within the range for sanders Engraved vessels. FIGURE 4.10. slipped and decorated rim sherds: a, sanders Engraved rim with engraving on inner surface; b, clement Redware (448641); c, sanders Engraved with engraving on outer surface. 3 8 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y they also have a black grog temper and paste identical to the other sanders sherds. two rim sherds (448642s12 and 448645s04) exhibit engraving on the outer rim sur- face and are from bowls that have standing rims (Figure 4.10c). the lip is direct and rounded both for the bowls with the engraving on the inner surface and for the bowls with engraving on the outer surfaces. Body sherds from these vessels that show no evidence of the engraving were collapsed into the sanders Plain category. Sanders Plain/Clement Redware there are only two sherds of this type (448641s16), characterized by a deep profile and scalloped everted rim (Figure 4.10b). they appear to be from the same vessel. the temper is a black grog mixed with grit and a small amount of shell similar to that found in some of the sand- ers sherds. the sherds are not from a carinated vessel form. the alternate designation as clement Redware de- rived from personal communication with James a. Brown (25 March 1992). othEr CEramiC PiECEs other ceramic elements include nodes, lugs, or han- dles. there is one strap handle from a highly polished red slipped vessel. the handle is 4.81 cm long by 3.44 cm wide, with a thickness of 1.89 cm, and consists of two refitted sherds (448647). an irregularly shaped, sand- tempered node or appliqu? that is 2.87 cm long (448645s05) ex- hibits indentations and a yellowish red color. a miniature pinch pot (448930) no bigger than a fingertip (2.33 cm in diameter and 1.83 cm high) completes the inventory of miscellaneous vessel parts. othEr warEs in addition to these wares, there are sherds that ap- pear to be from a far- removed archaeological context. in particular, there are six sherds from black- on- white painted vessels, white slipped and blackened sand- tempered wares (Figure 4.11). they are characteristic of southwestern ceramics styles. although it is possible that these sherds demonstrate interaction and trade with cul- tures of the american southwest, it is also possible that these sherds are not from spiro at all. in his study of over 15,000 sherds from the wPa excavations, Brown (1971b, 1996) does not record a single sherd of southwestern con- struction or decoration. Because of the loss of provenience for this material as it was passed from antiquities dealer to collector, it is possible that some artifacts were reputed to have come from spiro, when, in fact, they did not. all FIGURE 4.11. non- caddoan southwestern sand- and grit- tempered sherds with black- on- white design (423151). N U M B E R 4 9 ? 39 of the southwestern sherds were from the collection made by harry trowbridge, and he did not receive these sherds until six years after the commercial artifact mining ceased. the black- on- white sherds remain enigmatic. one polychrome sherd from a bowl rim is present. its source is unknown. the colors are red, orange, and black. this thinned rim has a flat lip and is inverted. the sherd is only about 5 cm wide, and relatively little of the design is visible. the decoration consists of a red rectilinear de- sign outlined in black against an orange field. the color of the sherd interior is light red, and the paste is gray. this sherd is of the same lot as the black- on- white material (423151s09). cultural affiliation for this piece is unknown. in addition, from the same lot (423151s10) there is a piece of glazed earthenware. the glaze is on the inner surface and along the rim on the outer surface. again, the provenience of this sherd is questionable. the final group recorded as unclassified consists of five sherds, two from Meyer?s collection (448644s03, a base sherd, and 448645s05, an appliqu?) and three from trowbridge?s (423151s08). the appliqu? piece has sand temper without grog. it is globular in shape and has one deep, one shallow, and two irregular indentations. the exterior surface and paste are yellowish red; the interior surface is very dark gray. the back or bottom appears broken, as if it had been attached to something, hence the designation as an appliqu?. the base sherd is black on the interior surface and in the core and very dark brown on the exterior surface. the surface has been burnished, and the paste is tempered with a mixture that includes grog, grit, and possibly bone. the three unclassified sherds from trowbridge?s collection include two rims and one body sherd of a smoothed, gray, sandy ware tempered with very fine grog. these two rims are everted. the three sherds are not typical of Mississippian ceramics from the region. Shell cuPS shell vessels make up the bulk of the spiro collection (>130 kg). a minimum of 330 shells made up of over 915 fragments is present. whole vessels had been given separate catalog numbers by the nMnh, whereas individual frag- ments were usually grouped into lots (e.g., Plates 1?11). Joan s. gardner examined shells in the spiro collec- tion in the mid- 1970s. she performed necessary conser- vation measures and made a considerable effort to match similar shell designs and refit shell fragments. this was done partially in the context of the shell iconography study per- formed by Phillips and Brown (1978, 1984). Even after the original tracings of shells were made for the Phillips and Brown work, gardner traced shell engravings at the smith- sonian and then forwarded those to Phillips, who was able to refit a considerable number of such designs. gardner refit- ted many shells together and researched motifs represented. Phillips and Brown (1978, 1984) discuss many of these shells in their work on the shell engravings and iconogra- phy. in some cases the various designs on the shells can be linked to designs on other media, such as pottery and cop- per (Brown and Rogers, 1989). Because the shell compo- nent of the spiro collection provides such a rich and varied corpus of iconography, it has also been the best studied. For the present study, shell fragments were grouped for analysis using several different criteria. if style could be easily determined, style was used; if style was uncer- tain but subject matter was specified, then subject matter was used (e.g., feet or feathers). if subject matter was not distinguishable, shell portion present was used (e.g., distal tips, apices). this applies especially to 448880, a lot of fragments containing 178 fragments that were combined into 44 different groups for analysis. Condition and ConsErvation condition of shells as discovered by Joan gardner is noted on her record cards. in many cases shells had yel- lowed because of the breakdown of old consolidants or conservatives. she removed excess polyvinyl acetate (PVa) with acetone that lightened the shells significantly, so that very few are found today that have yellowed. the shells from the trowbridge collection are in poorer condition than those from the Meyer collection. they are whiter, chalkier, and lighter in weight overall. (Refer to the section on columella pendants in chapter 7 for a brief discussion of possible causes for differences be- tween the Meyer and trowbridge collections.) the unengraved shells in the collection had not been given the attention that the engraved shells received. some of these were never treated, yet others have thick coats of a preservative that has yellowed (especially 448883). those that have never been treated are chalky, flaking, and fragile. the best- preserved interior residues are found on unengraved shells. in many cases this interior residue is thick and powdery. it is likely that decreased handling of unengraved shells has helped to preserve this residue. raw matErial the prevalent type of shell used is that of the si- nistrally whorled whelk, Busycon sinistrum (catalog 4 0 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y notations use the nomenclature B. contrarium conrad 1840 or B. perversum Linn? 1758). see Figure 4.12. the taxonomy of the genus Busycon has been reexamined for archaeological purposes (Pendergast, 1989:105) on the basis of a revised taxonomy developed by malacologists (turgeon et al., 1988). the name Busycon pulleyi hollis- ter 1958 now subsumes the classifications B. perversum Linn?, B. perversum pulleyi hollister 1958, and Fulgur perversa. this group includes the prickly whelk found in the gulf of Mexico from just east of the Mississippi Delta to the Yucatan. the name Busycon sinistrum hol- lister 1958 now refers to the lightning whelk, previously called Busycon contrarium conrad 1840. this mollusk ranges from new Jersey to the gulf of Mexico. in the gulf of Mexico, it blends with B. pulleyi. the lightning whelk ranges in length from 10 to 40 cm and is the shell primarily used for the spiro shell engravings. it appears that the left- handed whorl is the typical pattern for such whelks. several pieces of Pleuroploca gigantea (conch) are present (Figure 4.13). Cassis madagascarensis shells were utilized especially for gorgets, for which they were cut, the ends smoothed, and the edges incised with small, hatched lines. shells other than Busycon are rarely engraved, al- though there are fragments of Pleuroploca in the collec- tion that have been engraved (423288). rEsiduEs Basketry impressions are frequently visible on the outer surfaces of the shells (n = 28). in most cases the residue is located on the body of the piece, that is, on the under side when the shell is held as a vessel (Figure 4.14).this location indicates that shells were probably placed in the mound with the cup opening up. in one case the opposite is true; 448860 exhibits textile- based residue along the edges, as though the vessel were seated in a thick pile of fabric, with the shell?s design showing. on the interior surface of some of these shells are resi- dues that appear to have resulted from evaporating fluids. the result is a series of rings on the inside. Usually, the residue found in these shells resembles a silty sediment and is light brown or yellowish brown (Figure 4.15). in other cases darker brown stains are visible on both inner and sometimes outer surfaces. these stains appear to be the result of a natural taphonomic process operating on the shell. there is also a very different residue that is distinctly yellow in color. it usually stains the inner surface and is not found on the outer surface. where thin, it resembles a dull stain; where thicker, it appears grainy and in some FIGURE 4.12. Busycon or whelk shell, outer surface (542538). FIGURE 4.13. Engraved Pleuroploca or conch shell (423288). N U M B E R 4 9 ? 41 areas resinous. a different powdery brown residue occurs on some unengraved shells. cup 423269 is a Braden c style vessel with red grainy or crystalline residue on the inner surface. the surface is stained as though a fluid evaporated. this could be a cul- turally relevant residue. manufaCturE Phillips and Brown (1978) describe in depth the man- ufacture of the shell containers. in making a shell cup, the columella and a portion of the outer wall are removed (Figure 4.16). Usually, the right side of the shell (when viewed from the outer side with the apex positioned at the top) is left unworked. the shoulder is sometimes cut away at the opening. other times the shoulder is left intact and then cut to remove the columella. this occurs along the body wall approximately 1 cm from the shoulder. the surface is usually ground smooth. Most Busycon have some form of ridge or knobs occurring at the shoul- der. Knobs are usually removed, although in rare cases, these have been modified as a part of the decoration. in one case these knobs were incorporated into the design. FIGURE 4.14. Unengraved shell cups with exterior basket impres- sions (423232s02 and 423232s04). FIGURE 4.15. Residue on interior shell surface (423239). FIGURE 4.16. well- preserved cup showing form of finished artifact (542538). 4 2 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y human faces were engraved around the knobs such that the knobs became the noses for the faces. in some cases, toolmarks from grinding the surface can still be seen. For cups, the inner surface is never decorated; for gorgets the inner surface is often decorated. Engraving or incising appears to have been done using a sharp tool, probably lithic, given the V- shaped profile of the engraved lines. shell is a hard and brittle material and, unless fresh, is difficult to work. Many of the incised lines are typified by false starts and overcutting, typical results of working rounded surfaces using chipped-stone cutting tools (Figure 4.17). in most cases the designs are incised; however, in a few cases the designs have excised elements. it appears that a dark pigment was rubbed into the incised lines in order to make the design stand out. holes are often drilled into the distal and/or apical portions. these are consistently drilled from the outside through to the inside. on the apex, the hole is usually lo- cated in the center of the opening that has been cut to cre- ate the vessel. it is usually approximately 1 cm in toward the apex from the rim. More rarely, this hole is on the body wall between the shoulder and the rim. holes drilled through apical and distal end range in diameter from 0.3 to 0.7 cm. the mean is 0.47 cm, suggesting that medium to medium- large drills were used. the drills were proba- bly stone because the holes are either contracting or, more rarely, biconical. Furthermore, shell is a hard material and not easy to drill. it would require a durable drill bit. it is possible but by no means certain that abrasives were used in the drilling process. usE these cups appear to be well used. in many cases the interior surface exhibits fine, long striations as though the inner surface had been scoured. often, these striations underlie residue or sediment, testifying to the prehistoric origin of these damage lines. inner surfaces are naturally smooth, and there seems to have been little reason to scour the inner surface in the context of manufacture. on one Braden B fragment (448880s29) there is a curious arrangement of four tiny holes drilled around a larger worn hole (Figure 4.18). Perhaps these holes are the vestiges of an attempt to patch a cup that developed a leak. this arrangement implies reuse, but as of yet there is no way of knowing how long the use life of one of these ves- sels might be. they are breakable and might easily be dam- aged during use. Most of the shells, even nearly complete ones, lack the distal tip. Few unattached tips are present in the collection. it is possible that this type of breakage typi- fies use damage when a heavy vessel containing fluid was FIGURE 4.18. Drilled holes in shell apex perhaps indicating a repair (448880s29). FIGURE 4.17. Detail of incised surfaces showing toolmarks and overcutting (448858). N U M B E R 4 9 ? 43 held by the weakest part, the tapering handle. if such dam- age had occurred during use, it would follow that distal tips might be discarded elsewhere. however, the artifact miners could easily have overlooked distal shell fragments even had they been present in the craig mound. overall, the best indicator of use may be the residues remaining in the vessels. hamilton (1952) mentions that unengraved shell cups were reported to have held clay pigment. such a use is in- triguing and plausible. a spongy, powdery brown residue found on unengraved shells does appear claylike. another pigment, probably glauconite, is present in the collection from spiro, but none was noted on the inner surfaces of unengraved shells. instead, it is found on the interior sur- faces of ceramic vessels. taPhonomy Many shells have holes in varying sizes on one or more faces. Usually, these holes are accompanied by greater deterioration of the shell. these holes were caused either by intentionally punching holes in them (perhaps in an effort to ceremonially kill the objects) or by preferential deterioration of those areas that were in contact with the ground surface on which the shells rested (Figure 4.12). if the shells were simply broken, then there seems to be no reason why these areas would have incurred greater deterioration in areas adjacent to the holes. Most appear worn through, not punched through. if this is the case, the location of holes might be indicative of the positions and orientations that the shells may have assumed in the mound, i.e., resting on the rounded surface. this wear pattern corroborates information from basketry impres- sions that imply the shells sat with the openings facing up. worm action is also indicated. dEsign stylEs Each engraved shell was given the style designation assigned by Phillips and Brown (1978, 1984). there are six possible styles based on two major schools with three phases in each. the primary division is between the Braden and craig schools. these styles are distinguished accord- ing to the way in which the space available on the shell surface is used by the artist, the iconographic elements, and workmanship. several items in the nMnh collection are not pictured in Phillips and Brown, and these were as- signed to styles where possible (table 4.4). Braden style engraving differs from craig in the ex- ecution of the incising. the overall appearance of Braden a and B style engravings is one of greater control. there is little overcutting where lines that were meant to stop at a given boundary cross the boundary limit. craig style engraving is replete with this characteristic of execution. with Braden style engraving, areas to be filled or hatched were incised first, and then the outlines were done. the best way to achieve such control, based on my experi- ments with shell engraving, is to outline the area to be filled with a partially carved line. this line clearly marks a delimiter and provides a natural stop for a tool meeting the incision from another direction. then, the interior of the design element can be incised. Finally, the outline areas are finished and deepened, thereby eradicating many small overcuts made when the interior portions were carved up to the outline. with Braden style engraving, areas to be filled or hatched do appear to have been completed first, and then the outlines were done. with craig engraving, lines tend to be shallower and narrower, and there is much less excising. Excising is common with Braden style en- graving, especially with the earliest, Braden a. apparently, the Braden and craig styles are more or less contemporaneous. within each school the style of ex- ecution changes over time. Braden a and craig a styles contain the most detail. Designs are intricate and often very finely done. human figures are common among the mo- tifs. During Braden B and craig B, designs become bolder and less detailed. animal figures abound. Braden B is de- void of full- bodied humans. craig B offers a plethora of stylized animals, fantastic figures, and composite animal/ human creatures. During Braden c and craig c the styles become the simplest. highly stylized patterns, often poorly executed, are frequent. Braden c examples appear crude compared to Braden a. craig c designs are bold, large, and simple. they adhere closely to specific themes. craig c is usually sketchy, and the execution shows much overcutting and tool slippage. however, when a desired line was actu- ally cut, it was deepened and widened to the point where the slippage seems inconsequential or hardly noticeable. there is variability in the execution of incising shells that extends beyond style boundaries. sometimes there is variability on a single vessel, with some lines being deep and confident and others being sketchy. the cups present in the nMnh collection are subdi- vided into the groups shown in table 4.3. Maximum size values are included to give the upper size limit. the maxi- mum sizes for each group have been averaged to produce an average upper limit on size. Maximum values are given in order to emphasize just how large these artifacts are. the total minimum number of individuals (Mni) of 330 underestimates the total number of separate shell 4 4 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y vessels represented. some analytical units consist of shell fragments having similar designs. Each of these is counted simply as one vessel in the table 4.3. of the total Mni, 80 are unengraved, placing the estimate of engraved shells at 220. it is possible that there may still be matches among the shell fragments present. with the aid of Phillips and Brown (1978, 1984) a concerted effort was made to match fragments; however, it is possible that refits were overlooked. Braden A thirty- eight entries fall under Braden a. there are many loose fragments in this designation. Meticulous ex- ecution of design and greater detail typify Braden a shells. Usually, shell designs in Braden a are not oriented accord- ing to the axis or surface area presented by a shell. instead, motifs may be scattered at random across the surface. the shell apex is not treated as a separate plane. Rather, the design spills over onto the apex (Phillips and Brown, 1984:ix?x.) humans are shown in naturalistic propor- tions. often, the body is decorated with closely spaced hatching. accessories are plentiful and highly detailed. Beads are usually rendered as spherical. Bilobed arrow headdresses are common. some of the more frequent themes include intertwined snakes, a motif referred to as ?amphisbaena? by Phillips and Brown (1978) (Figure 4.19). some snakes are shown from above so that only the dorsal markings show, but most are viewed from the side and exhibit two distinct bands of design, representing the dorsal markings and un- derbelly of the snake. in one case the design spills over onto the columella stump. Using the stump as part of the design field is extremely rare and occurs only with the Braden a intertwined snake theme cups (448838). TABLE 4.3. characteristics of shell cups in the nMnh collection. Maximum Maximum Maximum Fragment Shell Weight length width height b Style count MNI a (g) (cm) (cm) (cm) Braden a 115 38 8,667.2 30.0 18.5 11.5 Braden B 63 35 14,182.7 35.2 20.2 13.0 Braden c 30 20 9,421.2 33.2 22.3 16.2 craig a 90 29 6,126.3 33.4 18.4 12.8 craig B 155 63 20,268.4 34.5 21.0 15.0 craig c 57 41 22,449.2 34.9 21.0 14.3 Unengraved 230 80 46,072.1 31.3 20.0 12.8 Unclassified 121 18 4,668.4 24.5 19.8 11.8 Various 46 6 2,543.1 16.0 10.7 - Total count or mean value 907 330 134,398.6 30.3 19.1 13.4 a Mni, minimum number of individuals. b a dash (-) indicates data nonexistent or not available. FIGURE 4.19. Braden a style entwined snake motif (448791). N U M B E R 4 9 ? 45 there are also motifs that are similar to pottery styles, primarily spiro Engraved (Figure 4.20). on these shells the entire surface is filled with spirals and curvilinear pat- terns. on another variety with geometric motifs, rayed concentric barred ovals and rectilinear designs referred to as Davis rectangles are present and are usually combined on the same shell. sometimes these are bordered by a saw- tooth pattern that runs along the rims. overall, the impression is of considerable detail and great skill in incising. the time spent in carving individual shells must have been considerable. careful incising has created smooth curves, finely rendered circles, and deep excision of some design elements. one of the more unusual motifs (448865) is a mor- tuary figure identified by his long, spindly, knobby, skel- etal legs. the mouth is shrunken in death. included in this figure is the only rendition of human genitalia seen on humans depicted on shells (Phillips and Brown, 1978: pl. 21). again, the detail on this cadaverous form is finely rendered. Braden B these designs consist primarily of animal themes and include some human heads. no full human figures are pres- ent. there are 33 separately analyzed examples. common iconographic motifs include broken maces, arrows, and bones. Disembodied human heads are frequent. snakes and intertwined snakes are still present, but the style is less controlled and more variable (Figure 4.21). Birds and composite animals are also present. some designs adhere to a horizontal layout. in addition, the apex begins to be treated as a separate plane and is incorporated into the overall design accordingly. cup 423257 shows two birds, one facing up, the other down (Figure 4.22). they have long beaks, and the feet are well detailed. a serpentine border design shows tails cross- ing at the distal tip. this is an example of orientation using so- called ?court card? symmetry after the type of symme- try found on playing cards (Phillips and Brown, 1978: pl. 89). this layout is found also in craig B designs. another interesting shell combines a number of mo- tifs. cup 448785 (referenced in Phillips and Brown, 1978: pl. 62, as 488880- B) exhibits human heads with forked and streaming eyes, bilobed arrows, ogees with stream- ers, maces, perforated pulley earspools, facial markings, FIGURE 4.20. Braden a style cup showing spiral designs similar to spiro Engraved ceramic styles (448827). FIGURE 4.21. Braden B snake motif (423249). 4 6 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y beaded forelocks, and a headdress. one mace is tied with a raccoon binding, and another has a tassel hanging from one of the protuberances on the top of its crown (Figure 4.23). Perhaps this is what these protuberances were de- signed for. Fiber or something adheres to the inner apical surface, and there are stains from fluid evaporation on the inner surface, but these have been scoured. at any rate, the motifs in Braden B designs are oriented toward weaponry and warfare. Unfinished shells found their way into the deposit as well. For example, 423263 is a relatively small Braden B cup that is unfinished. the outer shell surface exhibits many striations. a snake figure is visible from only one body arc. Braden C these are crude, simple patterns splayed out over the shell surfaces. Execution is typically very poor with con- siderable tool slippage, overcutting, and jerky rendition of curves. the number of examples is less common. there are 20 separate data entries represented. one Braden c cup (423259) has copper residue on the inner surface along the right edge. this cup had been rest- ing on copper, a rarity among these shells. instead, most seem to have been placed on basketry, away from copper, judging by the paucity of copper residue or staining on engraved shells. serpent designs are common, but the rendering is quite different from those done in Braden a style. cup 448809 shows the popular entwined snake theme, but the craftsmanship is sketchy and simple (Figure 4.24). there is one unfinished snake cup (448793). scenes suggesting warfare continue. cup 448809 is sketchily executed and almost appears unfinished. in the design a face hangs from a mace and has one eye outside on its forehead. Raccoon bindings are present on another mace. these are combined with snakes that are cross- hatched and intertwined. Feathers hang from both the FIGURE 4.22. Braden B bird motif with court card symmetry (423257). FIGURE 4.23. Mace, ogee with streamers, human head, and bilobed arrows on two sides of Braden B style shell (448785). FIGURE 4.24. snake and warfare motifs in Braden c style (448809). N U M B E R 4 9 ? 47 maces and the head. another cup in which a snake attacks a lizard (448830) may be metaphorical for warfare. scenes of death are present as well. cup 448864 is unusual because it depicts a skull with hair. the execu- tion is poor, with considerable overcutting and slippage. this shell exhibits the characteristic deep yellowish brown discoloration on the inner surface, as though some fluid sat here and altered the shell (Phillips and Brown, 1978: pl. 103). one cup that is not featured in Phillips and Brown (1978) is 423252 from the trowbridge collection. it may depict an armadillo and, if so, is the only such depiction known. the long- snouted animal appears alongside a bird in flight and displays four vertical lines on the body. these lines could easily represent the segmentation characteristic of armadillo markings. the four feet are shown as flag- shaped ?hooves.? armadillos would not be unexpected in eastern oklahoma. geometric patterns are not uncommon (Figure 4.25a,b). spiral patterns like those found on the exqui- sitely carved Braden a shells are also present, but again, they are more simply rendered, with shallower incising. Craig A craig a is the least numerous among the craig style shells. twenty- nine analytical units are present (cups or small lots of similar shell). craig a style cups exhibit a fair amount of detail. craig designs utilize the shape of the shell as a field for the artwork. human figures are ren- dered naturalistically. in one instance the shell is not merely incised, it is carved in low relief. this harkens back to the excision present with some Braden a forms. cup 423296 features horizontal registers of human heads. the surface is well polished and there are no whittle marks. Raised carved beading (convexo- cylindrical beads laid end to end) sepa- rate registers of right- facing heads. this includes an apex (423296) and (423297) distal fragment (Figure 4.26). the apex has shallower relief, and the heads are frontal view, rather than profile. Phillips and Brown (1984: pl. 159) ten- tatively place them together, and the trowbridge numbers are sequential. there are so few low- relief carvings that it would not be unlikely that they match. (For the pres- ent study, a fragment numbered 448846 was added to this context.) the heads so modeled present one of the rare representations of unperforated earspools among the de- pictions of people on the shell cups. other motifs include raccoon bindings, petaloid bor- der treatments that frame portrait heads, crosses, and avian themes. the birdman theme is prevalent. For example, 448825 is a shell with a very ornate and complicated design showing a figure wearing three very long vertically hang- ing plumes (Figure 4.27). one knee is raised to the figure?s right in dancing, and the arms swing to the figure?s left. the character is quite animated. the man wears an apron that has a square plate in the center. Lines extending out from the plate could be hair decorated with convexo- cylindrical FIGURE 4.25. spiral patterns executed in Braden c style: a, 448816; b, 423250. ba 4 8 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y beads. the part- human face has a beak. the person wears a beaded forelock, feather plumes, columella pendant in- scribed with apical banding, apron or scalp plaque below the waist, perforated pulley earspool, and two rolls about the waist. there is an orange organic residue inside the ves- sel. Manufacture is complex as well in that there is both incision and excision. Joan s. gardner (unpublished notes, 1970?1971, national Museum of natural history, Depart- ment of anthropology conservation files) observed that the hand is similar to that of the mortuary figure on cup 448865. one unusual cup is 448871s01. the subject matter here is a bison that has been shot (Figure 4.28). this de- sign is one of the few that refer directly to animals of the great Plains. another (448874s03) also depicts a bison; however, this time the head is shown in frontal view. Craig B the craig school offers up a remarkable bestiary. More of the shells (n = 63) fit into craig B than into any other style. common themes include hands and hand- in- eye mo- tifs arranged in horizontal registers, birdmen that occupy the FIGURE 4.26. Distal fragment of craig a style cup excised in low relief with incised decoration (423297). Depicts human heads ar- ranged in horizontal registers. FIGURE 4.27. craig a style carving showing birdman theme (448825). note beaded apron, forelock, avian characteristics, and leg raised in movement. FIGURE 4.28. Fragment of craig a style cup (448871s01) depicting a bison flank (in lower section) shot with an arrow. N U M B E R 4 9 ? 49 full plane, and composite animals having serpentine, avian, arachnid, and feline characteristics (Figures 4.29, 4.30). in- teresting things are done with symmetry, including the ar- rangement of four serpent- birds into a swastika formation. Raccoons achieve prominence and several fabulous animals have raccoon eyes. Raccoon bindings are shown on objects and on humans, and these are wrapped thrice around the object to which they are bound. certain other character- istics exclusive or nearly exclusive to craig B include the depiction of a double hair bun on the top of the head, a bent feather headdress, and two beads on the forelock. spiders are present in craig B but do not appear in the Braden style of any phase. Bird representations abound (Figure 4.31). Fish and other fabulous composite animals appear. other motifs include the pear- shaped, forked eye (in which one fork is truncated) and cross- in- circle (Figure 4.32). as in the Braden school, figures are simpler in craig B than in craig a. Figures such as those shown in Figure 4.33 are decorated with bead bracelets and anklets; however, the beads are rendered as grids. in manufacturing, some- times the apex was cut back so that the vessel could lie flat when turned upside down. Designs were usually oriented vertically on the vessel. Craig C craig c is well represented with at least 40 exam- ples. workmanship is bold and designs are simplified and stylized. throughout the craig school humans become quite stylized. By the time of craig c, humans have lost FIGURE 4.29. craig B style cup showing fabulous animal with com- plex forked- eye motif (448821). the animal appears to be drinking from a shell cup. FIGURE 4.30. Fabulous beast with serpentine, avian, and raccoon features (423236). FIGURE 4.31. craig B cup depicting avian characteristics (448787). 5 0 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y the pelvic region entirely and legs extend directly from a dropped belted waist. Prevalent themes of this style include the snake and talon motif and the ?men facing a forked pole? or ?busk? theme. there are multiple examples of very similar designs in each of these categories. the workmanship and style represented in the talon and snake motifs are so similar that it seems likely that the shells were produced by the same craftsperson. the snake and talon motif shows one snake on the apex and another on the body, the apical snake usually faces right while the lower one faces left (Figure 4.34). three eagle or hawk talons are fit between the two snakes, and one talon is shown on the distal por- tion of the shell, below the lower snake. there are six shells with this design. the forked pole theme is another common one (Fig- ure 4.35). here two men face one another, separated by a forked pole (or sometimes a serpent pole). they usually FIGURE 4.33. craig B cup showing man and snake themes (448858). FIGURE 4.32. craig B petaloid sun icons free floating across the field (423287). FIGURE 4.34. craig c style cup at two different angles showing snake and talon motif (448813). N U M B E R 4 9 ? 51 have large woodpecker heads extending from the front of their belts. there are 11 shells having designs that follow this theme. similar designs depict a man facing a serpent staff. Five shells fall into this group. there is remarkable consistency in the designs falling into the major themes in craig c. Birdmen and hand- in- eye motifs reoccur as in craig B, but the designs become simpler and bolder. Barred ovals are common. overall, the styles change from being highly involved and painstakingly executed to less carefully done and sim- pler. although most of the materials in the smithsonian?s collection were reported in Phillips and Brown (1978, 1984), several were not. these fall into all of the major styles and are presented in table 4.4. they include com- mon themes such as the snake and talon motif shown in cup 423253 (Figure 4.36). unEngravEd shElls the trowbridge collection contains 31 shells cataloged as number 423228. these shells were not FIGURE 4.35. craig c cup depicting two men facing a pole (423243). TABLE 4.4. Engraved shell fragments in the nMnh collection not reported in Phillips and Brown (1978, 1984). Catalog number Fragment type Style Motifs / notes 378246 gorget type 9 hatching 378262 gorget type 1 sun disc; possible refit with catalog number 448741 423238 Damaged cup craig c Birdman holding serpent staff; shown in hamilton (1952: pl. 92B) 423252 cup, distal portion Braden c Large bird in flight with armadillo-like animal; assigned Braden c because of poor missing execution and lack of detail 423253 apex and body craig c Rectangles, chevrons, woodpecker heads; complete except for large hole in body; snake and talon theme; may fit center in Phillips and Brown (1984: pl. 333c) 423277 Body Braden B Long bones 423291 Distal Braden B snake markings vessel 423294 gorget type 5 Plain with drilled holes 423343 gorget irregular Bone gorget with concentric circles and scalloped line 448740 gorget type 1 sun circle (aka Moundville circle) 448743 gorget Mask apical fragment that uses the shell apex as the nose for the otherwise incised facial feature 448744 gorget Unclassified Fenestrated; possible cross-in-circle design 448757 gorget type 2 Unclear decoration on inner surface 448822 cup craig c Paired figures facing a pole; similar to Phillips and Brown (1984: pl. 317) 448824 cup craig c Perhaps two figures with a pole; woodpecker head at waist 448839 2 distal craig c Man, mace, bird head belt 448840 apical craig B Pear-shaped motif floats about the apex; the rest is shown in Phillips and Brown (1984: pl. 267) 448843 Left distal Braden a concentric semicircles, possible Davis rectangle; hamilton (1952: pl. 137a) 448853 3 various craig B hands with eyes 448858s03 apical body craig c Bird man 448858s11 Fragments craig a, B snake man 448860 shoulder and apex Braden a Bilobed arrow, human head, forked eye; Phillips and Brown (1978: pl. 12) show body fragment, but large piece of shell has been added through refitting (continued) 5 2 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y specifically tagged with subnumbers because they can all be cross- linked by using their trowbridge number.1 Meyer?s collections also contained a large number of unengraved shells. these are grouped into lots numbered 448883 and 448728. the unengraved vessels show that while shell cups were often engraved, this was by no means standard. WooDen VeSSelS two fragments from wooden effigy bowls are present (Figures 4.37, 4.38). one represents a handle on which an animal effigy, in this case a feline, stands (448893). this handle was carved from one solid piece of cedar, as pictured in hamilton (1952: fig. 29B). toolmarks are very vague. the face is disproportionately large compared to the body. Four legs are realistically modeled with paws exhibiting four toes. a long tail is connected to the flat semielliptical handle. the face is turned slightly toward the left, and the mouth has been carved out. Mouth and snout are damaged. size is 12.4 cm long, 9.84 cm wide, and 12.8 cm high. Large TABLE 4.4. Continued Catalog number Fragment type Style Motifs / notes 448866s01 6 various Braden snakes, twined snakes 448867s02 apical fragment craig a May match Phillips and Brown (1984: plate 182); woodpeckers 448869s01, 02 6 various craig Feathers and forelocks; barred oval 448871s02 1 distal (448866) Braden B Broken arrow 448871s03 apical Braden a arrowheads, arrow fletching 448876s02 16 various Braden a one linked to Phillips and Brown (1978: pl. 37a), others are not; rayed concentric barred ovals 448876s04 10 various Braden a spiro engraved ceramic design 448877s06 11 various Braden a Davis rectangle and barred ovals 488880s05 1 distal, 1 apical craig a Eagle, bead grid 448880s08 9 various Braden a intertwined snake theme, concentric dotted circles; 4 found in Phillips and Brown (1978, 1984), 5 not 448880s13 3 distal Braden B Bi-triangular arrows 448880s14 apex and body craig B Barred ovals, snake 448880s16 Body craig c Birdman 448880s17 Body craig c Forked pole theme 448880s26 Body craig a one small fragment 448880s29 1 body Braden a or B amphisbaena 448880s34 39 various Unclassified Various (both schools) 448880s38 9 various Unclassified Exfoliated surfaces 448880s40 1 body Unclassified Bird or snake form 448880s42 10 various Unclassified human form, sash, snaky bits 448880s43 7 various Unclassified Poor condition, uninterpretable 448880s44 2 worked scales Unclassified worked alligator gar scales 448881 3 fragments Unclassified Brickwork pattern, hemicircle, arrow point 448882 shoulder Braden a amphisbaena 515786 1 body craig wings FIGURE 4.36. apical fragment of craig c snake and talon cup (423253). N U M B E R 4 9 ? 53 eye spots have been excised for inlays. the bowl sides have concentric rings incised below the handle. the second fragment is that of a bird head effigy fig- ure perched on the rim of the bowl (448894). the bird faces inward. it has disproportionately large shell disk eyes. Rectangular shell insets were glued along the crest of the head and down onto the beak. Four concentric rings are incised around the outside of the bowl. there is a trace of basketry matting on the crest (the fiber is 0.4 cm wide). Residue remains in the locations where shell insets are missing; perhaps this is aboriginal glue. Large eyes are made of inset concave shell discs (5.50 cm diameter aver- age). the face is narrow at 2.5 cm and along the width of the head is a crest formed by three insets of shell. the posterior inset (5.25 cm long) is missing. the middle is 4.9 cm and has the matting impression on it. the anterior inset is 2.8 cm long. the beak is 4 cm long and also had insets, but those are missing. the bird faces out from the bowl. this entire vessel from which the piece was broken would have been huge judging by the shallowness of the arc. there are a total of seven shell insets. the piece is also pictured in hamilton (1952: fig. 29a). Stone boWlS there are two stone vessels in the collection (Figures 4.39, 4.40). Both appear to be marble (or other calcare- ous stone) and are badly fragmented but have been refitted. the manufacturing technique is similar for both. an effigy figure is shown peering out from the rim on both bowls. on the opposite side is a tail- like handle. Vessel 423149 has three incised lines around a raised collared rim (similar to the wooden vessels). these lines dip down to become concentric half circles around the human effigy and the handle. the human head is only 3.00 cm wide and 7.05 cm high. it rises 4 cm above the rim. Features are modeled in relief, but they are very faint because of the surface erosion. in spots, the original surface is visible, and where so, it is highly polished. Ears and hair bun or ponytail are modeled on the sides and back of the head. Facial features include a low nose, what may have been an open mouth, and faint eyes. the neck protrudes from the side of the bowl like a figurehead on a ship. the vessel consists of 13 fragments refitted and held with epoxy. Dimensions are 21.2 cm long ? 19.0 cm wide ? 12.5 cm high. weight is 1,170.90 g. the other bowl (423150) weighs 870.40 g, is 21.2 cm long, 18.8 cm wide, and 11.4 cm high. Five concentric rings are incised horizontally around the raised collared rim. they also dip down to become semicircles around the FIGURE 4.38. wood bird head effigy perched on bowl rim (448894). FIGURE 4.37. wood animal effigy bowl handle (448893). 5 4 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y effigy snake head. toolmarks are obliterated by deteriora- tion. a section of wall has fallen out because of a weakened epoxy join. the effigy snake head is 2.65 cm wide, 6.36 cm high, and 5.35 cm deep. it is similar enough in design, ex- ecution, and raw material to have been made by the same artist that produced 423149. in each case the handles are flat shelves extruding out from the rim but not elevated above the rim. the incising simply follows the contours of the handle, without dipping down below the handle. this bowl consists of 35 fragments refitted and held with epoxy. other containerS other containers include baskets, bags, sacks, or pouches. Basketry, because it resembles textiles in manu- facture, is treated in the next chapter. other bags or con- tainers could be made from rawhide, leather, or textiles. chapter 8 includes a discussion of leather. Brown (1996) includes cedar pole litters as containers. although tech- nically these are containers, in this publication they are included along with other special ceremonial objects and discussed in chapter 8. note 1. a master list of equivalent numbers can be found in appen- dix a. FIGURE 4.39. Limestone bowl with human effigy adornment (423149). FIGURE 4.40. Limestone bowl with snake head adornment (423150). t he textiles from spiro are among some of the best preserved examples of native north american textiles outside the arid southwest. included are examples of woven fabrics, plaited strips, and cordage. there are also examples of matting and basketry. all of these types are grouped here in a chapter predicated on grouping artifacts of similar manufacture and raw material usage rather than on function per se. Uses for the textiles include bindings or wrappings, clothing, coverings, and containers. Most of the textiles derive from the trowbridge collection. trowbridge attempted to salvage the material and described his acquisition thus (trowbridge, 1938:52): on the whole, the material is in a remarkable state of preservation. the first specimens reached me about three weeks after they had been excavated, and were quite damp. it is thought this dampness is a fea- ture of recent years only, pits having been sunk by relic hunters and letting surface water into the central portion of the mound. Most of the fabrics appeared to have been folded when placed in the mound, but responded, in the case of the larger pieces, to careful manipulation. . . . these were pinned to fibre board and underwent a stretching process of weeks, the pins being moved outward a trifle each day. at the same time, quantities of lime, probably from shell, and perhaps from soft shell ornaments originally decorating the gar- ments, were removed bit by bit. some of the materials have been examined or analyzed at various times, usually for identifying fibers used in their construction. trowbridge (1938) was the first to publish on these textiles when he wrote up a report based on analysis performed at the national Bureau of standards on samples that he had sent to the smithsonian in 1937. in 1941, a. c. whitford at the american Museum of natural history reported on the identity of vegetable fibers that he examined (whitford, 1941:15), and his designations have been retained in the nMnh catalog records. hamilton (1952: pls. 140?152) includes plates of several of trowbridge?s textiles, including most of the large pieces. willoughby (1952) also wrote about the textiles, in a companion piece to hamilton?s study of the arti- facts. Burnett (1945) discussed spiro textiles primarily from the collection of the 5 Fiber artifacts April K. Sievert 5 6 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y Museum of the american indian, heye Foundation, now in the collections of the national Museum of the american indian. Baerreis (1947) produced a short article on the bas- ketry. Brown (1976) reported on textiles in his report on the materials from the wPa excavation. King and gardner (1981) produced an article reviewing the techniques and materials used in producing the textiles in the nMnh. in 1978, hoffman produced a Master?s thesis on the subject of the conservation of spiro textiles. the most in- depth analysis was that of Kuttruff (1988), who included spiro textiles in her sample for her study of caddoan textiles. Kuttruff?s work contains significant analysis of techniques, raw materials, and styles as well as a complete review of research done on textiles from spiro. she used 37 samples from textiles from the smithsonian collections. Rogers et al. (2002) analyzed 101 feather samples from numerous textile fragments to determine the bird species represented. the analysis presented here is primarily an inventory. none of the reports listed above has summarized the ex- tent of the nMnh collection or listed the contents of the collection by type. the collection comprises several large, nearly complete pieces, numerous fragments of all sizes, and odd lots of jumbled fibers. Many of the fragments may have derived from single artifacts; however, with time and handling, much of the association between fragments has now been lost. a brief discussion regarding numbering and cataloging is warranted to put trowbridge?s compli- cated collection in perspective. trowbridge housed his collection of textile fragments and relatively complete pieces in walnut boxes (Figure 5.1) or under glass in walnut frames and gave them all cata- log numbers. in most cases, he assigned separate numbers for separate items, provided they were stored as separate discrete objects, which applies to most of the larger textile fragments. smaller fragments were stored in boxes or cases and were numbered by him according to the case in which they were stored. For example, his artifacts numbered 2719 were all from one box. he differentiated among items by assigning alphabetical suffixes. Presumably, his numbers reflect best the original association among pieces. when these items were transferred to the smithsonian, museum catalog numbers were assigned. these reflect trowbridge?s original groupings. trowbridge?s cases with numerous frag- ments were each assigned a single number. For example, all pieces designated 2719 by trowbridge were given a catalog number of 423373. through time, the assignment of trow- bridge?s numbers has become muddled. For example, there are items marked 423373 (2719- F, h, L). in this case, the item is a single fragment of twined cloth. since it was known that trowbridge?s numbers referred to discrete objects, there is no way a single fragment could be from each of 2719- F, 2719- h, and 2719- L; rather, it must be from only one of these. Because of this problem, the relationship of such objects to other objects labeled 2719- F, h, or L cannot be reconstructed. trowbridge did, however, leave original photographs of the items in his collection. these include photographs taken of his boxes, and the objects within the boxes are labeled with their subletter. these photographs are in the trowbridge catalog among waldo wedel?s cor- respondence in the national anthropological archives. after coming to the national Museum of natural his- tory, most textile pieces were conserved and placed in safe mounts that facilitated examination. During the 1970s conservator Joan s. gardner placed many of these materi- als between sheets of Plexiglas. all but two large pieces and several smaller fragments were so conserved. Fragments similar in style were combined within the same subnumber context. subgroups were designated within catalog num- bers on the basis of storage status. some of the lots with large numbers of pieces were stored in Riker boxes. Be- cause of the fragility of these materials, artifacts so housed were not handled for this study, and subnumbers were not FIGURE 5.1. the contents of one of trowbridge?s boxes of textile fragments (423372). N U M B E R 4 9 ? 57 written on the boxes. at most, the lid was removed for a better view. During the 1990s, conservation of the textiles was further upgraded with new storage boxes with glass lids. Materials can now be observed without handling. any object of interwoven or plaited fibers was desig- nated as a textile. in many cases, a single catalog context will contain both textiles and other materials such as shell beads or copper fragments. still, there is a limited number of object types, which include the following: 1. cordage is spun or unspun fibers plaited or twisted into a yarn, rope, or braid. 2. cloth is textile made from fibers spun into cordage and then interwoven to form a sheet. the term is used in- terchangeably with ?fabric,? a term employed by Kuttruff (1988) to refer to the same sort of artifact. 3. Basketry is material made from strips of vegetable material, plaited, coiled, or interlaced to form a container or sheet. in some cases the objects were classified as mats if it was clear that the original piece had been flat. 4. sample refers to bits of textiles used as samples. Many of these are prepared as microscope slides. the above categories comprise the larger divisions among the textile groups. note that basketry is included here along with other fiber constructions. the similar processes of manufacture link these artifacts even though the functions of the artifacts vary. cordage either acted as the first step in the construction of woven textiles or was the desired product and was used for tying and hold- ing objects and stringing beads. cloth may have been used for garments, including capes or kilts, or for blankets. Bas- ketry is used to make mats or containers into which other burial accessories were placed. textile styles are listed in table 5.1. the count column refers to the number of catalog contexts for which the style is recorded. By multiplying the number of contexts by the count per context, the minimum number of fragments in the category is derived. sizes of the largest members of the category are included as well. raW ma terialS an array of plant and animal fibers was utilized (see table 5.3 on page 72 for a summary of identifications). Fur and feathers were spun into yarns. the predominant animal hair used was rabbit. in his 1977 analysis of spiro textiles, Michael a. Bogan (national Museum of natural history, Department of anthropology, smithsonian insti- tution, unpublished manuscript) suggested that the most likely species included Sylvilagus floridanus (eastern cot- tontail), Sylvilagus aquaticus (swamp rabbit), and Lepus californicus (black- tailed jackrabbit). Feathers were ac- quired primarily from turkeys and also from geese. Rogers et al. (2003) examined 101 textile samples from the col- lection microscopically in order to determine more specifi- cally which species of birds are represented. of this sample, which yielded 85 examples of feathers, 77.6% were tur- key (Meleagris gallopavo), 17.6% were goose (Branta canadensis), and 4.7 were trumpeter swan (Cygnus, cf. buccinator) (Rogers et al., 2002:245). these authors point out that although raptorial birds and woodpeckers are present in artistic representations, their feathers do not appear to have been used in textile production. Vegetable fibers include cane and pawpaw fibers (whitford, 1941) as TABLE 5.1. styles of textiles in the national Museum of natural history (nMnh) collection. Minimum Maximum Maximum fragment length width Object Style Occurrences count (cm) (cm) cordage Braid 13 53 41 18 Yarns 17 47 17 18 wrapped basketry 3 7 11 4 cloth twined tapestry 36 76 28 33 spaced weft twining 24 70 61 139 wrapped warp twining 24 113 31 64 simple weave 1 1 45 148 Total 118 367 5 8 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y well as numerous unidentified fibers. Plant fibers include very few seed hairs (Kuttruff, 1988). corDaGe For cordage, style was assigned on the basis of ply and twist direction. therefore, the two major classifica- tions of cordage were Z- twist and s- twist. Ply was simply designated as one- ply, two- ply, and so on. cordage was encountered in the context of the textile collection and in combination with artifacts that are stored separately from the textiles. in addition to braided or twisted cords, i also included a type of weaving in which basketry fibers, such as cane or grasses, are wrapped with yarn. yarn Yarn refers to spun or unspun fibers that are twisted into a usable length (Figure 5.2). Usually, the yarns will be composed of more than one ply, most often two but oc- casionally three or more. Yarns with more than two plies seem to have been preferred for stringing beads and are not common in woven cloth. Yarns will be the compo- nents of other textile constructions. on the other hand, they may also be utilized as for stringing beads or possibly sewing. Yarns are manufactured from both vegetable and animal fibers. Yarns vary greatly in diameter. the finest are less than 1 mm in width, while the average is 0.3 cm. when pos- sible, ply number and twist directions were recorded. For the yarn category, the fragment count is underestimated. Much of the material in this category consists of loose yarns from unraveled cloth. in cases for which there were several groupings of yarn fragments present in a context, the number of separate groups of yarn was counted, rather than the number of strands within each group. there are a number of pieces of yarn or cordage uti- lized for tying objects together or stringing beads. the composition of these yarns varies. in some cases, the yarns or cords appear to be of animal hair; in others they are vegetable. no yarns made of feathers appear to have been manufactured into cordage that would be used for fasten- ing or stringing. the only example of cordage demon- strably used for tying is artifact 448939s01.2, a piece of two- ply, Z- twist cordage used for making a bundle of four mammal leg bones (Figure 8.22). it is a thick yarn of a stiff vegetable fiber. it is glossy and well preserved, suggesting that some preservative has been used on it. Beads are strung on cordage in several cases. Lot 448714 contains four examples of cordage as part of a group of tubular copper beads and copper pin fragments. the cordage was used to string the beads and may have been well preserved because of its contact with the cop- per. in one instance (448714s06.2) the cordage consists of a two- ply, s- twist string on which two copper beads re- main strung. a thicker gauge Z- twist yarn (448714s07.2) manufactured by braiding three strands of a two- ply cord together retains three intact tubular copper beads (Figure 7.21a). there are also two loose fragments of string in this context (448714s08 and 448714s09). these consist of a three- ply s- twist cord and a narrow, fine, Z- twist two- ply string, respectively. the same context contains tubular copper beads that retain the original cordage inside the center bore but for which the characteristics of the cord cannot be determined (Figure 7.21b). another composite artifact (448934) is a large rect- angular basket filled with beads, many of which are still strung on the original cord (although in most cases this cord has decomposed). the cord in this case is of vegetable fiber. composite artifact 423381 (Figure 5.3) is a piece of wood wrapped with two- ply Z- twist cordage. FIGURE 5.2. Yarns (423375). N U M B E R 4 9 ? 59 braid Braid is manufactured of several plies that are plaited or interlaced together. Plaiting is an easy way to create ropes or cords of different sizes that can then be used in multiple ways. Plaiting can be accomplished to produce wider cord than that produced by twisting, and some of these may have been used for headbands, belts, or ties. Mean length is 5.1 cm for braided cordage in the collec- tion. one knotted piece of braid (423384) adheres to a piece of copper (Figure 5.4). at least six strands of two- ply s- twist yarn are plaited into a rope. there is a second type of cord consisting of a thicker Z- twist two- ply yarn that extends out from under the knot on the copper sheet. the fiber appears to be hair. some braids were often quite colorful. Planned use of red, black, and light brown yarns yielded a braid with a chevron design (Brown, 1976: fig. 68c). Braiding was a technique also used in finishing off garment edges. in this case, hanks of loose warps or wefts were left long, then plaited into a braid about 3 cm wide, and then divided and plaited, usually into four smaller braids, thus effec- tively creating a tassel. other braids are quite thick. Figure 5.5 shows many strands of two- ply Z- twist yarn combined into a thick rope (423382). wraPPEd baskEt fibEr this category presents a different concept in creating cordage. strips of cane or other vegetable material (like those used in making matting or basketry) are wrapped with medium- fine yarns (423372s03, 423373s22). Usually the finished cordage is about 1 cm in diameter. it would have been flexible when fresh, but stiffer than other braids produced solely of spun fibers. the yarns used in wrap- ping were invariably dyed. Red, black, and a natural light brown are evident. it is possible that what appears now to be light brown was once a brighter yellow. combining these resulted in cordage that was colored in alternating bands having high contrast, resembling the black, red, and yellow markings on king or coral snakes. FIGURE 5.3. cord- wrapped paddle (423381). FIGURE 5.4. Braid (423384). 6 0 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y CLOTH stylEs styles were assigned according to how the cloth was woven. Methods of weaving present in this and other spiro collections are outlined by King and gardner (1981:132) and include plain weft twining, spaced weft twining, twined tapestry, oblique twining, wrapping, oblique in- terlacing, and plain weave. twining is the most common and involves interlocking pairs of weft yarns around the warps. in twining, the warps may be single, paired, or paired alternating (Brown, 1976; Kuttruff, 1988). Fur- thermore, the wefts can be wrapped, resulting in an over- all s- or Z- twist. Fabrics can be constructed using one or both of the twisting directions. in some cases, such as the tapestries, the direction varies from row to row, resulting in a countered weft twining. Much of the cloth is fragmentary; however, there are eight large pieces that represent unrolled bundles. these appear to be the major portions of actual garments. they are referred to as mantles and skirts in the collection re- cords. one (423353) has ties on either end, which suggests that it was indeed tied about the body, possibly about the waist or neck (Figure 5.6). Large fabrics without ties may have been draped over the back (Brown, 1996:620).FIGURE 5.5. Braided rope (423382). FIGURE 5.6. twined and resist- dyed cloth with ties (423353). N U M B E R 4 9 ? 61 sPaCEd wEft twining Medium- gauge yarns are used in making a spaced weft twined cloth. there are seven bundles of spaced weft cloth that have not been unfurled (Figures 5.7, 5.8, 5.9). the remainders, which have been unfurled, demonstrate that several different cloth types are manufactured using this technique. in one type, the warps consist of a rela- tively wide gauge yarn approximately 3 mm in diameter. wefts are sometimes missing in this type, leaving skeletal FIGURE 5.7. Bundles of spaced weft twining: a, 423361; b, 423360, wefts decomposed; c, 423362, wefts decomposed. 6 2 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y warps. the appearance now is that of large hanks of fringe, composed solely of the well- preserved warps as in bundle 423361 (Figure 5.7a). these are the so- called fringed skirts reported by willoughby (1952). it is reasonable to agree with Kuttruff?s communication to James Brown (1996:621) that these were not intentionally fringed?they only appear that way because the wefts are missing. this cloth is most often made from what appears to be rabbit hair and dyed red. the red has faded to a muted reddish brown. the missing wefts have, however, left impressions on the warps. the wefts appear to have been a completely different yarn from the warp, very fine, and of a material that has decomposed. the spacing between wefts averages 1 cm, meaning that the overall weave was fairly loose. another type of spaced weft twining involves warps similar to those described above, but they are twined using weft of similar composition and weight to the warp. Five of the large bundles unrolled by trowbridge fit this category: 423353 (Figure 5.10), 423354 (Figure 5.11, 5.12), 423355, 423356 (Figure 5.13), 423357 (Figure 5.14). in these cases both warps and wefts were preserved. spacing between wefts was less than 1 cm. in the case of large framed textile 423353 (Figure 5.6), the wefts are spaced 0.6 cm apart and are made from similar yarn to the warp. this particular cloth exhibits three fringed selvages. the pattern of twin- ing for 423353 calls for alternate paired wefts. a third type of spaced weft twining involves a com- pletely different yarn used to make an unusually fine small weave (423379). the yarn is fine, approximately 1 mm in FIGURE 5.9. Bundle of cloth (423366), spaced weft twining. FIGURE 5.8. Bundle (423364), spaced weft twining. N U M B E R 4 9 ? 63 FIGURE 5.10. spaced weft twining (423353) showing resist dye pattern. FIGURE 5.12. spaced weft twining (423354) close- up. FIGURE 5.11. spaced weft twining (423354). 6 4 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y diameter, and appears to be vegetable (Figure 5.15). the warp and weft are of the same fiber. King and gardner (1981:126) cite whitford?s report that this fiber may be Amsonia ciliata, dogbane or bluestar. weft spacing is ap- proximately 3 to 4 per centimeter. this produces a tight weave. the border is close, forming the selvages. wraPPEd warPs For some cloth the warps are spun and twisted from vegetable or other fiber, and then this cordage is wrapped with dyed feather strips. the effect is a strikingly soft and lustrous fabric. these fabrics were dyed in black, red, or yellow. the wefts are in most cases missing, but they appear to have been fine yarns worked in a spaced weft twining. the wefts are invisible, as they are hid- den beneath the tendrils of the feather from the wrapped warps. these fabrics were bundled, as were the other twined textiles. Bundles of this type tend to be smaller and more fragile than the twined cloth constructed of hair. there are four bundles that have not been opened. FIGURE 5.14. spaced weft twining (423357). FIGURE 5.13. spaced weft twining (423356). N U M B E R 4 9 ? 65 twinEd taPEstry tapestry refers to twined fabrics with tightly packed, paired wefts. twining with single colors was done in patches or blocks, which followed a pattern to create an overall design. these fabrics are usually made from fine- gauge predyed yarns. Yarns of contrasting colors are used to provide outlines (usually black, but sometimes red). the most intricate fabric designs known from spiro are produced using this method (Figures 5.16, 5.17, 5.18). in twined tapestry, the black outlining wefts are usually of thicker yarn than the inner areas. this thread does seem to outlast the others and often extends beyond the lim- its of the preserved mesh. in some cases, traces of a pat- tern outline are present, but the yarn used to fill in the space is largely missing. the colors found in this type of weaving are strong and, when originally produced, would have been brilliant. when samples of tapestry wefts from fragments of textile number 386188 were analyzed at the smithsonian?s conservation analytical Laboratory (caL, now renamed the Museum conservation institute) in 1976, they were found to consist entirely of feather and contained no mammal hair or vegetable material.2 Down, most likely from turkeys or waterfowl, was spun into this type of yarn. Kuttruff (1988:206) reports that rabbit hair was used for some of the twined tapestry weaving, including the large fragment showing human faces with forked- eye motifs (423373). the yarns used in the tapestry twining are quite fine, approximately 1.5 mm in diameter. although some tapestries were worked in three colors, red, black, and light brown (referred to as tan in the litera- ture), in some, a fourth color, brown, is indicated as well. there is considerable variation in shade. in the tapestry cloths, the warps have often decom- posed, leaving skeletal wefts. therefore, warps must have been of a different material entirely. Kuttruff (1988:206) reports that for some, unspun strips of vegetable material were used for the warps. simPlE wEavE this weave is a simple over one and under one weft weave. it is rare but does appear in one large decorated fabric piece (423358s01) shown in Figure 5.19. although the weave is different from the spaced weft twining, the yarns and colors used appear to be similar. FIGURE 5.15. spaced weft twining, tight weave (423379). FIGURE 5.16. twined tapestry (423373s24). 6 6 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y fabriC dECoration For twined tapestry textiles, braids, and feather- wrapped fabrics, the yarns appear to have been colored prior to weaving. Kuttruff (1988:117) noted that as many as six different colors may have been used in textiles from spiro. Dyeing was presumably done using traditional dyes, such as black walnut and madder. the other major method of decoration was the creation of patterns using a resist dyeing method (batik is one such method). the tech- nique is clearly depicted on several of the large unrolled textiles from the trowbridge collection (Figures 5.6, 5.10, 5.19, 5.20). these textiles are twined, with the exception of the one simple weave cloth. Patterns are placed around the borders and comprise concentric hemicircles (423355), ogee, and key motifs (423353). Usually, the fabric is a soft shade of red or dusky red, while the design is now a light brown (tan or buff) color. the designs are most intricate among twined tapes- tries because the method of weaving using different col- ored yarns allows creating patches of different color with greater facility. the twined tapestry motifs include geo- metric and human facial motifs (Figure 5.18). Many frag- ments are small, and therefore, it is difficult to estimate the range of variation among motifs employed. bEads and tExtilEs a note from trowbridge (national Museum of natu- ral history, 1960) says ?all the beads in tray 2721 were exposed when the entire textile collection was being cleaned and unfolded.? therefore, these could be from any textiles in the entire collection, and there is no way to match them with the textile with which they may have been associated originally, nor is it known whether they were attached to the cloth in some way or simply depos- ited in association. FIGURE 5.17. twined tapestry (423373s25). FIGURE 5.18. twined tapestry showing human face (423373s27). N U M B E R 4 9 ? 67 funCtion and usE the majority of the textiles were bundled. there is no evidence to suggest that textiles were placed unbundled into the great Mortuary. trowbridge unfurled eight large textiles during the late 1930s and placed them between glass in walnut frames.3 Joan gardner remounted six of these. several of the large fragments are still bundled. two textiles that were on exhibit during gardner?s project were not remounted (423359 and 423355) and as of 1992 were still within trowbridge?s original walnut frames. baSKetrY the term ?basketry? is used to refer to any object in- terwoven of relatively wide strips of vegetable material. Fibers used in basketry include cane or reed. there is one complete basket, several fragments of baskets, and a con- siderable number of basket impressions on other pieces. Baskets can be distinguished between twilled and woven baskets and wicker. wicker differs from other basketry in that it involves the use of a support structure around which the baskets are made. there was very little wicker present in or suggested by the collection. in at least one case a mat contained a support structure, and this support is of similar fiber as the matting it supports. in most cases it is difficult to tell whether a basketry fragment derived from a shaped basket container or from FIGURE 5.20. close- up of simple weave cloth with resist dye deco- ration (423358). FIGURE 5.19. simple weave cloth with resist dye decoration (423358). 6 8 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y a flat mat. since many baskets were rectangular, no curva- ture would be expected for their fragments. often, the frag- ments are small, and although weave can be reconstructed, overall size and shape cannot. selvage edges are rare. Manufacturing is usually done in a twill pattern, al- though there are examples of coiled baskets. twill plait- ing is done by lacing wefts over and under more than one warp and then offsetting the next weft to create a herring- bone pattern. Brown (1976, 1996) uses a simple method for characterizing woven patterns by counting the num- ber of warps that a weft crosses over, then the number it crosses under. For example, if the weft crosses over two and under two, it can be denoted 2- 2. this is similar to the way in which Baerreis (1947) refers to baskets. the width of fibers used in basketry ranges between 0.3 and 0.57 cm. the average width is 0.43 cm, with a standard deviation of 0.09 cm for fiber width. the total number of discrete basketry fragments or lots of fragments is 34. the incidence of basketry is recorded in table 5.2. squarE shallow baskEts the single complete specimen is a large flat basket lined with leather (Figure 5.21). it contains beads, copper fragments, and shell fragments. this artifact (448934) is unique in that it retains much of its contents (hamilton, 1952). the basket is well preserved and measures 54 cm long by 29 cm wide by approximately 10 cm deep. the basket has collapsed in on itself, but the selvage or bas- ket rim is present on three sides. one side is damaged. a second basket fragment with intact edge sits on top (the lid), covering some of the beads within. this second frag- ment is not shown in the illustration in hamilton. this suggests that it was replaced on top during rehousing at some point. there is no reason to doubt that this fragment does belong to this artifact. the size is in the same range as that reported for a group of twelve similar flat rectangu- lar baskets in the collection of the University of arkansas (Brown, 1996:414; horton and sabo, 2011). the weave is a twill done with fibers of 0.3 cm diam- eter in a 3- 3 pattern. some of the material found within the basket has become separated and is stored in small boxes alongside the basket. the basket is lined with fabric TABLE 5.2. characteristics of basketry in the nMnh collection. a dash (-) indicates that data are nonexistent or not available. Catalog Subnumber Length Width number (if any) Object Style (cm) (cm) Count Comments 423388 Basketry twill 6.5 4.5 3 Fragment 423387 Basketry twill 13 5 1 selvage fragment 448778 s01.2 Basketry twill 8.1 3 2 Fragments 386185 s01 Basketry twill 15.7 4.43 1 Rim fragment 423386 Basketry twill 2.5 1.5 1 Fragments 423224 s01.2 Basketry twill 7 6.6 5 Fragments 448922 Basketry twill 75 11.5 2 Mat selvage fragment 448934 s01.1 Basketry twill 54 29 1 complete 386186 s01 Basketry twill 17 12 1 Fragment 448914 Basketry Loose fibers - - 1 lot Fragments 448927 s01.2 Basketry twill 7.35 4.13 1 Fragment 448934 s07 Basketry Unclassified 25 1 12 Fragments 448934 s05 Basketry Unclassified - - 1 lot Fragments 423389 Basketry coil 7 7 1 Basket bottom 423390 Basketry coil 20 6.5 1 Fragment FIGURE 5.21. square shallow basket, 54 ? 29 cm (448934). N U M B E R 4 9 ? 69 that supports numerous strings of shell disc beads. this cloth is 41 cm wide by at least 8 cm long. it is difficult to determine the original form of the cloth because the fabric is wadded up and smashed under the weight of the other objects that were in the basket. along one edge, running along the basket selvage, is a braided tassel, which is really the only part for which a clear style can be seen. over 500 shell beads repose on the fabric. Many of these are still strung on a vegetable yarn. in addition, there are approxi- mately 1650 more beads that have become detached and are stored in a separate box. also in separate boxes, but presumably from the same basket, are three medium- sized spherical beads with straight bores, one piece of engraved shell, loose basketry fibers, and approximately 60 frag- ments of sheet copper weighing 100 g. CoilEd baskEts there are two examples of coiled baskets in the col- lection (Figure 5.22). these baskets were manufactured by stitching together a coiled structural fiber using a nar- rower, thinner fiber. one example is the center or bottom of a coiled basket (423389). the other is a piece from the side of a coiled basket or mat (423390). For this piece, the center or beginning of the coil is not present. Both are from the trowbridge collection. coiled baskets are un- usual at spiro, according to Baerreis (1947), who discusses the basketry present in the University of oklahoma collec- tion and in the McDannald collection. twillEd matting or baskEts twilled matting refers to basketry done in charac- teristic twill plaiting but without the curvature expected for some containers. Most pieces are too small to exhibit curvature, so much of what appears to be matting may actually be small fragments from containers. Rectangular shallow baskets will not exhibit curvature, even though they are containers. Baerreis (1947:7) discusses the inci- dence of matting as a common artifact among historic FIGURE 5.22. coiled baskets: a, 423389; b, c, 423390. 7 0 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y groups of the southeast, such as the natchez, chitima- cha, and taensa. the largest piece of basketry that may be matting in the nMnh collection is a long selvage edge fragment, 75 cm long and almost 12 cm wide (448922). it consists of two fragments laid end to end (Figure 5.23) and is a large matlike creation consisting of three layers of twill work. this piece from the selvage and shows clearly how one layer was folded over and reworked to finish off the edge. a long wide piece of woody cane runs along the piece between the layers. this probably served for sup- port. the twill is a large weave; however, the twill for both pieces does not match. the larger piece is plaited in a pat- tern of 7- 3, and the smaller piece is 6- 2. Brown (1996:414) cautions that for those artifacts that are demonstrably mats, the twill pattern is always a regular weave of 1- 1. therefore, the twill in this case better resembles that of a basket. Both fragments are three layers thick. Edges are frayed more on the smaller piece. other examples of twilled basketry are present in much smaller fragments. some adhere to copper (Figure 5.24a). the twill weave can be used to produce variation and hence regular patterns in the overall design (Figure 5.24b). oPEn- work wEavE another type of woven mat is a very loose weave, with as much as 2 cm between fibers in the basketry. this is evi- dent in two cases: one is a small bit of basketry adhering to copper plating; the other is an impression only. the latter is an impression on a large piece of leather. the leather ap- parently rested on the matting and conformed to the mat surface (448934s08). although this object has the same catalog number as the large shallow basket containing strung shell beads, it is uncertain whether or not this truly came from that artifact. twinEd matting there is one example of twined matting made from wide yarns of vegetable bast fibers (448913). Kuttruff (1988:207) reports that this fragment was not dyed and was woven in a spaced twined mesh using single warps. she reports six yarns per square inch. the entire piece is over 15 cm long and 10 cm wide. baskEtry imPrEssion Many artifacts exhibit impressions of basketry. no- where is this more evident than with the engraved shells. thirty shell cups carry basket impressions. on the shells the basket impressions appear as patterned staining on the surface. they occur in small patches or cover a large portion of the shell surface. the impressions are almost invariably on the rounded surface of the shell, indicating that the shells were placed aperture up in the mound. Bas- ket impressions are seen on clay and stone objects too. FIGURE 5.23. Large mat or basket fragments (448922). N U M B E R 4 9 ? 71 at least three limestone earspools exhibit clear basket im- prints, as does a figurine made of galena. fiber iDentifica tion several attempts have been made to identify fibers used to manufacture these textiles and baskets. the first person to have fibers analyzed was trowbridge. he sent materi- als to the smithsonian, which had them analyzed by the national Bureau of standards (now the national institute of standards and technology) in 1937. analytical results of trowbridge?s nBs inquiry were published the following year (trowbridge 1938). several samples (B, c, D, F, g, J, K, and L) were reported to be rabbit hair, with a remote possibility of them being from a related animal, such as either squirrel or rat. two samples (D and g) were rabbit hair wrapped around a vegetable material warp that might have been wood or bark. two other samples (a and E) were feathers. sample a was composed of feathers with strips of skin wrapped around a two- ply vegetable material thread. a more extensive analysis of feathers used in textiles was FIGURE 5.24. twill basketry: a, curved basket section (386185); b, twill matting (432385). 7 2 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y conducted by Rogers et al. (2003). Finally, sample h was identified as hair, possibly ?buffalo or bear, or perhaps dog or horsehair? (trowbridge 1938:53). we now know that the sample could not have been horsehair, as the archaeo- logical context dates at least two hundred years before the spanish introduction of horses to north america. whitford studied vegetable fibers in 1941. Joan gard- ner elicited identifications from the conservation analyti- cal Laboratory and from Michael a. Bogan, who produced a short summary of his findings. supplemental notations on Bogan?s report were made by Roxie Laybourne, an ex- pert on feathers. the same slides were also examined by Douglas Deedrick of the FBi, who agreed with the identi- fications (M. a. Bogan, unpublished; see table 5.3, foot- note b). Kuttruff (1988) examined fibers as part of her research. textile identifications made during these earlier studies are summarized in table 5.3. Rogers et al. (2002:246) published a report on the mi- croscopic study of 106 feather samples used in a number of textile fragments from the nMnh collection. Most preva- lent were wild turkey (Meleagris gallopavo), canada goose (Branta Canadensis), and swan (Cygnus sp.), in that order. Feathers from raptorial birds appear not to have been used in constructing textiles (Rogers et al., 2002:249). noteS 2. see caL reports 2089, 2090, 2091 (smithsonian institu- tion, Museum conservation institute, unpublished). 3. object 423352 was never located (a large fabric in Plexi- glas). a temporary specimen removal card for 423352 dated 1989 and signed by natalie Firnhaber indicates the destination was smithsonian?s Museum support center, where the object is probably located at a120305003. TABLE 5.3. Fiber identifications for craig Mound textiles. Catalog number a Identification Source 423372 (2718K) Nolina georgiana (yucca) whitford (1941) 378240 Rabbit and unidentified Bogan (unpublished) b 386188 Feather caL report 2089 c 423353 Rabbit Bogan (unpublished) b 423353 Fur fiber Kuttruff (1988:205) 423361 Rabbit and canid? Bogan (unpublished) b 423365 (2717c) Asimina tribola (pawpaw) whitford (1941) 423366 (2717E) Asimina tribola (pawpaw) whitford (1941) 423367 (2717h) Asimina tribola (pawpaw) and Arundinaria tecta (cane) whitford (1941) 423367 Rabbit and rodent? Bogan (unpublished) b 423368 Feathers (goose and turkey) Bogan (unpublished) b 423368 Vegetable fiber caL report 2090 c 423370 turkey feather, rabbit Bogan (unpublished) b 423371 (2722D, i) Arundinaria tecta (canebrake) whitford (1941) 423372 (2718K) Asimina tribola (pawpaw) 423372 Rabbit and some turkey Bogan (unpublished) b 423373 warp, vegetable; weft, fur Kuttruff (1988:206) 423373 Rabbit and unidentified (mustelid?) Bogan (unpublished) b 423374 Rabbit and unidentified Bogan (unpublished) b 423377a Rabbit, small rodent, turkey Bogan (unpublished) b 423377b Rabbit, turkey, small rodent Bogan (unpublished) b 423378 Rabbit Bogan (unpublished) b 448913 Bast fibers Kuttruff (1988:207) 448917 Rabbit Bogan (unpublished) b 448917 Bundled plant material caL report 2091 c a numbers in parentheses are the original trowbridge catalog numbers. b Michael a. Bogan, smithsonian institution, national Museum of natural history, Department of anthropol- ogy, spiro textile analysis, unpublished manuscript, 1977. c Unpublished reports of the Museum conservation institute (formerly conservation analytical Laboratory [caL]), various dates. smithsonian institution, washington, D.c. t his chapter treats objects used in resource extraction?farming, manufacturing, or hunting?as well as tools that may have served in warfare- related activities. these artifacts are generally made from lithic materials. the category includes weaponry, artifacts presumably used for hunting or warfare (smaller projectile points), and boatstones or other parts of composite artifacts. in addition, there are lithic artifacts that probably represent a range of functions. special or symbolic weaponry?large, ornate, and decorated objects such as maces or sword- form bifaces?probably served important social or ceremonial functions. Extractive implements comprise tools that were designed to obtain or process resources or raw materials and include chipped- and ground-stone spades, drills, and scrapers. chiPPeD lithic WeaPonrY chipped lithic weaponry can be divided into three major groups: small haft- able points, larger haftable points, and large bifaces that vary widely in form and may or may not have been hafted. Excavation techniques utilized by the Pocola Mining company may not have been conducive to the recovery of small lithic tools, and that there are any small projectile points at all is remarkable. caching may have contributed to the recovery of small chipped lithics. there is little or no minimally modified lithic material, either because debitage was overlooked by the excavators as worthless or because it was not included among grave offerings. stylE chipped lithic tools were examined for edge modification, morphological characteristics, traces from use, manufacturing traces and workmanship indica- tors, raw materials, and styles. Fairly complete point typologies exist for the southern Plains and can be useful for making chronological interpretations. therefore, small projectiles and larger haftable bifaces were assigned to known types styles for the sake of communication and comparability. characteristics 6 weaponry and Extractive tools April K. Sievert 7 4 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y of small point styles that occur in and around spiro were compiled from Bell (1958, 1960), Perino (1968, 1970), Brown (1976, 1996), and Justice (1987). Metric attributes include total axial length, stem length, maximum width of blade, distal stem width (minimum haft width for Brown), proximal stem width (basal width), and thickness. nomenclature for bifaces is shown in Figure 6.1. two ratios in particular are useful in assessing mor- phology and workmanship. these are the thickness to width (t/w) ratio and the proximal stem width to blade width (Psw/Bw) ratio. the t/w ratio offers the relative degree of bifacial thinning. the Psw/Bw ratio gives the relative narrowness of the base, which can be useful for looking at the differences between stemmed and notched points or in distinguishing side- notched Reed points from wide- based washita points in ambiguous cases. manufaCturE Edge modification was described in simple terms of bifa- cial or unifacial retouch and the relative fineness of the knap- ping. other manufacturing variables of importance include characteristics of basal and tang retouch and the presence of barbs. Blade outlines were categorized into triangular, ovate, incurved, excurved, and recurved forms as shown in Figure 6.2. Base shapes can be concave, convex, or straight. hafting nearly all points in the small projectile class have a re- touched hafting element of some type, consisting of shoul- ders, stems, or notches. in Figure 6.1, the hafting element is referred to as the stem. the biface shown in Figure 6.1 displays long extended barbs. not all tools exhibit such an exaggerated barb. if a piece has no over extended barb, it is FIGURE 6.1. Diagram of a flaked lithic artifact showing nomenclature and measurements. after Brown (1976, 1996), Rogers (1980). Ti p Blade Shoulder or Barb Length T ang Thickness Width Stem Stem length Base (Proximal end) Distal Stem Width (DSW) Proximal Stem Width (PSW) Distal end FIGURE 6.2. attributes of biface form. N U M B E R 4 9 ? 75 said to possess a shoulder. in some cases, the base, stem, or shoulders have sustained some damage from twisting mo- tion. this suggests that at least some of the projectiles found at spiro may have been used. the majority appear not to have been used. there are no traces of hafting materials, such as mastic. given the excellent preservation at craig Mound, this seems unusual. it seems unlikely that these tools were hafted at the time when they were placed in the mortuary feature because contact with copper should have contributed to preservation of some hafting compounds had there been any. on the other hand, cleaning by collec- tors may have removed traces of hafting materials. there are no traces of bindings or wooden handles present either. lithiC raw matErials Raw materials derive from a wide variety of locations, and these are important for understanding cultural rela- tionships suggested by social contact and trade. identifica- tion of lithic material was based on visual characteristics and therefore must be considered ?apparent? rather than certain. identifications were made with the assistance of a reference collection of lithic raw materials collected and described by Dickson (1989). samples of all known variet- ies from the ozark highlands, Flint hills, and ouachita highlands are present in this reference collection. Dick- son (1989) classifies these according to geological forma- tion, including subtypes and varieties. Using this reference set, raw material designations could be made. Banks? (1990) study of major raw materials contains a compre- hensive study of raw materials from a wider range than represented by the available reference collection, and this source proved to be invaluable. Lithic raw materials derive primarily from four general locations. Southern Ozarks Most of the lithic materials are expected to derive from sources in the southern ozarks. this area produces the varieties of chert sometimes referred to as Boone chert, but there is considerable variation in the materials from the formations in the region. the following chert varieties from various formations have been identified for the spiro material from the nMnh. type descriptions were com- piled from Dickson (1989) and Banks (1990). Ordovician Cotter. ordovician cotter, including Jefferson city cotter, has distinctive banding in gray shades. Pitkin. Pitkin chert is a highly distinctive dark bluish gray to black chert with a slightly milky appearance. Mississippian Compton. Mississippian compton can be waxy and occurs in blue- gray and tan combinations. Mississippian Reed Springs. Missis- sippian Reed springs cherts come in grays and tans; one variety is green. several varieties have dark splotches sur- rounded by a light ?halo.? texture is usually fine, and many varieties are vitreous or have a waxy luster. Blue- gray varieties weather to tan. this might not be easily dif- ferentiated from the Boone varieties discussed below. B urlington/Boone. cherts from the fol- lowing formations have been included under the name Boone: Keokuk, Elsey, and Burlington. the Elsey For- mation yields cherts in cream to light gray, with a rough texture and dull luster. Keokuk Formation cherts are gen- erally lighter in color than Reed spring, often with wispy and hazy darker or lighter patches. Most varieties are peppered with tiny spots, giving an overall hazy or dusty appearance. these are duller and less vitreous than Reed springs varieties. they occur in larger nodules than Reed springs; consequently, they are used for large artifacts. Burlington chert is usually whitish to blue- gray and is im- proved by intentional heat treating. Flint Hills this area of southern Kansas and north central okla- homa produces some highly distinctive cherts known to have been used prehistorically. Florence (Kay County). this highly dis- tinctive chert from the Florence Formation is character- ized by a light brownish gray to pinkish gray color and concentric narrow banding in alternating light and dark colors. at spiro it appears especially among large ovoid bifaces (Brown, 1976, 1996). Ouachita Mountains Arkansas Novaculite. novaculite ac- counts for 31% of the total chipped lithic collection. the presence of novaculite is culturally relevant because it clearly indicates contact with areas to the southeast of spiro. the texture of novaculite when unheated is distinc- tive. Extremely fine grain size and crystalline structure re- sult in a sugary texture. novaculite may also have been heated to improve flaking qualities. the relative increase in translucence from heating is variable. novaculite flakes in such a way that lipped hinge terminations result in fracture planes being clearly visible on the surface, even though most novaculites are fairly opaque. this results 7 6 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y in translucent slivers that are clearly visible on artifact surfaces. Devonian Woodford. the Devonian woodford Formation (from the ti Valley) contains a dark, vitreous, and fine chert, excellent for knapping. Mississippian Pierson. the Mississippian Pierson Formation produces reddish and bluish varieties, which can be waxy. Red River Red River Jasper. at least one variety of jasper, a dark brown to olive green variety, derives from the Red River basin in texas. this opaque silicate was used for some projectiles in the nMnh spiro collection. Brown (1996:649) refers to this material as ranging in color from buff to dull orange, brown, and red. Texas Panhandle Alibates A gatized Dolomite. alibates agatized dolomite derives from Upper Permian formations along the upper reaches of the canadian River. a major source of alibates used in prehistory is the quarry located in the texas Panhandle, although very small amounts of alibates also occurs in the terraces of the washita River and in gravels along the arkansas River (Potter county historical survey committee, 1964; thurmond, 1991). alibates is characterized by waxy luster, some translu- cence, and wide concentric banding. one large mace from the spiro collections may be manufactured from alibates, a material that resembles agate. Other Sources smoky hills (niobrara) jasper comes from Kansas. other varieties such as Dover chert from tennessee and Mill creek chert from illinois appear in sword- form bi- faces and maces. Mill creek chert is a medium- grained material, opaque, and with a matte luster. Fine- grained Dover chert is often streaked with darker black or gray vitreous inclusions (Brown, 1996:647?648). numerous other lithic materials of unknown source were utilized for chipped- and ground-stone tools. usE and Curation it is expected that tools will exhibit some forms of damage that might have occurred while the object was used kinematically (with active motion) and curated (kept and transported). the evidence for use is discussed in the context of each artifact style. general characteristics of use that can be evaluated for assessing the function of tools are discussed briefly as follows. types of damage that are useful in assessing wear in- clude edge damage, striations, and polishes. Most polishes are invisible under stereomicroscopy, so these are of lesser importance here in assessing damage. Edge damage can de- rive from use, manufacture, and postexcavation damage. Edge damage consists of small flake scars removed from along the edges of tools. certain types of edge damage are highly diagnostic. For example, impact fractures are small flake scars that are removed at the distal tip of a tool upon impact when that tool is used in projection (using a bow or atlatl). the point of impact is clearly shown as a cone of percussion. Lateral edge damage may signify use in cutting or sawing. Damage to the tang, base, or barbs can occur as impact forces the load onto the area of the tool that is firmly affixed to the haft or handle. snap fractures occur if a tool breaks through the medial section as a result of use, manufacture, or postutilization processes. a snap fracture generally exhibits no bulb or cone of percussion, although it may possess a lip or hinge. the other type of fracture that is common is the shear fracture caused when a ro- tating projectile experiences torque when striking a solid. this spin causes fragments of the tool literally to twist off. types of visible damage are shown in Figure 6.3. Damage from curation will be more subtle and can arise from prehistoric or modern manipulation. it devel- ops in places where artifacts rub together during storage, either in a box or drawer after it has been collected or while wrapped loosely in cloth or leather sacks during in the prehistoric past. curation wear appears as smoothing to dorsal ridges, especially if the tools are carried or moved about. the wear need not always occur only on facial ridges, however, as barbs or tang elements may be dam- aged while the arrows are grouped in quivers or bundles. small ProjECtilE Points small projectile points include any bifaces or unifa- cially worked pieces displaying pointed distal tips and modified hafting elements. a wide variety of small point styles can be expected during harlan and spiro phase oc- cupations in the region of the spiro site (Figures 6.4?6.11). clements et al. (1935?1938:263) noted in their wPa report on the lithic materials from craig Mound that projectile points are uniformly small, thin, and finely flaked. Many of them have serrated edges N U M B E R 4 9 ? 77 FIGURE 6.4. small projectile points: a, 423181s01 showing impact flake at tip; b, 423181s02; c, 423182s03 agee point with damage to barb; d, 423182s02 agee point; e, 423182s01 agee point; f, 423181s01; g, 423181s02 Reed point; h, 423181s03. FIGURE 6.3. common forms of damage to projectile points. FIGURE 6.5. small projectile points (423192): a?j, 423192s01?s10 agee; k?o, s11?s15 corner- notched points; p, s16 Maud point; q?aa, s17, s19?s27 ashley (recurvate), agee, and Perdiz (stemmed) points. 7 8 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y and either diagonal or transverse notches. they belong unmistakably to the Lower Mississippi culture, and were found literally in thousands. in many cases they seemed never to have been used, but were manufactured solely as votive offerings. they were commonly found in heaps of several hundred piled between heads of a burial or were sometimes arranged in a definite pattern with the points turned one way. small projectile point styles and raw materials are shown in table 6.1. Raw material usage is described more fully in discussing each point style individually. Stemmed Points stemmed points include alba, hayes, hayes short, Perdiz, and Bonham point styles. Brown (1976) uses the characteristic of wide haft opening to demarcate this group of points. they are barbed and the stems may be well thinned to a semicircular basal outline (alba), narrow and pointed (Perdiz), or left unretouched (hayes). Vary- ing degrees of retouch on bases and tangs can muddy the distinction between alba, hayes, hayes short, and Perdiz. in some cases, what appears to be extremely fine retouch on tang elements may be spontaneous retouch produced during notching. Alba Brown (1996:439) describes alba points as having wide haft openings and wide stems with straight sides and links them to spiro ii grave lots. Most of the nine alba FIGURE 6.7. small projectile points (423192): a?e, s58?s62 Reed and Keota points; f?m, s64?s71 Reed/Keota variety; n, s63. FIGURE 6.6. small projectile points (423192): a?h, s28?s35 Perdiz, alba, and agee points; i, s36 Martindale point; j, s37 haskell; k, s38 side- notched point; l?t, s39?s47 Reed points; u, s18 Reed point; v?dd, s49?s57 Reed and Keota points. N U M B E R 4 9 ? 79 points in the smithsonian collection are barbed. Brown (1996) stresses that retouch on the base and tang for alba points varies, and in one case from the smithsonian collec- tion (423192s31) the basal retouch is minimal. in all cases but two, the lateral retouch is of very fine or fine quality, with regularly spaced pressure flaking and maximal thin- ning. one point is only shallowly retouched, and another is unifacially retouched. of the nine alba points, two are recurved, and the rest triangular. none are broken. Use. one point sustained snap fractures on both barbs (448651s43). two more are damaged at the shoul- ders. all of this damage could be related to excavation damage as easily as wear. Residues. Residues are present on two ar- tifacts. on one, the residue consists of a reddish brown sediment; on another it appears resinous. Raw Material. cherts primarily from the southern ozarks are represented, including Reed springs and Boone varieties. there are three tools of novaculite, and one is Red River jasper. Brown (1996:439) found rela- tively more novaculite in the sample from wPa excava- tions, where it comprises 85% of the total. Hayes hayes points have minimally retouched and some- what pointed stems, but they are otherwise barbed like alba points. Brown (1996:440) places them firmly in spiro ii contexts. For hayes points in the smithsonian collection, thinning and pressure flaking is less fine than on the projectile points designated alba. three are essen- tially unifacial. one is serrated (423192s32). none of the nine points really have pointed bases, which according to Bell (1958:32), is a fundamental characteristic. however, FIGURE 6.8. small projectile points (448650): a, s01 agee; b, s02 Rockwall; c, s03; d?f, s04?s06 sequoyah; g, s07 haskell; h, s08 Morris; i, s09; j?l, s10?s12 agee. FIGURE 6.9. small projectile points: a?f, 448650s13?s18 various; g, 448651s01 Edwards; h?l, 448651 s02, s07, s17, s21, s20 various. 8 0 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y seven have bases that are highly convex and that form an angular juncture with the proximal corner of the notch. two are nearly straight based but are classified as hayes because of their lack of stem retouch. Blade outlines are triangular in two cases, incurved in three cases, and re- curved in four cases. Use. Four points exhibit signs of damage that could be either from use or other mechanical process. one has polished ridges (423192s22) that might indicate curation. Residues. there are no residues beyond sedi- ment on 448651s19. Raw Material. the nine tools are nearly evenly divided between chert (n = 5) and jasper (n = 4). in one case the jasper is banded. Brown (1996:440) reports that jasper made up the larger portion of hayes points in the wPa collection. Hayes Short Brown (1966:440) classifies them as perhaps later than alba and hayes points, with caches associated with spiro iii and iV grave lots, while being found singly in spiro ii grave lots. there are two hayes short points in the smithsonian collection. one has a triangular outline, and the other is incurved. Bases are strongly convex. Bifacial retouch is either fair or moderately fine. Dimensions are 1.68 and 2.04 cm in length, 1.09 and 1.55 cm in width, 0.35 and 0.37 cm in thickness. Residue. none. FIGURE 6.10. small projectile points (448651): a, s03 Reed; b, s04 Keota; c, s05; d, s08 alba; e?f, s24?s25 alba; g, s06 ashley; h, s28 agee; i, s42 agee; j?k, s09?s10; l, s12 haskell. FIGURE 6.11. small projectile points (448651): a, s13 Reed; b, s15; c, s16; d, s14; e, s18; f, s27; g, s38; h, s14; i, s23; j, s11; k, s31, l, s26 Fresno; m, s22 unshouldered; n, s29; o, s30; p?s, s32?s35. N U M B E R 4 9 ? 81 TABLE 6.1. small projectile point styles and raw material summary. Style Style Quartz Novaculite Chert Jasper Unclas. Unknown total agee 1 12 10 1 0 1 25 alba 0 3 5 1 0 0 9 ashley 0 2 2 1 0 0 5 ashley chocoville 0 0 1 0 0 0 1 Bonham 0 0 1 0 0 0 1 Bonham tahlequah 0 0 2 0 0 0 2 collier 0 1 2 0 0 0 3 coryell 0 0 2 0 0 0 2 Edwards 0 0 1 0 0 0 1 Fresno 0 0 2 0 0 0 2 haskell 0 0 4 0 0 0 4 hayes 0 0 5 4 0 0 9 hayes short 0 1 1 0 0 0 2 homan 0 1 0 0 0 0 1 Keota 0 1 9 0 0 0 10 Martindale 0 0 1 0 0 0 1 Massard 0 2 10 0 0 0 12 Maud 0 0 1 0 0 0 1 Morris 0 0 3 0 1 0 4 nodena 0 0 2 0 0 0 2 Perdiz 0 0 6 0 0 0 6 Reed 0 2 10 0 0 0 12 Reed/Keota cluster 0 5 6 0 0 0 11 Rockwall 0 0 2 0 0 0 2 scallorn 0 3 9 0 0 0 12 sequoyah 0 1 3 0 0 0 4 Unclassified 0 0 8 0 0 0 8 Unclassified corner notch 0 0 4 0 0 0 4 Unclassified side notch 0 0 1 0 0 0 1 Uniface 0 0 1 0 0 0 1 washita 0 1 1 0 0 0 2 washita garvin 0 0 2 0 0 0 2 washita chaffee 0 0 1 0 0 0 1 washita peno 0 0 2 0 0 0 2 Young 0 0 1 0 0 0 1 Material total 1 35 121 7 1 1 166 TABLE 6.2. alba points measurement summary (n = 9). Standard Measurement Mean Extremes deviation Length (cm) 2.85 1.89 / 4.51 0.75 width (cm) 1.56 1.2 / 2.11 0.27 thickness (cm) 0.32 0.28 / 0.36 0.03 Proximal stem width (cm) 0.77 0.59 / 0.96 0.13 Distal stem width (cm) 0.68 0.57 / 0.82 0.08 stem length (cm) 0.71 0.55 / 0.82 0.09 thickness/width ratio 0.21 0.17 / 0.3 0.04 width/proximal stem width 2.07 1.36 / 2.69 0.45 TABLE 6.3. hayes points measurement summary (n = 9). Standard Measurement Mean Extremes deviation Length (cm) 2.45 2.07 / 2.71 0.18 width (cm) 1.41 1.14 / 1.64 0.15 thickness (cm) 0.35 0.23 / 0.52 0.11 Proximal stem width (cm) 0.72 0.5 / 0.95 0.14 Distal stem width (cm) 0.64 0.55 / 0.85 0.09 stem length (cm) 0.56 0.45 / 0.7 0.08 thickness/width ratio 0.26 0.15 / 0.46 0.11 width/proximal stem width 2.04 1.2 / 2.8 0.42 8 2 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y Raw Material. one is novaculite and the other is chert (possibly Burlington). Homan the single point classified as homan is novaculite, in- curvate, and 1.75 ? 1.08 cm in size. it is finely retouched and has a use- related break at the tip. Brown (1996:444) found that novaculite was the preferred material for homan points, which are characterized by narrow haft openings. they were included in a spiro iB cache. Bonham the one Bonham point (448651s33) has bifacial re- touch with slight serration. Burination- type flake removal forms both sides of a narrow tang. this might have been either deliberate in manufacturing design or a use (impact) stress fracture. as it stands, the parallel sides of the stem were used as the criterion for assigning it to this type. Length is 1.83 cm, width is 1.03 cm, and the artifact is 0.41 cm thick. Proximal stem width is 0.38 cm, whereas the distal stem width is 0.45 cm, indicating that the stem contracts slightly. a high width to stem width ratio of 2.71 is in keeping with Bonham styling. the base is pointed and the blade triangular. Bonham Tahlequah there are two points of this style, one each from the trowbridge and Meyer collections. overall, retouch is me- dium fine. the point numbered 423192s30 is 2.52 cm long by 1.6 cm wide and 0.4 cm thick. the straight- sided and straight- based stem narrows from 0.5 to 0.41 cm to the proximal end, resulting in a very high width to stem width ratio of 3.9. it is made from a heat- altered chert. the sec- ond, 448651s09, is made from the distinctive Pitkin chert and measures 2.62 cm in length, 1.06 cm in width, and 0.42 cm in thickness. it has sustained a shear fracture to the base and slight shoulder damage. Brown (1996:440) found Bonham tahlequah points in an early cache, in a spiro ia grave lot. Perdiz there are six Perdiz points, and all but one of these has a triangular blade outline; the other is ovate. Bases are either pointed or convex but narrow. Mean length is 3.23 cm, the shortest is 2.5 cm, and the longest is 3.79 cm. width ranges from 1.57 to 2.09 cm, with an average of 1.75 cm. the tools are moderately thin, ranging from 0.26 to 0.43 cm in thickness. the thickness to width ratio is low, ranging from 0.14 to 0.27. stem length ranges from 0.49 to 0.67 cm. one is heat altered and one shows me- chanical damage. Perdiz points are distinctive because in three cases they were made on thin flake blanks (contribut- ing to the low t/w ratio). one is coarsely serrated. Brown (1996:439) notes that Perdiz points do not appear to have been cached in quantity in the craig Mounds graves but appear as isolated finds in spiro ii and spiro iV grave lots. Use. one has edge damage near the tip (423192s24). two display either irregularity or edge dam- age on one edge. Residues. none. Raw Material. at least four appear to be manufactured from chert from the Reed springs Forma- tion. one may be Reed springs or Jefferson city?cotter Undifferentiated. the sixth is probably a Boone chert. Corner- Notched Points the next group includes points that are corner notched or seen to have a narrow haft opening according to Brown (1976, 1996). the distinction between agee, Rockwall, ashley, and homan in this group is sometimes unclear. collier and coryell styles are included here as well. Agee there are 25 agee points in the collection, making agee points the most common style represented. sum- mary measurements are given in table 6.4. Fourteen are serrated, and nearly all are characterized by fine bifacial retouch. Bases may or may not be retouched. ninety- two percent have either triangular or incurvate blade shapes, with triangular blades more common and accounting for TABLE 6.4. agee points measurement summary (n = 24). Standard Measurement Mean Extremes deviation Length (cm) 2.42 1.88 / 3.82 0.43 width (cm) 1.59 1.29 / 2.1 0.2 thickness (cm) 0.37 0.24 / 0.7 0.1 Proximal stem width (cm) 0.83 0.6 / 1.16 0.16 Distal stem width (cm) 0.62 0.47 / 0.83 0.09 stem length (cm) 0.56 0.39 / 0.73 0.1 thickness/width ratio 0.23 0.12 / 0.36 0.6 width/proximal stem width 1.95 1.25 / 2.53 0.36 N U M B E R 4 9 ? 83 60% of the total. one point is slightly ovate and one is slightly excurvate. two were clearly manufactured on flake blanks. Brown (1966:441) places agee points in early com- ponents at spiro on the basis of inclusion in spiro ia, iB, and ii grave lots. they may also have been reinterred in later spiro iV contexts in the great Mortuary. Use. Damage from use is expected to occur in the form of shear fractures to barb or tang elements, distal damage, and damaged serrations. among these artifacts, 10 have damaged barbs or shoulders, some apparently resulting from torque. Four tools are blunted or hinged at the tip, suggesting use. the most likely candidates for tools actually having been fired include 448651s16 and 423192s04. Point 448651s16 has a small bending snap fracture at tip and one broken barb. the base appears to be missing its corners. Point 423192s04 has snap/shear fractures on the shoulders, and the tip has been blunted by a hinge fracture, which could have been caused during manufacture or use. one tool (423192s08) is polished on the ridges, and this could have developed from curation. Residues. no residues are present, with the exception of soil on two points. Raw Material. the most common raw ma- terial is novaculite, which accounts for 48% of the total. Brown (1966:441) found a much higher proportion (85%) of novaculite from the wPa samples of agee points. Forty percent are chert from the southern ozarks. of the ozark chert points, two are of Pitkin chert, and three appear to be of the general Keota and Burlington Formations. there are also two that appear to be from the Reed springs For- mation. in moving to sources farther away, one point ap- pears to be made from Kay county chert, although this is tentative because of the difficulty of assigning chert types on the basis of the small surface area sample presented by small points. at spiro, Kay county chert was used pri- marily for making large bifaces, not small points. there is also one agee point finely chipped from quartz crystal. one point is from Red River jasper, and another is of an unknown material. Rockwall closely related to the agee style is the somewhat larger and more robust Rockwall style. Rockwall appears earlier than agee over much of its range and may well be the precursor to the agee style. Brown (1996:441) shows Rockwall points in conjunction with spiro ia and spiro ii grave lots, where they often occur along with agee and ashley points. Rockwall points are barbed, and they may have deep serrations. there are only two Rockwall points recorded for this collection. one is serrated coarsely, and the other is non- serrated. Both are well barbed. Lengths are 3.15 and 3.44 cm for points 448650s02 and 448651s42, respectively. the points are 1.83 and 1.54 cm wide, respectively. Sequoyah sequoyah points are similar to the agee general pat- tern except that they customarily have long narrow blades often embellished with coarse serration. Four sequoyah points are present in the smithsonian?s collection from spiro. the basal morphology is variable, with two being straight and one slightly convex. there are three such points present in the Meyer collection. one problematic point from the trowbridge collection (423184) was ten- tatively assigned to this category. the artifact has a rather wide blade for sequoyah points, but its serration is coarse, and the point would fit into the group on that basis. the blade tapers somewhat at the tip, another characteristic of sequoyah points. these points range from 3.11 to 4.23 cm in length, 1.22 to 1.41 cm in width, and 0.37 to 0.39 cm in thickness, yielding a high t/w ratio of 1.07?1.75. Justice (1987:223?224) treats sequoyah as an out- growth of the scallorn cluster dating from ad 1000 to the end of the spiro phase. he considers these the most com- mon and typical points found from spiro phase contexts at spiro. Furthermore, Justice reports that although se- quoyah points spread as far north as southwestern indiana, they are not found in appreciable quantities to the south of spiro, as the alba cluster varieties are. Brown (1966:443) reports that sequoyah point caches are present in spiro ii and spiro iV grave lots; however, these points do occur in small amounts spanning the time from spiro ia to spiro iV. Raw Material. one is a light pink novacu- lite (423184), and three are unidentified cherts. Brown (1996:443) reports a range of materials for the points from the wPa collections, including jasper, various cherts, and a small amount of novaculite. Ashley ashley points are similar to agee in style, but with longer stem lengths relative to the tool width. Brown (1996:442) reports that one cache of ashley points de- rived from an early (ia) craig Mound grave lot and that its later inclusion into the great Mortuary was fortuitous. the five ashley points are all finely bifacially re- touched. they are, by definition, strongly recurved in out- line and well barbed. the amount of retouch on the base 8 4 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y varies. the mean length is 2.52 cm, with a minimum of 2.21 cm and a maximum of 3.09 cm. width ranges from 1.24 to 1.78 cm; however, this range underestimates the width because of damage to the barbed shoulders. thick- ness ranges from 0.29 to 0.36 cm. Use. all ashley points are damaged at the shoul- ders; however, they are not damaged at the tips. Raw Material. these are equally split be- tween chert and novaculite. one of the chert points is a fine- quality black chert that may be of the Mississippian Bayou Menard Formation. Ashley Chocoville the single projectile in this style is a classic example of the type, which Brown (1996:442) refers to as a sociotech- nic counterpart to the ashley type (Figure 6.13a). it has a pair of extra notches along the lateral edges and is made from a fine- quality, lustrous light gray chert that appears to be from the Reed springs Formation (423187). Coryell the two coryell points are bifacially retouched in a fine or medium- fine manner. they are 2.0 and 3.4 cm in length and 1.42 and 1.94 cm in width, respectively. the smaller is 0.44 cm thick, and the larger is well thinned at 0.28 cm thick. Use. artifact number 448651s47 has polish on most ridges, probably from prehistoric curation. Raw Material. the polished artifact is of an unknown chert; the other, 448651s20, is made from a chert likely from the Reed springs Formation. Collier there are three collier points. all have fair workman- ship in that bifacial retouch is minimal, steep and shallow, or just shallow. two exhibit stepping or humping from poorly controlled flaking. they vary in size. Lengths range from 1.94 to 3.85 cm. Use. one exhibits edge damage. Raw Material. white novaculite was used for one. the other two are chert, from southern ozarks sources. Martindale a single Martindale point, 423192s36, exhibits bi- facial retouch and a manufacturing error on one lateral edge. stepped terminations create a hump, so the piece is irregular in cross section. it has the characteristic ?fish tail? base (Bell, 1960) and is somewhat translucent. Mar- tindale points were used between 5000 and 1000 bc, so they represent an earlier intrusion into the collection. the length is 3.67 cm, width is 1.81 cm, and the tool is 0.59 cm thick. Residues. a very dark brown residue is pres- ent on basal ridges. the probable age of this piece would make it highly unlikely for this to be mastic for hafting. Raw Material. the waxy lustered raw ma- terial is dotted with whiter inclusions and resembles Reed springs chert. Massard Massard points are corner- notched varieties without barbs. Brown (1996:440) lists these as appearing in spiro ii through spiro iV grave lots. among the 12 Massard points, one is ovate, one is incurved in outline, and the rest are triangular. Four have straight bases, whereas eight are subconvex to convex. two are broken at the shoulders. Measurements are given in table 6.5. Use. six points exhibit damage traces that may be use related. these are manifested as shear fractures or burination at the shoulders and, in one case, distal damage consistent with impact. Residues. one has a brown stain, and another has a brownish sediment adhering to the surface. Raw Material. one is clearly novaculite from the ouachita Mountains of arkansas. another is possibly a heat- altered novaculite that resembles chalced- ony. chalcedony would be an unusual material for this area. heat alteration might alter the novaculite, giving it TABLE 6.5. Massard points measurement summary (n = 12). Standard Measurement Mean Extremes deviation Length (cm) 2.38 1.87 / 2.68 0.28 width (cm) 1.18 0.84 / 1.51 0.17 thickness (cm) 0.39 0.3 / 0.52 0.07 Proximal stem width (cm) 0.88 0.59 / 1.21 0.16 Distal stem width (cm) 0.6 0.45 / 0.72 0.07 stem length (cm) 0.7 0.54 / 0.85 0.1 thickness/width ratio 0.34 0.21 / 0.47 0.8 width/proximal stem width 1.36 0.95 / 1.78 0.22 N U M B E R 4 9 ? 85 the waxiness and translucence usually characteristic of chalcedony. two novaculite tools would be fewer than ex- pected on the basis of Brown?s (1976, 1996) study, which indicates Massard (especially the variety designated ?B?) points are produced predominantly from novaculite. this can be explained partially in terms of the small sample size of Massard points in the smithsonian collection and partially in terms of the subjectivity involved in doing point typologies. three of the cherts could not be identi- fied with any confidence. one may be Keokuk, Prior ag- gregate variety, and another has the characteristic banding of the ordovician cotter or Jefferson city Formations. another may be from the Reed springs Formation, also in the southern ozark highlands. Scallorn scallorn points are summarized in table 6.6. the 12 scallorn points are distinguished by corner notching and are usually barbed. Blade outline is predominantly trian- gular (62%), although there are ovate outlines (23%) and one each of incurvate and asymmetrical triangular outlines. one- half of the bases are straight, five are convex, and one is concave. serration does occur on the point edges. work- manship is good, with fine bifacial retouch being typical, although there is one unifacially worked point that falls into this class. Justice (1987:220?222) places this cluster in the Late woodland to Mississippian transition, about ad 700?1000 or the Fourche Maline, Evans, and harlan phases. Brown (1996:442) reports on scallorn points from spiro ii and iii grave lots, implying that scallorn points were being made later, from ad 1100 to 1300. Use. shear fractures on the barbs is the most frequent damage recorded. three may be altered by heat. Point 423192s04 is serrated, with a retouched base and snap or shear fractures on the shoulders. the tip is blunted by a hinge fracture that could also have been caused dur- ing manufacture. artifact 448651s15 exhibits a bending fracture at the tip that could easily be from use. Residues. two points have some form of resi- due that may be prehistoric. Point 423192s63 has orange residue near the tip on the numbered face. Raw Material. nine of the 12 are chert, and three are of arkansas novaculite. the chert points all ap- pear to be of Keota, Burlington, or Reed springs Forma- tions. one may be of the Pierson Formation. Side- Notched Points side- notched points include Reed, Morris, Keota, haskell, and washita and its variations. the side- notched group as a whole usually exhibits fine thinning and sec- ondary retouch, but points are unbarbed. Reed this straight- based point style is listed by Brown (1996:444) in grave lots as early as spiro ia but is pres- ent in larger numbers in caches in spiro ii/iii and spiro iii grave lots. the 12 Reed points reported here are charac- terized by good to fine bifacial thinning and basal retouch (Figure 6.12). none of the Reed points are broken. By and large, they are thin and finely worked, although several are steeply retouched. Eight are triangular, and four are ovate in outline. two of the Reed points are serrated. Use. Fine edge damage is present on one point, but otherwise, characteristics of use are lacking. Raw Material. all are chert except for two, which are novaculite. Keota ten projectile points have the convex bases and rel- atively narrow side notches characteristic of the Keota style. three bear traces of heat alteration. Eighty percent have a triangular blade outline; 20% are ovate. the base may or may not be retouched. tools having a lower t/w ratio also tend to have retouched bases. there is a varying degree of fineness displayed in finishing knapping. Keota points are found in a variety of grave lot styles, occasion- ally in spiro iB and ii and cached in spiro iii and iV grave lots (Brown, 1996:444). Use. Point 448651s53 (Figure 6.12p) has steep unifacial retouch to form a thick tip. a steep intersection at the tip would make a useful graver or incising tool and TABLE 6.6. scallorn points measurement summary (n = 12). Standard Measurement Mean Extremes deviation Length (cm) 2.67 1.74 / 3.82 0.63 width (cm) 1.38 1.06 / 1.69 0.2 thickness (cm) 0.47 0.27 / 1.44 0.3 Proximal stem width (cm) 0.96 0.61 / 1.49 0.26 Distal stem width (cm) 0.62 0.53 / 0.78 0.9 stem length (cm) 0.62 0.39 / 0.77 0.12 thickness/width ratio 0.34 0.21 / 0.93 0.18 width/proximal stem width 1.51 0.94 / 0.33 0.39 8 6 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y would be unusual for projectile points. the tip edge resem- bles scraper retouch but is pointed very sharply. also, the wide blade would strengthen a tool that would be used for gouging, although there are no obvious direct traces of use. Residues. Yellowish sediment on one, black resinous residue on another. Raw Material. of 10 Keota points, nine are chert and only one is a highly lustrous novaculite. artifact 448651s53 is a small Keota point made of pink semitranslucent, nonwaxy novaculite that is homogeneous and nongrainy. of the chert points, one appears to be the deeply blue- black waxy Pitkin chert, one is unclassified, and the remainder appear to be of the Burlington, Keota, Elsley cluster (Boone). of the Boone group, one has tiny dark gray flecks in the matrix. another is a very distinctive mottled and variegated highly lustrous chert in grays with some brown swirls. it resembles samples of Keota Forma- tion, Prior aggregate variety. another is probably Reed springs. all raw materials with the exception of the no- vaculite are regionally available, and therefore, the Keota points do not present much in the way of exotic material. Reed/Keota Variation there are side- notched points made using a slightly different design from either Reed or Keota. these non- barbed points are asymmetrical in cross section and are not nearly as finely thinned as are others in the Reed, Keota, and washita macrogroup. they have long stems and wide side notches. Bifacial retouch does not always reach the axis of the piece, showing clearly the ventral surface of a flake blank on which these were almost invariably made. the retouch is in most cases steep and shallow. Blade out- lines are primarily triangular but include incurvate shapes FIGURE 6.12. small projectile points (448651): a?b, s36?s37; c, s39 alba; d, s40 Reed; e, s41; f, s43 alba; g?h, s44?s45; i, s48 Massard; j, s46; k, s47 coryell; l, s45 agee; m, s54; n, s50 hayes; o, s51; p, s53 Keota; q, s52 ground washita point. TABLE 6.7. Reed points measurement summary (n = 12). Standard Measurement Mean Extremes deviation Length (cm) 2.27 1.59 / 3.88 0.63 width (cm) 1.12 0.75 / 1.94 0.35 thickness (cm) 0.38 0.26 / 0.97 0.2 Proximal stem width (cm) 1.11 1.89 / 1.52 0.2 Distal stem width (cm) 0.63 0.49 / 0.86 0.13 stem length (cm) 0.64 0.55 / 0.86 0.1 thickness/width ratio 0.37 0.14 / 1.29 0.3 width/proximal stem width 1.0 0.77 / 1.69 0.24 TABLE 6.8. Keota points measurement summary (n = 10). Standard Measurement Mean Extremes deviation Length (cm) 2.53 1.52 / 3.49 0.65 width (cm) 1.19 0.97 / 1.43 0.13 thickness (cm) 0.39 0.28 / 0.57 0.03 Proximal stem width (cm) 1.1 0.98 / 1.25 0.1 Distal stem width (cm) 0.7 0.52 / 0.89 0.11 stem length (cm) 0.76 0.64 / 0.98 0.09 thickness/width ratio 0.33 0.24 / 0.45 0.7 width/proximal stem width 1.08 0.9 / 1.37 0.16 N U M B E R 4 9 ? 87 as well. three have a tip that tapers to a nearly needlelike point. Four of 11 are serrated, one of which shows pro- nounced serration. identical points are shown for the har- lan site (Bell, 1984:234). there are two varieties, straight based and convex based. the straight- based variation is clearly side notched, but the notches are very wide, giving the piece a long stem. the base is unretouched. included are 423192s40, 423192s44, and 423192s67. all are very finely serrated. these are all novaculite. the convex- based variation has exactly the same notching and blade form as the straight- based examples, but the base is retouched convex (though the basal retouch may be unifacial or minimal). included are 423192s45, 423192s64, 423192s65, and 423192s66. the two raw materials used to make them are novaculite and dark blue- gray Pitkin chert. Five of 11 points in this combined clas- sification are Pitkin chert. Both varieties can easily be combined into one type that overlaps the Reed and Keota types, hence the group- ing here. the tools listed as both above, combined with 423192s69 and 423192s71, all seem to go together. one could also add 423192s51 and 423192s55, which are es- sentially Reed points but are very like 423192s65 both in workmanship and raw material. although singled out here as a variation on the Mississippian styles of Reed and Keota points, which share some characteristics with ca- hokia (washita) points in style, Justice (1987) would prob- ably place these points into the same group as sequoyah. Use. none of these points are broken, nor are there any other indicators that would suggest they had ac- tually been used. Residues. none. Morris there are four Morris points that have the character- istic deeply concave basal morphology. Brown (1996:444) places these points between ad 1000 and 1200. the mean length is 2.36 cm, with a minimum of 1.47 cm and a maxi- mum of 3.13 cm. Mean width is 1.26 cm, minimum width is 1.05 cm, and maximum width is 1.48 cm. three of the points are finely serrated. all appear to have been altered by heat. workmanship is very good. Use. one exhibits a shear fracture on the base. Residues. Point 423185s02 has a band of dark material that may be a residue. Raw Material. three of these points are from chert, and one is an unidentified material. Brown (1996:444) found high levels of heat treating for the Mor- ris points in the wPa collections. Haskell two concave- based haskell points are finely flaked using well- controlled pressure flaking techniques. one (423192s37) resembles a Reed point (423192s46) in both raw material and flaking style and could have been produced by the same flintknapper. this would support the contention that some of these pieces do derive from cached materials. the third haskell point is steeply re- touched on the edges only. a fourth was surface collected at the site after the wPa excavations. haskell points from the wPa work were found cached in spiro ii and iii grave lots (Brown, 1996:444). Use. the two finely flaked haskell points each have damaged tips. Residues. one has a deposit of sediment. Raw Material. all are chert. Brown (1996:444) reports that haskell points he analyzed were primarily made of a dark gray, red- banded chert. Washita there are five washita points in the varieties washita garvin, washita peno, and washita chaffee. washita- style projectiles are characterized by fine flaking, thin cross sec- tion, and relatively highly placed narrow side notches. these are typical of the cahokia site in illinois. two more washita points were not classified into subtypes. accord- ing to Brown (1996:445), the washita garvin style was found cached in spiro iii grave lots and is also found in some spiro ii contexts. TABLE 6.9. Reed/Keota variety points measurement summary (n = 11). Standard Measurement Mean Extremes deviation Length (cm) 2.49 1.89 / 3.27 0.42 width (cm) 1.12 0.91 / 1.29 0.11 thickness (cm) 0.37 0.3 / 0.5 0.06 Proximal stem width (cm) 1.09 0.92 / 1.24 0.1 Distal stem width (cm) 0.59 0.42 / 0.75 0.1 stem length (cm) 0.8 0.6 / 0.99 0.1 thickness/width ratio 0.33 0.25 / 0.42 0.05 width/proximal stem width 1.04 0.89 / 1.24 0.12 8 8 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y Use. the washita peno points both show signs of use in the form of a blunt tip on one (423192s38) and a burination from impact on the other (448651s03). Raw Material. the two washita garvin points are made from gray Reed springs chert and Boone chert. the washita peno points are both of Reed springs chert. one unspecified washita point of novaculite is heavily ground on both faces (Figure 6.12q, 448651s52). one thick novaculite point is unusual in that the edges have been ground on both faces (Figure 6.12q). the grind- ing nearly, but not quite, eliminates the flake scars from knapping. this type of grinding is rare for projectile points from north america. the other unspecified washita ex- ample is chert of unknown source. the washita chaffee point is possibly made from a high- quality chert of the Devonian woodford Formation from the ouachita hills. Unclassified Side- Notched Points one side- notched point (448650s16) was left unclassi- fied because it exhibits characteristics of more than one type. the small tool has a typical Reed base but is coarsely serrated and more ovate in form than most Reed points are. it may be a variety of Reed point or a resharpened tool, giving it the shorter overall aspect and more rounded form. the small point of light olive- gray Reed springs chert measures 2.32 cm in length, 1.29 cm in width, and 0.24 cm in thickness. Unshouldered Points Unshouldered points are rare and include Fresno (straight based), Young and nodena (convex based), and a single Maud point with serrations part of the way down the edges along with a concave base. Maud there is one Maud point of a very fine grained black chert with a matte luster. although rare from the spiro craig Mound collection, Maud points are reported else- where at spiro phase localities nearby. For example, Rohrbaugh (1985a:70) records the incidence of four Maud points at the geren site (34Lf36). Fresno small projectiles with a straight base and triangular blade outline are classified as Fresno points. there are two such in the spiro collection (378271s03 and 448651s26), measuring 1.48 and 1.88 cm in length, 1.26 and 1.07 cm in width, and 0.28 and 0.26 cm in thickness, respectively. they are both chert. one appears to be consistent with Boone chert varieties, and the other is unidentifiable. Brown (1996:446) finds Fresno points in relatively late contexts (spiro iV) based on his grave lot seriation. Young and Nodena there are three unshouldered, convex- based points with ovate blade outlines. the two nodena points, 448651s22 and 448652s06, average 1.98 cm in length and 0.95 cm in width. the t/w ratio is 0.24. the Young style (448650s18) is characterized by a wider blade (1.27 cm). it is longer overall than the nodena points at 2.58 cm. Unclassified Points Eight points were unclassified. in some cases diag- nostic features are broken, hampering classification. two of these (448651s51 and 448651s46) are similar arrow points with wide shallow side notches. they are made from Reed springs chert. another narrow, thick point (378271s01) was manufactured from a reddish yellow chert, possibly a variety of Boone or Keokuk chert. Small Projectile Point Summary Heat Alteration heat treatment was almost certainly done to improve flaking quality of some of the raw materials. however, in- tentional heat treatment may produce only subtle effects in some materials and may be confused with nonintentional heat alteration in others. conversely, it is important not to assume intentional heat treatment every time an arti- fact displays signs of having been heated because there are many other factors that can bring tools into contact with heat, including hearth proximity, prairie fire, or cremation. Postmanufacture heating is suggested by eight projec- tile points that have a pink or reddish tip consistent with heat alteration (table 6.10). if the reddening is from heat, it must be postmanufacture because otherwise it would not affect just the tip unless mostly finished preforms were being heated. this alteration could be some form of pur- poseful treatment as part of the mortuary procedure or simply a characteristic of chance contact with heat. the characteristic of postutilization heat alteration needs to be examined for contexts with better provenience, especially in nonburial situations. all in all, few points ap- pear to have burned after use. Burning after use might be N U M B E R 4 9 ? 89 expected (1) as a function of random disposal in domestic contexts or (2) as a function of mortuary ritual in which some materials might be burned. Postutilization heat al- teration is usually uncontrolled (as opposed to controlled heat treatment to improve flaking quality) and often results in crazing, thermal fracturing, potlid scarring, and resin- ous or sooty residues. none of the specimens from spiro exhibit any of these dramatic damage indicators. Caches some similarities among points suggest cached depos- its with points having been made of similar material by a single manufacturer. there are definite similarities among points in the trowbridge and Meyer collections. cach- ing was studied by Brown (1976, 1996). several points in the trowbridge collection are of an unusual olive brown lusterless opaque jasper, which probably comes from the Red River area (Brown, personal communication, 1992). these are similar morphologically as well and fit into the hayes and alba groups. they could easily represent a group made by a single manufacturer and perhaps were cached in a single burial context or feature. Likely candidates for cache composition among the small points include (1) group a, 423192s51, s55, s65; (2) group B, 423192s66, s67; and (3) group c, 423192s17, s18, s19, s20, s21, s22, maybe 423192s08 with s33. Use there are some indicators of use. in most cases they consist of damage to shoulder, barb, or base, characteristic of damage from projection when the force of impact is loaded onto the haft. however, such damage may also be produced by mechanical, excavation, and curation (keep- ing a tool) processes. there are few broken pieces. the propensity for tools to be complete might result from the preference among both diggers and dealers to collect com- plete pieces to which they accord a higher value. shear fractures on barbs would be unlikely to affect the overall value greatly or might not even be noticed by a collector. the possibility for arrows to have been held together in a quiver might account for the relatively high amount of shoulder damage. if arrows are held together in a tight cluster, the portions of the tools that will come in contact with one another are the barbs or shoulders, and these will be more likely to become damaged. this is an example of damage that may occur as a result of curation: keeping and perhaps transporting objects together. small projectile points recovered from nonmortuary contexts at spiro appear to have incurred a greater de- gree of damage. Most of the points illustrated from the 1979 excavations at spiro (Rogers, 1980:172) exhibit some form of breakage consistent with wastage during use. small points in the nMnh collection from the craig Mound are remarkably whole, given the total number of small points. although the projectiles may not have been in a pristine state when they were deposited, as a group they were definitely utilized only to a slight degree. largE ProjECtilE Points Larger bifaces that display a haft element but are too large to have functioned as arrow points are present in the collection and were classified as large points. Most are stemmed as opposed to notched, although there are excep- tions. these may have been utilized as knives or projectiles for use with atlatls or spears. among the large points, the most common styles are gary and Lange. there are also others that do not fit well into known types, and those were listed as unclassified. Gary gary points are distinguished by ovate blade outlines and stems with well- rounded convex bases. they range from Late archaic to woodland contexts and are associ- ated with Fourche Maline phase sites in arkansas (Jus- tice, 1987:189). since these tools are generally earlier than the spiro phase great Mortuary deposit, they were pos- sibly redeposited in the craig Mound as fill taken from older habitation areas during mound construction. there TABLE 6.10. Presence of postmanufacturing heat alteration in small projectile points. Catalog number Style Notes 423192s05 agee already red; tip and barbs slightly darker 423192s14 Unclassified corner notch 423192s24 Perdiz grayish red tip on a light brown tool 423192s47 Reed slightly pinker at tip 423192s49 Keota Red tip 448651s50 hayes More pink at tip 448651s24 alba although entire piece is red, tip is darkest area on tool 448650s08 Morris Redder tip 9 0 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y are four gary points, one of which has a midblade snap fracture. the other three range from 5.74 to 9.02 cm in length and 2.7 to 3.87 cm in width. thickness ranges from 0.88 to 1.76 cm. stem length ranges from 0.87 to 2.17 cm. stem width at the distal juncture with the blade ranges from 1.72 to 1.99 cm. Bifacial thinning is deep, and in three cases secondary edge retouch is minimal, although stepped terminations are common. Use. the gary points do appear to have been utilized. one (448652s01) has a medial break that could have been incurred in any number of ways, including use. another (448652s03) has burination originating at the shoulder and continuing more than one- half of the dis- tance up one lateral edge. it also shows edge rounding es- pecially at the tip. artifact 448929 exhibits clear damage and blunting to the tip. the fourth tool has a flake taken out of one shoulder, which may be from either manufac- ture or use. all of these damage types are consistent with the action of projection (sievert, 1992), and none would typify utilization along a lateral edge in cutting or sawing. Residues. no residues were present. Raw Material. all are chert; however, in most cases the type of chert is unknown. one may be chert from the Jefferson city Formation. two are of a material that has apparently been heated. Lange Lange- style lithic artifacts are characterized by straight or slightly expanding stems (Figure 6.13c,d.). the blade is usually narrow relative to length. there are four such tools in the spiro collection. three are large and range from 9.48 to 11.42 cm in length, 3.23 to 3.69 cm in width, and 0.8 to 1.15 cm in thickness. the fourth (448652s02) is considerably smaller at 5.4 cm long, 2.65 cm wide, and 0.57 cm thick. Lintz (1978:141) discusses similar Lange points recovered on survey in haskell county, oklahoma. Use. Point 378372 exhibits highly polished ridges and a green stain that is probably copper. two others have damage consistent with impact; 423175s01 exhibits burination on the stem and a corresponding shear fracture on the shoulder of the same edge. Point 448652s02 has distal damage typical of impact. Raw Material. all are chert. however, in one case (423175) the biface is of an unusual purple color, which possibly resulted from heating. another (423176) is TABLE 6.11. Large projectile point styles and raw materials. Catalog Style number Raw material gary 417939s02 Jefferson city?cotter chert 448652s01 Keokuk/Boone chert 448652s03 chert 448929 chert Lange 423175s01 Keokuk chert 423176 Jefferson city?cotter chert 448652s02 Keokuk/Boone chert Johnson 423189 Boone or Reed springs chert graham cave 423190s01 chert 423177 chert Expanding stem B 423190s02 chert Eccentric 448653 Keokuk/Boone chert FIGURE 6.13. Large and unusual points: a, ashley chocoville point (423187); b, very early point, graham cave (423177); c, d, Lange points (423175s01, 423176); e, Kay point (423175). N U M B E R 4 9 ? 91 a distinctive banded chert not inconsistent with the cherts of the cotter Formation in the southern ozarks. Johnson one haftable point exhibits a pattern of inverse re- touch that creates a bevel and is tentatively assigned to the Johnson category. it has shallow serration and stepped termination on the secondary retouch flake scars. it is unbarbed, with a minimally retouched base. it resembles Ellis points (Bell, 1958) with beveling along the sides of the blade, but the stem does not expand significantly. Johnson points are usually dated to the Late archaic or Early woodland periods (Perino, 1968). Raw material is a typical Boone chert. Graham Cave one biface has an ovate blade outline, a definite side notch and a slightly concave base (423190s01). it is made from a dull brown chert and measures 5.87 ? 2.07 cm. an- other (423177, Figure 6.13b) resembles 423190s01 except that the blade is excurvate, the base is markedly concave, and the ears are more rounded. these both appear to be archaic period point styles, with resemblance to the Early to Middle archaic graham cave styles found in the region (Justice, 1987). along with the Johnson point, these two early points may have been deposited in the craig Mound through the mound building process, or they may have been intentionally collected from earlier sites by the spiro people and redeposited in the secondary burials placed into the great Mortuary. Expanding Stem Points artifact 423190s02 is a medium- sized haftable point having an excurvate outline and wide expanding stem. it is 4.98 cm long, 2.59 cm wide, and fairly thick at 0.98 cm. Raw material is a fine- quality waxy- lustered translucent brown chert. Eccentric Points there is one truly eccentric haftable biface produced from a good quality white chert not unlike Burlington (448653, Figure 6.14e). it is a complete tool 7.45 cm long, 4.02 cm wide, and 0.65 cm thick. Distal stem width is 2.2 cm, which contracts to a proximal stem width of 1.21 cm. the stem is 1.41 cm long and notched in the center of the base. there are three sets of notches along the triangular blade edges. the distal notch is shallow, the medial notch is wide and deep, and the proximal notch is shallower and narrower than the medial notch. the shoulders are barbed. largE bifaCEs there are various types of larger bifaces ranging from ovoid bifaces to sword- form knives and maces. clements et al. (1935?1938:263) noted that finely shaped, very thin stone blades are typical of this culture. these blades range from eight inches to twenty- eight inches in length and seem to have been manufactured for ceremonial use as they are far too fragile to have served any utilitarian purpose, although the larger ones may have been used as knives. the only large points that were found were ?V? shaped with small tangs at the apices of the ?V? and average about five inches in length. Raw materials used in producing large bifaces vary (see table 6.12). Banks (1984) mentions argillite, which was used for some celts and reputedly comes from a ?silt- stone? ledge in the atoka Formation, which abuts the arkansas River. smoky hills jasper, also called niobrara jasper, comes from Kansas but was utilized at spiro, pri- marily for large bifaces. novaculite, quartzite, and variet- ies of chert complete the inventory of materials used for large bifaces. in most cases, these large bifaces probably served a social or ceremonial function. haftablE bifaCEs Ovoid Bifaces there are two examples of the type that Brown (1976:127?131; 1996:464) refers to as ?ovoid bifaces,? type a (Figure 6.14f, h). they are made from Kay county chert of the Flint hills region of northern oklahoma. this chert has the characteristic banding, and the bifaces have lopsided blade outlines, with one edge convex and the other straight. one is broken, such that only the proxi- mal end is present. they are notched a short distance from the base. the complete biface has thick brown sediment adhering to one face near the tip and is well polished on dorsal ridges. near the tip, along both lateral edges the residue is abruptly truncated about 1 mm from the edge, leaving a narrow residue- free band. Function of these tools is uncertain. 9 2 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y an unusually shaped ovoid biface is present (423179, Figure 6.14g.). this is a fine, thin blade made from what is probably heat- altered novaculite. it has a clear red color and waxy luster. the tool has a well- defined shoulder and shal- low notches in the lateral edges set high on the blade distal to the shoulder. some edge damage is visible on one lateral edge near the tip. the tool measures 12.19 cm in length and 3.81 cm in width. another biface fragment is made from a dense grainy material that appears to be siltstone (423191s03). it is broken and tapers near the tip of the wide, thin blade. Kay one biface corresponds to the morphology reported for Kay points (Brown, 1976); however, it is crudely made of an unidentified yellowish brown grainy chert instead of the more usual Kay county chert (Figure 6.13e). this point (423175s02) has rounded corner notches and a bu- rination extending up the stem from the base. there is an odd whitish residue on one face of this artifact that could be adhesive. it is 11.53 cm long and 3.95 cm wide. FIGURE 6.14. Large bifaces and drills: a, expanding stemmed drill (423193s01); b, barbed drill (423193s02); c, drill made on a pebble (448655); d, quartz graver (423191s04); e, eccentric point (448653); f, h, ovoid bifaces of Kay county chert (423178, 417939); g, novaculite biface (423179). N U M B E R 4 9 ? 93 Preforms there are nine pieces that may be preforms. of these, three are large bifacially worked tools (423170) of novac- ulite or chert with triangular outlines and broad flat bases. it is conceivable that such tools could have been utilized as celts; however, it is also possible that since they are of nonlocal raw material, they may have been imported in this form, to be used or finished by local knappers. one is of a high- quality grayish novaculite (Figure 6.15), similar in raw material to that used to manufacture biface num- ber 423180 (a long, thick lanceolate biface which likewise could be a preform but is more likely a resharpened knife form). ovoid and lanceolate bifaces of catalog number 448654 consist of four chert forms and one novaculite form, all of which appear unfinished. SPecial chiPPeD-Stone WeaPonrY Brown (1996:470) divides special weaponry clubs into three types: maces, axes, and sword forms. Both chipped- stone and ground-stone manufacturing methods are used for celts and maces. sword- form tools are chipped stone. ceremonial weapons were perhaps designed to resemble more utilitarian weapons; however, they were embellished or altered to reflect a symbolic component. contexts of use will involve display or demonstration. sword- form bifaCEs sword- form bifaces are large bifacially worked tools that fit into several morphological classes, including gen- eral bifaces, Duck River bifaces, and fusi- elliptical form bifaces. these forms may have served as batons or clubs rather than as knives or other cutting tools. they are listed in table 6.12. Duck River spiro is known to have yielded fragments of very long narrow blades, often pointed at both ends. a recon- structed biface and five fragments (Figure 6.16) are pres- ent. these artifacts are similar in morphology and raw material to the Duck River cache of bifaces found in ten- nessee (Brown, 1996:474) and have been given the name ?Duck River? for that reason. Brown (1976:159) notes that all of the ones in the wPa collection were broken and writes that they can be expected to occur in highly fragmented form, suggesting intentional breakage or cer- emonial ?killing.? Pronounced hinge fractures as well as FIGURE 6.15. Large triangular preform or celt- form biface (423170s01). TABLE 6.12. Large biface types and raw material summary. Catalog Type number Raw material ovoid biface a 417939s01 Kay county chert 423178 Kay county chert ovoid (irregular) 423179 novaculite Kay 423175s02 chert celt form 423170s01 novaculite 423170s02 novaculite 423170s03 chert 423173 chert Preform 423180 novaculite 448654s02 chert 448654s03 novaculite 448654s04 chert 448654s05 chert general sword form 423174s01 chert 423174s03 chert 423191s03 siltstone Duck River sword form 423172s01 Boone chert 423174s04 Mill creek chert 423174s05 Mill creek chert 423174s06 Mill creek chert 423174s07 Dover chert 423174s08 Dover chert Fusi-elliptical sword form 423171 Quartzite hoe 423174s02 chert Unclassified 448654s01 chert 448690s01 Unknown 9 4 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y battering and retouch on the edges of medial breaks sup- port the idea that the blades were deliberately broken. the pieces in the smithsonian collection are all smeared with green pigment, and the five nonreconstructed pieces bear ocher stains as well. this residue is truncated by hinge fractures; in one case (423174s05) the hinge fracture ex- tends for 2.81 cm onto the face. therefore, the pieces must have been smeared with pigments and then ceremonially broken and strewn about. the fragments range from 4.39 to 8.43 cm in length and 3.86 to 2.53 cm in width. the reconstructed biface is composed of at least four other fragments clumsily fit together to make a 30 cm blade (423172). trowbridge acquired the piece and re- corded in his catalog that this blade was constructed from similar but nonrefitting fragments by J. w. Balloun, from whom he purchased it. one join has given way so that the tool is now in two pieces. an outstanding feature of the reconstructed biface is the retouch on the edges of one of the breaks. it appears to be recent and occurs on an edge that is broken at an oblique angle. it does not seem pos- sible that this edge could be the base of the fragment but perhaps was nicked away at the edges by Mr. Balloun in order to produce fragments that would fit together in a straight line, thereby reducing the true edge of the break and creating a false ?fit.? Residues on the tool are plentiful and include traces of glauconite, copper, and ocher. Mod- ern residues include plaster or wood filler, green pigment, and adhesive. the entire surface is smeared with recon- structive material. Beneath the material, relatively fresh flake scars can be seen along the broken edges, which indi- cates that the dealer who reconstructed the blade battered away at the broken edges in order to create a smooth fit. the chert is homogeneous and peppered with very small dark speckles. the same raw material has been used in FIGURE 6.16. Duck River bifaces: a, reconstructed biface (423172); b, fragments (423174). N U M B E R 4 9 ? 95 each piece. hamilton (1952: pl. 46) shows several bifaces that may have been similarly reconstructed. there are two types of chert represented (table 6.12). one is a grainy brownish chert that is probably mate- rial from Mill creek in illinois. the other form is a finer- grained, lighter gray, slightly mottled chert typical of materials from Dover, tennessee (Brown, personal com- munication, 1992). the reconstructed biface appears to be of the Dover variety, although the raw material of some of its component fragments is obscured by applied recon- struction material. General Sword- Form Bifaces in the collection is a complete but broken biface of a white chert (423174s01, Figure 6.17b). it consists of four large fragments and one small fragment that refit to a length of 25.5 cm. the maximum width is 6 cm. the blade is pointed distally and rounded proximally. the pointed end is refitted from a fragment from lot 423191, and this break occurred before excavation. there are at least two breaks that occurred after cataloguing at the nMnh. a brown residue adheres to one face only and is present on the oldest break surface. the point of impact for the ear- liest break is on the face of the piece, leaving a cone of percussion on the break surface. this is a typical result of hitting a biface with a stone hammer to break it intention- ally (sievert, 1992). the white chert has a homogeneous texture and sparse darker elliptical spots with very light halos suggestive of the Reed springs Formation. Biface fragment 423174s03 has an asymmetrical out- line with one edge being more convex. the straighter edge is clearly beveled, with the opposite edge retouched. the retouched convex edge is polished, and the distribution of wear suggests use as a side scraper. it is of a pinkish gray chert of unknown origin. Brown (1996:472) reported this type of biface from spiro ii and spiro iV grave lots. he found the raw materials to include smoky hills jasper, Kay county chert, and other unidentified cherts. Fusi- elliptical Bifaces these long bifaces are wider near the distal end. artifact 423171 has a distinct shoulder occurring 3.96 cm from one end. Bifacial thinning with secondary edge retouch has produced a sinuous cross section. there is polish on either end and on dorsal ridges. it is hard to say which end is distal and which is proximal. gener- ally, there is more polish on the shorter, wider end, which could be from hafting or from use, such as digging. Polish extends well up toward midblade on some ridges. it is clearly fusi- elliptical in outline but broader and thicker than other similar bifaces described by Brown (1976:161, 1996:474), which could put it into a class more similar to the Ramey knife described by Perino (1963). the raw material is a dark quartzite with sparse darker gray to black mottles and streaks surrounded by halos of a rusty hue. it is unlike any other lithic material encountered in this collection. maCEs there are four large, ornate, and impressive crown- form bifaces of varied morphology. these bifaces have been referred to as crown maces and fall into a number of FIGURE 6.17. general sword- form bifaces: a, 423191; b, 423174. 9 6 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y documented morphological types (Brown, 1976:166?173; 1996:474?476). the first is represented only by the elaborate haft ele- ment consisting of a stem and wide flanges proximal to deep notches (423196, Figure 6.18a). the margins of the haft element are discolored to a very dark grayish brown, possibly related to the way it was hafted. the discolor- ation may be due to mastic or bindings. this discoloration follows the edges, where bindings would be expected, not in the center, where mastic might be expected. none of the tools in this collection exhibit any trace of mastic, which implies that mastic was simply not used. the mace has been shown in hamilton (1952: pl. 44a). During the 1970s, a. segovia (University of Mary- land, Department of geology) identified the raw mate- rial as novaculite (Joan s. gardner, unpublished notes on conservation measures taken for the spiro collections at the national Museum of natural history, smithsonian institution, 1970?1971). however, the texture, color, and inclusions are unlike that of any of the novaculites in the reference collection. Rather, this tool appears to be a dark grayish brown fine- quality chert, resembling those of the woodford Formation. there are at least two kinds of resi- due on the piece, a greenish clay that resembles glauconite and a brown sediment that overlays the glauconite and probably represents sediment from the mound fill. a second large crown- type mace made of what was at first thought to be jasper was loaned to the nMnh in 1936 by Braecklein for casting (Figure 6.18b). sev- eral casts were made, and there are two in the collection (377378, 377379). the original was later turned over to trowbridge and entered the nMnh collection more than two decades later (423198). Braecklein (1936), in a letter to the smithsonian, indicates that the artifact was ?found in hematitie [sic] bed, one end next to a cremation.? this mace is an impressive piece of lithic workmanship of the shape that Brown (1996:475) refers to as type 1. this crown mace has two projections on the convex expanding distal end, one in the middle, and one at the side. Presum- ably, there was a third on the opposite corner, but this cor- ner has been fractured. there is a pair of large ?ears? on either side in the medial section of the artifact, and below these ears are two short, downward- pointing barbs. this artifact is pictured in hamilton (1952: pl. 41), Fundaburk and Foreman (1957: pl. 92), and Merriam and Merriam (2004: fig. 15). the mace is quite similar to 18/9334 of the national Museum of the american indian (Burnett, 1952: pl. xvi). the mace pictured by Burnett lacks the thorny barbs and is painted. the upper projections or ears may have served as anchors for tassels that are shown dangling from such maces in the iconography of the incised shells (Phillips and Brown, 1978: pl. 62) the tool is made from a large block of high- quality cryptocrystalline material with a waxy luster. Brown (1996:475) reports that all known examples are made of a semitransparent flint known as Kaolin chert, which de- rives from illinois. there is red ocher smeared over the piece (fortunately, it was never washed). the raw mate- rial is characterized by subtly shaded concentric bands in gray, yellowish brown, and red. one red vein runs along the axis of much of the blade. this is possibly coinciden- tal; however, in light of the treatment of the third blade described below, such an intentional design element can- not be put beyond the skills of the lithic craftspeople who made these implements. a third mace is pointed and has a pair of lateral pro- jections (423197). one longitudinal half of the artifact is deeply stained with red ocher (Figure 6.18c). this tool was either painted or rubbed with ocher. Furthermore, tiny bits of yellow residue overlay the red pigment. the line formed by putting ocher on the blade follows the axis of the piece. in shell depictions, crown maces are FIGURE 6.18. Maces: a, 423196; b, 423198; c, 423197; d, 423195. N U M B E R 4 9 ? 97 sometimes shown crossed with lines (see Figure 6.19). this mace is broken at the proximal end, so the original length is unknown. the proximal portions, or handles, of these maces are extremely narrow given the overall massiveness of the blades (table 6.13). this artifact is pictured in both hamilton (1952: pl. 44a) and Funda- burk and Foreman (1957: pl. 92). Brown has labeled this shape ?type 2.? the fourth mace (423195) is considerably smaller and corresponds to a type described by Brown (1976, 1996:476) and labeled ?type 5,? which Brown believes derives from tennessee. it has two sets of lateral projections above the shoulder (Figure 6.18d). Flaking consists of moderately well controlled bifacial thinning and secondary retouch. the dorsal ridges are smooth and appear polished, perhaps from some sort of scabbard. this mace carries no residue. it was pictured in hamilton (1952: pl. 42). there are no ground-stone maces in the nMnh collection. CElts these tools fall into two major categories, those for utilitarian or extractive tasks and those for special or cer- emonial use. they may be utilized to accomplish everyday tasks, or they may be reserved for very special occasions. some probably served as part of composite weapons. For celts, indicators of use include chips and crushing along the working edge; asymmetry of the working edge, which may indicate resharpening; and, sometimes, striations. al- though in some cases it may be unclear whether or not the axes or celts served utilitarian purposes, all celts, utilitar- ian or symbolic, will be treated in this section. For celts several attributes were recorded in addition to those noted for chipped-stone implements: width of the working edge, width of the poll, and shape of the working edge. celts vary in morphology, but they all have a work- ing edge and some form of haft device (although it may be undifferentiated from the bit, i.e., unshouldered). some celts resemble chisels, some resemble axes, and some are without analogy. celt nomenclature and measurement are shown in Figure 6.20. FIGURE 6.19. Depiction of mace on Braden B shell cup (423266). TABLE 6.13. Mace types, characteristics, and raw materials. Stem Proximal Catalog Length Width Thickness length stem width Weight Raw number Type (cm) (cm) (cm) (cm) (cm) (g) material 423195 type 5 15.5 5.56 1.33 5.46 2.73 74.0 chert 423196* Unclassified 18.25 8.18 0.95 7.58 7.3 187.6 chert 423197* type 2 28.9 10.47 2.25 6.5 4.25 540.8 chert 423198 type 1 27.6 11.58 1.52 12.38 5.47 424.8 alibates * incomplete. 9 8 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y Elongate Celts Elongate celts are, by definition, long and narrow and often have a flared or splayed bit. in cross section they are rectangular to subelliptical. there are six elongate celts in the collection, ranging in size and shape. Most resemble the celts in the collections of the national Museum of the american indian illustrated by Burnett (1945: pl. xxii h,i.). Elongate celts from the nMnh collection are listed in table 6.14. only one is flared out into a rounded bit (448670). this celt has been broken and glued back together (Fig- ure 6.21a). the pieces were likely in different deposi- tional contexts, judging by the difference in appearance of the readhered sections. the center section is more de- teriorated than the two end fragments. this long lime- stone celt is illustrated in hamilton (1952: pl. 24a), where it is incorrectly referred to as a bone spatula. celt 448670 closely resembles a celt pictured in Brown (1976: fig. 33n). the smallest of the elongate celts is a small complete artifact of jasper (448676) and was made from a pebble. the poll end retains considerable cortex (Figure 6.21d). celt 448671 has been reconstructed (Figure 6.21b). the proximal end of the blade has been rebuilt using a com- pound resembling plaster. the reconstructive material is smeared over the original stone surface in some areas. the additive contains grit and also some form of very fine white fibrous material that resembles fiberglass or synthetic fiber, such as nylon. (Bits of filler material that were removed were retained in a small vial now kept with the artifact.) Spatulate Celts spatulate celts are wide and flat, usually with convex working edges. notable is the collection of cannel coal celts (Figure 6.22a,b). there is one complete celt and bit portions of two others as well as a few small fragments. these are wide and relatively thin. the condition is poor to fair because of the softness of the material and its ten- dency to flake and disintegrate over time. such celts could only have served a special function because the bits could stand no percussive forces, lest they split and fracture. one cannel coal celt has a pronounced shoulder and deeply convex bit. two others have convex bits that flare to the sides and a simple juncture with the poll. sizes of all spatu- late celts are shown in table 6.15. there are two large stone spatulate celts. one has wide parallel sides and a biplano cross section and is manufactured of a high- quality, dense, dark greenstone or serpentine from an unknown source (448673, Figure FIGURE 6.20. celts: nomenclature and measurement. TABLE 6.14. Elongate celt characteristics and raw materials. Working edge Poll Catalog Length width Thickness width Raw number (cm) (cm) (cm) (cm) material 448676 14.68 6.30 4.36 4.58 Unknown 448670 30.20 4.74 2.12 2.60 Limestone 423201 15.54 2.75 1.93 n/a sedimentary 448672 10.66 3.20 1.75 1.98 greenstone 448671 23.50 2.17 2.02 1.91 sedimentary 448676 9.05 2.90 2.70 3.00 Jasper N U M B E R 4 9 ? 99 FIGURE 6.21. Elongate celts: a, flared bit (448670); b, reconstructed celt (448671); c, elongate celt (423201); d, small jasper pebble celt (448676s01); e, wide celt (448676s02). FIGURE 6.22. cannel coal spatulate celts: a, expanding bit (448690); b, shouldered (448675). 1 0 0 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y 6.23b). the other is a large tool, perforated in the center (448674, Figure 6.23a). it was made from high- quality white marble and was broken but has been repaired. a small missing triangular section was reconstructed by the dealer or collector. the center perforation was drilled from both sides. this celt is a morphological twin to a celt reported from the wPa excavations (Brown, 1976: fig. 179f; 1996: fig. 2- 178f); however, the tool at the nMnh is larger. Both large stone spatulate celts were shown in hamilton (1952: pl. 52). Other Celts there are four other celt fragments that either fell outside the above two categories or were too frag- mentary for certain identification. of these, three are cannel coal and probably represent parts of larger spat- ulate celts. the fourth (423200) is a small flat celt of a sedimentary stone 8.04 cm long, 4.45 cm wide, and 1.0 cm thick. it exhibits some damage to the bit (Fig- ure 6.25a). there are no monolithic axes in the nMnh collection. other WeaPonrY comPonentS other components of composite weapons are pres- ent. these include ground-stone artifacts, such as atlatl weights, and decorations. Pieces of wood that may or may not derive from weapons are included with other wooden objects in chapter 8. there are also some ceremo- nial copper covered wooden blades that are also treated in chapter 8. Four artifacts are classified as atlatl weights (448677). they are all boatstones, that is, hollowed out crescent- shaped devices (Figure 6.24). all are manufactured from what appears to be greenstone, described as porphyritic igneous material and as being altered (J. gardner, un- published notes on conservation measures taken for the spiro collections at national Museum of natural history, smithsonian institution, 1970?1971). the color is a dark olive gray. all four sit on flattened keels. the ends of the boatstones are notched with grooves ranging from 0.35 to 0.71 cm in length. they are highly polished, and sur- face cavities appear to contain residues that in two cases appear to be ocher. toolmarks are visible on the inner sur- faces of the cavity. these boatstones are shown by ham- ilton (1952: pl. 59), who reports that six such boatstones were recovered from one of the smaller southern mounds in the craig Mound group. two keeled boatstones were recorded by Brown (1976:115; 1996:465?466) for the wPa excavations from pre- harlan phase contexts at spiro. the wPa boatstones were not of greenstone as the four specimens from the relic- hunting excavations are. Boatstones are usually associated with earlier woodland occupations when atlatls were in greater use. however, Brown (1996:465) reports that they were still in use in the caddoan area until ad 1000 and may have been used for symbolic purposes after they ceased to be used directly for weaponry. TABLE 6.15. spatulate celt characteristics and raw materials. Working edge Poll Catalog Length width Thickness widtha Raw number (cm) (cm) (cm) (cm) material 448690s06 16.0 18.8 1.84 n/a cannel coal 448690s05 9.68 13.79 1.48 n/a cannel coal 448673 b 23.3 15.7 1.83 12.76 greenstone 448675 18.0 16.4 2.14 n/a cannel coal 448674 b 24.9 18 1.57 6.36 Marble a ?n/a? indicates data are not applicable. b complete artifact. FIGURE 6.23. spatulate celts: a, perforated marble celt (448674); b, large greenstone celt (448673). N U M B E R 4 9 ? 10 1 there may also be segments of wooden weapons, possibly bow or atlatl (see chapter 8 for discussion of wooden objects.). the wooden segments present are too fragmented to be clearly linked with a particular artifact form (see table 8.7). utilitarian artifactS although there are utilitarian implements, they are definitely in the minority in this collection. such artifacts most likely served needs for resource procurement and processing, or needs associated with the manufacturing of other artifacts. Both chipped-stone and ground-stone ex- tractive tools are in the nMnh collection. the collection of utilitarian tools is summarized in table 6.17. ChiPPEd stonE Hoes and Spades there are two hoe- like implements chipped from chert. these are relatively thin and finely made. one is a broken hoe from a heat- altered, good quality chert (423174s02). the other is a large, thin biface from a brown and white mottled chert (423173) that shows clear evidence for grinding or insertion into an abrasive substrate. the ridges and edges of the piece are quite smooth. it is 13.34 cm long and 6.79 cm wide. there are three massive, thick, crude spade- like im- plements that have definite shoulders and stem elements. one is made from a fossiliferous chert and is quite crude in manufacture (a large brachiopod mold is evident in one face). the source of this chert is indeterminate. it is heav- ily patinated to a deep brown surface color. two more shouldered spades are made from a dense siltstone or ar- gillite (Brown, 1976), which Banks (1984) sees as deriving from the atoka Formation. these siltstone tools (448656) measure 15.12 and 12.65 cm in length, 6.94 and 7.35 cm FIGURE 6.24. atlatl weights or boatstones (448677). TABLE 6.16. atlatl weight characteristics. Raw material is green- stone for each specimen. Catalog Length Width Depth Cavity number (cm) (cm) (cm) depth (cm) 448677s01 15.02 2.66 4.07 2.22 448677s02 16.00 2.78 3.49 2.90 448677s03 19.10 2.70 4.00 3.42 448677s04 19.80 2.57 3.97 3.16 TABLE 6.17. Frequency of utilitarian tools by type. Tool type Count Engravers 1 abraders 2 Manos 3 Polishers 2 Drills 4 hoes 2 hammerstones 3 Flakes 4 worked pebbles 5 Total tools 26 1 0 2 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y in width, and 1.83 and 2.48 cm in thickness, respectively. the smaller of the two is shown in Figure 6.25b. Drills there are four drills, each having a unique morphol- ogy. there are two large drills made from whitish cherts (423193s01, s02, Figure 6.14 a,c.). one is stemmed and has long barbed shoulders (6.96 cm long). the bit is thick, wide, and damaged. its companion in the same catalog context has a simple expanding base and a thick, wide bit (6.08 cm long). its lateral edges are heavily rounded. the third drill (423191s01) is a unifacial tool made from novaculite and has a widely expanding base. all three of these were probably hafted. the fourth is a small tool manufactured on a pebble that has been retouched to a short fine point (448655). the base of the tool retains cor- tex. only the latter tool could have been used for doing delicate work, and it would have to have been held in the fingers. Unifacial Graver one tool would have served well as a graver or incising tool. this a unifacial piece with a projection (423191s04, Figure 6.14d.). it is made of quartz crystal and so would be hard enough to use effectively on hard materials such as shell or bone. Scrapers there are three small end scrapers made from local chert. they have straight to convex working edges of ap- proximately 2 cm in width. they are all from surface col- lection done by R. King harris and are not necessarily from craig Mound contexts at spiro. the only other scraper is a unifacial obsidian scraper donated by J. g. Braecklein (378273, Figure 6.25c). obsidian tools would be unusual for spiro (no other obsidian tools are present in collec- tions from controlled excavations), and therefore, this tool is highly interesting. the tool has shallow engraving on the ventral side reading ?Le Flore Mound okla.? other scratches also occur on ventral surface. it is not impossible that obsidian trade goods could have entered the site from far away, especially later in the sequence. Records indicate that Braecklein personally collected the obsidian artifact from the east tunnel of the craig Mound in December of 1935 (national Museum of natural history, 1937), so the provenience is more or less assured. Braecklein was a care- ful and knowledgeable collector. Barker et al. (2002) used energy dispersive X- ray fluorescence to obtain the trace element signature of the obsidian. Results indicate that the obsidian derived from the well- studied Pachuca source in the state of hidalgo, north of Mexico city. this obsidian artifact is therefore the only certifiably Mesoamerican ar- tifact to be found in a Mississippian context. Flakes there are only four small flakes and one clearly bipo- lar flake. the flakes (488688) appear to be of a grayish to reddish local chert, perhaps from river pebbles, judging from the cortex on two of the flakes. FIGURE 6.25. other stone tools: a, celt (423200); b, shouldered spade (448656); c, obsidian scraper originating from the Pachuca source in central Mexico (378273). N U M B E R 4 9 ? 10 3 ground stonE Abrading Tools abrading tools fall into three categories. the first in- cludes tools for polishing narrow objects and would in- clude shaft abraders and similar forms. these would be manufacturing tools. two tools fall into this category. one is a piece of unidentified stone with one wide groove and one smaller groove in its surface (423206, Figure 6.26a). the grooves are of even width and run the length of the stone, which would be expected for shaft polishers. the other abrader has been referred to in the nMnh catalog as a sharpener (423205). it has four grooves, two on either face (Figure 6.26b). the grooves are deepest in the center and are confined to the central area of the faces. the second type includes manos, which would have been ground against other stone surfaces, presumably in food preparation. grinding clay or pigments with manos would be distinct possibilities as well. Manos often have a pit in the center of one surface (448681, as shown in Figure 6.26c). Manos in this collection are similar to those reported from the excavation of domestic areas at spiro (Rogers, 1989: fig. 73). the third type comprises those stones that are pol- ished to a high sheen on one or more faces. these were either used in fine sanding activities and became polished through use or were intentionally polished to produce a mirror- like surface. they are classified as polishing stones and could have been employed in burnishing ceramics. there are two such stones in the collection. one is a large piece of quartzite polished to a high gloss on one face (423203, Figure 6.26d). it is biplano in cross section. the other face is smoothed but not glossy. the second such stone is a black stone, which is plano- convex in cross sec- tion. the flat face is glossy. the opposite face is faceted into a roughly five- sided pyramid. these faces are not equivalent in size. the outline of the piece is circular. in the smithsonian records, the raw material has been al- ternately referred to as hematite (possible) or chalcedony (unlikely). it is a dense, black material. Hammerstones there are three quartz cobbles with extreme dam- age from battering. they are roughly equivalent in size. the damage is consistent with hammerstone use, and the overall shape and size place them in a similar class to hammerstones recovered from nonmound excavations at spiro (Rogers, 1980:166). hammerstones from the 1979 excavations were recovered in two areas, north and east of craig Mound. Worked Pebbles Four worked pebbles are present and differ in shape and material: a long, slender, broken sandstone pebble; a smoothly polished triangular pebble and a thin ovoid pebble, both of fine, dense, dark gray or black stone; and a broken plano- convex stone that appears to have been burned. the function for these pebbles is unclear. FIGURE 6.26. abrading tools: a, grooved abrader (423206); b, grooved abrader (423205); c, mano (448681); d, polishing stone (423203). o rnaments are decorative objects used or worn primarily for personal adornment or as components of clothing and headgear. such decora- tions include beads, earspools, plaques, pins, and pendants made from a wide variety of materials, including bone, shell, wood, and leather. earSPoolS one ubiquitous Mississippian ornament is the earspool. these are large ear- rings worn through extended holes in the earlobe. they are known from various Mississippian contexts, including several other sites in oklahoma, arkansas, and northern texas, with variation in style, size, and composition. Raw materi- als desired for making earspools include stone, copper, wood, ceramic, and shell. some spools are composed of two or more materials in combination, such as stone with copper foil wrapping or wood with a shell disc in the center. Both Baerreis (1957) and Brown (1976, 1996) described elements of earspools. Most are shaped like pulleys (hence the name) and have an outward facing flange that is often decorated and an inner flange that faces the neck of the wearer. they may or may not be perforated by a central bore. Earspools from the smithsonian collections fit into the style typology developed by Brown (1976, 1996). there are 53 whole or partial earspools and over 200 fragments. table 7.1 shows the distribution of earspools and fragments by style. the whole or partial count column reflects complete earspools or pieces large enough to be measured in at least one dimension. PullEy- shaPEd EarsPools Pulley- shaped earspools are by far the most common. they are character- ized by an outer flange, an inner flange, and a central core and are primarily constructed of stone. in many cases, vestiges of a copper flashing remain visible on the spool surfaces. often the core is perforated with a central bore. spools can be classified by the morphology of the flanges and the presence or absence of central bores. 7 ornaments and Decorations April K. Sievert 1 0 6 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y Perforated Pulley- Shaped Earspools Perforated earspools are common, comprising 47% of the total measurable earspool collection. in most cases the perforation is small (~1.5 cm in diameter), leaving ample surface space for decoration. Figure 7.1 shows sev- eral fragments of differently styled pulley earspools. the earspools are heavy, averaging 55.2 g per earspool for the whole and nearly whole ornaments. table 7.2 gives the dimensions for perforated earspools (see appendix a for measurement procedures). Manufacture the earspool central bore is parallel- sided and may be either smooth or ridged. a ridged bore has narrow incised grooves resembling threads within the bore. the threads are fairly even and were mechanically produced by the techniques used to drill out the stone. Both decorated FIGURE 7.1. a variety of perforated pulley- shaped stone earspools: a, complete example carved from argil- laceous siltstone (448698s01); b, large earspool carved from argillaceous siltstone (448698s02); c, fragment with narrow pulley groove (448698s03); d, fragment made of soft claystone (448698s04); e, unfinished specimen with central hole only partially drilled (448698s05); f, flanged example carved from sandstone (448698s06); g, fragment carved from siltstone (448698s07). TABLE 7.1. counts of ear ornaments (whole and fragments) by style. Whole or partial Ornament Style ornaments fragments Total Pulley shaped Perforated 31 7 38 Unperforated 7 58 65 Divided flange 6 13 19 nesting 1 0 1 central boss 2 0 2 Flanged ring 1 0 1 composite 2 0 2 other Effigy 2 0 2 wooden disc 1 1 2 Unclassified 0 132 132 total ornaments or pieces 53 211 264 total weight (g) 2,137 452 2,589 N U M B E R 4 9 ? 10 7 (448691) and undecorated (448699) earspools exhibit this feature to varying degrees. if drilling was accomplished using a hollow cylinder, such as a reed (perhaps with an abrasive), such a pattern might develop. similar ridges occur on the surface of other stone objects with wide drilled bores (e.g., pipes). the central bore may have a partially utilitarian function. By drilling out the inner pul- ley, the earspool becomes considerably lighter, thus lessen- ing the pull on the earlobe. in addition, the central bore might also facilitate removing the material between the two flanges. Placing a shaft through the bore would allow the spool to spin as if on a lathe. another manufacturing characteristic appears on the inner faces of the flanges where fine striae radiate out from the center of the pulley. this must be a result of grinding out the pulley. Pulling a cord or thick thong around the pulley could result in similar marks, especially if abrasives were used. the interior surfaces of the pulley shank are smooth, without obvious toolmarks. Raw Material the complete and nearly complete earspools are man- ufactured from either limestone or a fine- grained siltstone or sandstone. Baerreis (1957) and Brown (1976) refer to the latter as sandstone, and indeed, there is a fine crystal- line graininess visible in the surface of this stone. however, the grain size is very small. conservator Joan gardner noted that this raw material may be argillaceous siltstone on the basis of examination by geologist a. segovia from the University of Maryland in 1974. it is brown, usually in the 10YR hue range using the Munsell soil color chart (Munsell color co., 1973). in some areas the surface has become exfoliated as layers of the raw material flaked away. Exfoliation might be expected if the stone is, indeed, a siltstone. the stone is tough, and surfaces in general ex- hibit very little weathering, also typical of siltstones. this stone accounts for 13% of the total number of earspool fragments. this percentage may seem insignificant except that this material makes up 44% of the earspool collection by weight because the stone is durable and so many of the spools made from it are unbroken (table 7.3). the other commonly used stone is limestone, a white or light gray soft and fine- grained variety. in general, the limestone has fared poorly, and many fragments show weathering and leaching of the surface. several lime- stone spools are undecorated and without the residue that would indicate that they once had a copper foil covering. Limestone has a much greater tendency to be broken than does the siltstone. it is softer, more friable, and highly af- fected by acid. Associations copper foil is a common decoration for earspools. it occurs on both decorated and undecorated spools. Bell (1984:234) notes that the perforated stone earspools found at the harlan site are not decorated but are covered with copper foil. he notes also that they become larger in diameter through time. Brown writes also that the non- incised stone earspools are associated with harlan phase contexts and suggests that through time the inner flange becomes larger in proportion to the outer flange (Brown, 1976:279). Decorated earspools are found throughout the spiro phase. similar styles of decoration are shown among the earspools depicted in Burnett (1945: pls. xi?xii). Most of the earspools from the nMnh collection will be from spiro or late spiro phase burials, as suggested by the patterns recorded for earspools from the wPa ex- cavations (Brown, 1976:303, 1996). the distribution of perforated pulley- shaped earspools is wide, and they are found across the caddoan area (Brown, 1996:566). these earspools cease to be used in later caddoan sites, and at spiro they were in use primarily during the spiro ii and spiro iii periods (Brown, 1996:566). Decoration the majority of earspools are smooth. however, some are decorated with incised designs. Designs registered on perforated pulley- shaped earspools include both geometric and figural representations. Brown (1996:563) links the geometric designs with spiro ii burials and figural designs with spiro iii. Decoration occurs on the outer face only. TABLE 7.2. characteristics of perforated pulley-shaped ear- spools. all measurements are in centimeters. Standard Measurement (count) a Mean Extremes deviation thickness (31) 1.67 1.18 / 2.32 0.25 outer flange diameter (29) 6.63 3.27 / 9.06 1.61 inner flange diameter (23) 5.58 2.43 / 7.95 1.6 Pulley diameter (27) 3.36 1.43 / 4.2 0.7 hole diameter (28) 1.67 0.53 / 3.39 0.52 wall thickness (29) 0.48 0.3 / 0.68 0.1 a numbers in parentheses are the number of artifacts having this measur- able component. 1 0 8 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y Perforated spools are often covered with copper foil and decorated with an incised pattern of geometric forms. the copper foil would be carefully pressed into the relief de- signs carved on the surface of the flange. concentric circles occur on four examples of whole or partial rings (Figure 7.2). there are two earspools that are nearly identical in form and decoration. these could easily represent a pair. only one pattern occurs with any regularity. a chev- ron pattern repeated in four quadrants of the spool is present on four of the specimens from nMnh and was common in the wPa collections (Brown, 1976: fig. 58p; 1996: fig. 2- 119p). in three cases the chevrons are nar- rowly incised lines that radiate out from the central bore. there are four repetitions of the chevron pattern. in one case, the space between the lines in the chevrons is greater, thus effectively dividing the field into quarters. artifacts 448691s01 and 448691s02 represent a pair, with narrow incised lines forming the chevrons. the surface is with- out copper (Figure 7.3a,b). artifact 448692 is a similar earspool with wider incised lines and remnants of the orig- inal copper foil intact (Figure 7.3c). in most cases the decoration is incised into the surface, but in a few cases, it is excised. the single earspool from spiro known to depict a human figure (423162) is from the trowbridge collection, and it is excised (Figure 7.4a). the figure appears in low relief, with detail added by incising the sandstone surface. the lower body of the figure is de- picted in unusual form in that legs are not shown. Rather, the body terminates in what resembles a flared skirt. this might represent a desire to impart avian characteristics to the figure (a bird?s tail feathers), or the artist may simply have run out of room for legs. the head is shown in profile with the nose transformed into a hawklike beak. the eye is a simple diamond shape. Elbows are bent with hands upraised. in this particular case, the central bore seems not to be incorporated into the overall design. instead, it TABLE 7.3. summary of earspool styles by raw material. a dash (-) indicates that no earspools occur in this raw material category. Siltstone/ Style Style Limestone Sandstone Wood Copper Other total Pulley shaped Perforated 17 15 6 - 1 39 Unperforated 59 - 6 - - 65 Divided flange 16 3 - - - 19 nesting 1 - - - - 1 central boss - - 2 - - 2 Flanged ring - - - - 1 1 composite - - 2 - - 2 other Effigy - - 2 - - 2 Disc - - 2 - - 2 Unclassified 50 8 - 74 - 132 Raw material total 143 26 20 74 2 265 total weight (g) 1,131 1,209 187 24 38 2,589 FIGURE 7.2. Perforated pulley- shaped earspool with concentric circle decoration (448693). N U M B E R 4 9 ? 10 9 FIGURE 7.3. Earspools with chevron decoration: a, b, pair of earspools without copper foil (448691s01, 448691s02); c, earspool with chevron design and copper foil (448692). FIGURE 7.4. Perforated pulley- shaped earspool: a, human warrior figure design (423162); b, petaloid design (423163). 1 1 0 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y perforates the figure. this spool is illustrated in hamilton (1952: pl. 81). Brown (1996:566) refers to the figure as a standing warrior. human figures are rare on earspools and have been noted for only the norman and wilbanks sites (Brown, 1996:567). there is one complete earspool bearing a six- lobed double petaloid line around the central bore (423163) (Figure 7.4b). this sandstone earspool also has one ring incised close to the perforation and is pictured in hamil- ton (1952: pl. 81). it has no exact match in the other col- lections known to be from spiro. Brown (1996:566) refers to a similar motif as a double- line meander. traces of the copper facing remain. there is excellent information as to how earspools were worn. Perforated earspools are illustrated as adorning the ears of people depicted on engraved shells from spiro and on shell gorgets from across the southeast. Usually, the motif is executed by incising a circle and then starting a drill hole in the center of the circle, resulting in a per- fectly round dot. this motif appears frequently elsewhere in spiro art, for example, as decoration on the bodies of serpentine forms or as the eyes of zoomorphs. whether the earspool design reflects an iconographic formula or the iconography reflects a style of dress is debatable. Alteration several earspools also show evidence for having been reconstructed after excavation. artifact 448698s16 ap- pears to have been reconstructed with plaster or a simi- lar filler. two sections of the outer face were rebuilt and smeared with a green pigment to mimic copper staining. Unperforated Pulley Earspools Unperforated earspools are similar to the perforated form, except that they do not have holes drilled through the center. as shown in table 7.1, there are seven unper- forated pulley- shaped earspools. however, six of these are of wood, leaving only one unperforated earspool made of limestone. this earspool is incomplete and consists of the flanges of a limestone earspool that have been cemented together with copper. the unperforated stone earspool in this collection is not decorated. this contrasts with the unperforated stone earspools recovered by the wPa exca- vations, which generally have decoration excised in shal- low relief. Many of the fragments of stone earspools recorded as being from unperforated earspools may actually be from the perforated or divided flange types, but without a trace of the inner flange, this is difficult to ascertain. there is a fairly large number of such fragments recorded (n = 58), and they are all limestone. it is possible that they may be from the outer flanges of the broken earspools recorded as having divided flanges or from perforated earspool flanges. there are four matching pulley- shaped wooden spools with flanges of unequal widths (448902, 448903, 448904, 448905s01). they could represent two match- ing pairs. in each of these four spools one flange is wide, markedly concave, and extremely thin. it carries a narrow lip. this surface is not finely finished, and toolmarks are visible on this face in every case. the opposite flange is narrower in diameter, thicker, and excised to form a dish- shaped indentation. with the exception of 448904, which may have once held a shell disc, there is no evidence that another component was set into the shallow hollow space. if 448904 ever held a shell disc, it is not there now, and the earspool is otherwise so similar to the other unperforated wooden spools that it seems justifiable to classify it with them. the earspools are relatively plain and do not appear to have been covered with copper. a third pair of plain wooden spools differs from the other two pairs only in that the flanges are quite narrow (448905s02, 448905s05). al- though these are fairly thick, the absence of a noticeable flange makes them appear more like wide plugs. this pair is marked by flattening and deformation, so the stylistic characteristics may simply be harder to discern. the un- perforated wooden earspools have a mean weight of 7.98 g. they average 1.6 cm in thickness, 4.15 cm in outer flange width, 3.56 cm in inner flange width, and 3.4 cm in pulley width. two of the unperforated wooden earspools have traces of fiber wrapping around the pulley (448903 and 448904). Divided Flange Earspools Earspools with divided inner flanges are common (Figure 7.5). the divided flange is characterized by two equally sized, semicircular halves that are separated by a sharply defined channel (see Baerreis, 1957:32). this type may be prototypical to the later Foster type earspool in which the inner flange is reduced to two lugs. surfaces of the divided flange earspools are usually concave. there were no earspool fragments of the Foster type from the nMnh collection. the divided flange earspools are badly fragmented, with none being complete. the 11 limestone fragments in 448699s03 represent a minimum number of six earspools. Residues are plentiful, and nearly every fragment is stained with copper. one earspool has a well- adhered fragment N U M B E R 4 9 ? 11 1 of mica on the outer surface (448695s02), suggesting that mica may have been an alternative to copper for decorat- ing earspool surfaces. in addition, there are three flange sections having twilled basketry impressions on the sur- face. the impressions appear to be similar, and it is likely that they were from similar depositional contexts. average dimensions for divided flanges are as follows: thickness, 1.67 cm (n = 4); outer flange, 5.77 cm (n = 1); inner flange, 5.45 cm (n = 3); and pulley width, 3.41 cm (n = 5). Central Boss Earspools one pair of wooden spools with a central boss is pres- ent (448900s02 and 448905s04). these have the inset inner flange and a raised mound on the outer flange and were clearly covered with copper foil. Faint toolmarks around the outer flange of 448900s02 suggest that a tool with a working edge width of about 0.45 cm was used in shaping these. its probable mate is slightly dam- aged around the perimeter. the mean dimensions of these earspools are as follows: thickness, 1.89 cm; outer flange width, 4.66 cm; inner flange width, 3.6 cm; pulley width, 3.46 cm; and weight, 12.45 g. Composite Earspools composite earspools are manufactured from more than one element (Figure 7.6). For example, a wooden or stone spool may possess an inlaid disc of shell or stone. the two examples of complete composite earspools are wooden with inlaid shell, but they do not constitute a matching pair. the first is a wooden pulley with both flanges intact (448900s01). the inset size of the outer flange is 1.8 cm in diameter; the inner flange inset size is 2.1 cm. the shell disc in the outer flange is present. there is nothing in the shallow depression on the inner flange; however, there is a perforation on the perimeter of the inner flange as well as slight breakage. there are traces of fiber wrapped around the pulley. wooden earspool 448905s03 was also inlaid on the outer flange with shell, and it is stained with copper. it is embellished with small bumps that have been excised from the outer surface. there are six extant bumps, and there is room for two more in the missing, broken portion of the outer flange. this earspool carries only a faint impression that the inner flange was excised. this is a thin earspool pulley compared to other wooden forms. Nesting Earspools one small limestone earspool appears to be the male portion of a nesting ring earspool (448699s02). nesting spools consist of two parts that fit together, one flange nesting inside the other. this fragment is 1.03 cm thick and has a pulley width of 1.4 cm. Brown (1976:278) re- ports that this variety was also found in the burial mound at the harlan site. FIGURE 7.5. Divided flange earspools: a, 448695s01; b, 448695s02. FIGURE 7.6. composite wood and shell earspools: a, 448900s01; b, 448905s03. 1 1 2 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y Flanged Ring Earspools the single earspool that could be called a flanged ring is ceramic, not stone (448696). it displays a wide central opening and very narrow flanges. the paste is tempered with shell. it is complete and in good condition (Figure 7.7). othEr EarsPools Earspools not shaped like pulleys are much less com- mon. Furthermore, most of these are manufactured from something other than stone. Wood Disc Earspools one complete wooden disc is present (448901). it is flat except for a raised portion in the center of the disc. copper and a trace of matting or woven surface are vis- ible. the width of the fiber is not clear but seems fairly wide (0.5 cm) and indicates contact with basketry or matting. there is a perforation in the flat posterior face that is centered, approximately 0.2 cm deep, and appears to be drilled. there is also a fragment of a wooden disc (448909). Exactly how they were worn or used is unclear. Effigy Earspools the only effigy earspools are a pair of carved cedar canine heads (448895s01, 448895s02). tongues hang out as if the animals were panting (Figure 7.8, 7.42a). the figures have exaggerated snouts and excised eye patches (which might have held an insert, although there is no resi- due in the depression to attest to this) and display small, rounded upright ears. two pairs of small holes flank each ear and may have formed a means of attachment. Resi- due clogs these holes. this pair was shown by hamilton (1952: fig. 30a). the only other effigy earspool reported to be from spiro is known only from a photograph published by Brown (1996: fig. 2- 121d). it depicts the head of a bird of prey with shell inset eyes. Shell Rings and Discs shell rings or discs may have been utilized for pur- poses other than to decorate ears or earspools. however, given that inlays for earspools are often of shell and that shell earspools are a known type, any shell discs and rings are discussed here in the context of earspools. shell discs may either be planar or concave. whether these may have served as parts of composite earspools, figures or other artifacts, or as independent objects is unclear. there are 14 shell discs. all are well smoothed. to- gether they weigh 153.1 g. Five are perforated in the cen- ter, drilled straight through from one face to the other. Diameters range from 2.88 to 6.0 cm and average 4.34 cm (standard deviation, 1.03 cm). sizes do seem to cluster, with one size measuring close to 4 cm in diameter and a larger size measuring nearly 6 cm. Most are concave, exhibiting the natural curvature of the shell wall from which they were cut. signs of manufacture include scour- ing marks, especially on the inner surfaces. FIGURE 7.8. canine wooden effigy earspools, side and frontal views (448895). FIGURE 7.7. ceramic flanged ring earspool, both sides (448696). N U M B E R 4 9 ? 11 3 one is incised with a star design reminiscent of a seven- point Moundville star (448908). around the inside of the perimeter is brown prehistoric adhesive. copper foil remains in some of the scallops around the star. the star would have been manufactured of shell and glued into the center of another object, probably a wooden earspool (now gone). the outer surface would then be covered in copper foil carefully cut to fit the scalloped pattern en- graved into the shell disc. the result would be a flashy copper ornament with a lustrous central star. beaDS Beads are numerically the largest component of the spiro collection, and shell is by far the most common ma- terial used to make them. shell beads are present strung on the original prehistoric cordage, piled into lots (and some- times restrung on modern string), or as a component of other composite artifacts. Most of the beads are combined into lots. at the time of original cataloguing some beads were apparently sorted by type, while other lots were not. trowbridge had his beads strung in many cases, perhaps to display them better. Most of these were not resorted when they were accessioned, and trowbridge?s original groupings tended to stay together. the combinations of beads in trowbridge?s lots probably depended primarily on the circumstances of his acquisition?what he received together, he kept together. in some cases, several kinds of beads (primarily the small varieties) were mixed into large groupings that included other artifact types as well, espe- cially quartz and galena. shEll bEads Manufacture shell beads can be manufactured in three ways. (1) the columella of univalves (which in some cases can be extremely thick) can be used as raw material. shell thick- ness is greatest in the columella and is ideal for making large beads. Beads manufactured this way often display the groove of the mollusk?s anal canal. (2) Beads can be worked from the outer surfaces of either univalve or bi- valve mollusks. in these cases, original curvature of the shell may be visible. (3) Finally, whole shells can be used. For small shells such as the miniature species of Olivella, the apex may be ground down, leaving a circular hole. the bead can then be strung using this hole and the natural opening at the distal end. there seems to be no simple rule regarding which bead types are drilled from both ends and which are drilled straight through. the smallest disc beads are nearly always drilled from both sides. this may prevent breakage as the rim gets progressively thinner with deeper drilling. Beads with a round cross section are always drilled from both ends. some elliptical and convexo- cylindrical beads are drilled from one side only. some beads have relatively equal bore holes on each end and a straight bore, suggest- ing that the drill used was a reed or other similar material. overall, most beads have a biconical or contracting bore, suggesting the use of a stone drill. some beads were altered by heat. Brown (1996:576) be- lieves that burned beads were not typical of the litter burials in the great Mortuary because few other objects found with the large quantities of beads were burned. Burned beads would be more characteristic of the earlier burials that were redeposited for the construction of the great Mortuary. Beads had a wide use in personal decoration. neck- laces, hair ornaments, ear ornaments, bracelets, and anklets were prevalent and are identified on the shell ico- nography that depicts human figures. Styles Beads come in a wide variety of styles, as illustrated in hamilton (1952: pl. 66). a number of bead shapes are represented in the nMnh collection, as shown in Figure 7.9. Beads from the nMnh collection are not split into quite so many types as done by Brown (1976). For ex- ample, for purposes of this report, beads made from the columella are included with other beads of the type. Fre- quency and size range for beads are shown in table 7.4. Disc Beads the overall shape is round and flat, with relatively straight sides. there are four substyles. 1. For very small beads with flat or slightly convex sides, the diameter of the hole occupies a substantial pro- portion of the total diameter. these are included in the ?small? category in table 7.4. 2. Very thin discs (~1 mm) are distinguished by a low thickness-to-diameter (t/D) ratio. they are listed as ?thin discs? in table 7.4. these discs were noted in only two catalog contexts (423331 and 448721) and were al- ways mixed in with small disc beads and small convexo- cylindrical beads. the bore is relatively small, and the size is uniform at approximately 1 cm in diameter. one such bead was sculptured along the edges. 1 1 4 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y 3. Larger discs are similar to small disc beads except that the diameter is relatively greater, especially as related to the diameter of the hole. these correspond to the ?me- dium? and ?large? categories in table 7.4. 4. concave disc beads, although disc- shaped, do not lie flat (Figure 7.10). Disc beads number over 10,000, making them the most common bead type present in the collection. Fur- thermore, this number does not include beads in the com- posite artifact 448934, a basket containing leather, beads, and copper. it was not feasible to count the beads nestled among the other artifacts in that basket. Flat Circular Beads these are also disc shaped; however, the bore enters the bead from the side rather than the face. of these, there are only two (448735), and they are large (see table 7.4). thickness averages at 1.5 cm. the center bore is 0.82 cm in diameter at the surface and is biconical. the surfaces are well smoothed but have weathered. Elliptical Beads Elliptical beads are much less common. Beads are clas- sified as elliptical if their diameter exceeds their length and if the outline forms a fairly smooth curve. Elliptical beads in small, medium, and large categories are present. they are similar to disc beads in style, but the rounded outline implies greater investment in finishing. Convexo- Cylindrical Beads these are the common ?barrel- shaped? beads men- tioned often in the literature. they are less numerous than disc beads at a total of over 2,300. Length exceeds diam- eter. there is variation in the sizes here as well, but they do seem to cluster into three major categories. FIGURE 7.9. Manufactured shell bead types in the national Museum of natural history collection (after Brown, 1976:345). N U M B E R 4 9 ? 11 5 1. small beads are <1 cm long. Mean bead length in this group is 0.79 cm, and mean bead diameter is 0.55 cm. 2. Medium beads generally fall between 1 cm and 2 cm in length. 3. the largest beads are made from the columella of Busycon shells. Usually quite long (>3 cm) they almost always display the anal canal from the mollusk. this cat- egory includes all of the large columella beads and some other large convexo- cylindrical beads that may or may not be from columellas. Sculpted Beads this style is another variation on the convex- sided beads and includes any beads that were modified fur- ther by incising extra detail on the surface. Making these would involve added time expenditure. sculpted beads come in two sizes: small (<~1.5 cm) and larger (>2 cm). there are very few. the small ones comprise two pearl beads that were treated with extra incisions around the circumference (423337s02). the large beads are columella beads in which either the canal spiraling down the length of the columella is embellished or extra grooves are added. there are two large beads of this type (423323s01 and 448768s02). Asymmetrical Convex (Bowed) Beads these beads are convex in outline (or somewhat hex- agonal) but are bowed when viewed from the side and dis- play natural shell surface sometimes on both faces. they appear to have been made from either distal portions of TABLE 7.4. Frequency and characteristics of shell beads. a dash (-) indicates data are not available or not applicable. Weight Length (cm) Width (cm) Style Size Count (g) Maximum Minimum Maximum Minimum Disc small 10,377 1,423.1 0.67 0.11 0.78 0.31 Medium 135 284 0.92 0.29 2.85 0.67 Large 79 209.4 1.00 0.42 2.62 1.87 thin 162 10.4 0.18 0.07 0.83 0.68 Flat circular - 2 76.2 4.33 4.2 4.64 4.27 convexo-cylindrical small 2,130 651.7 1.22 0.24 0.94 0.25 Medium 237 299.7 2.50 0.71 1.54 0.52 Large 45 410.1 13.73 1.72 2.67 0.94 Bowed - 196 1,123.3 10.06 2.39 0.94 1.57 Drop-shaped - 3 3.69 2.07 1.80 1.62 1.33 sculptured Pearl 2 - 0.93 0.69 0.54 0.53 columella 2 39.0 8.25 4.65 2.10 1.82 spherical small 107 51.8 1.35 0.25 1.27 0.33 Medium 9 45.0 1.75 1.18 1.96 1.23 Large 180 2,121.8 3.58 1.90 4.22 1.15 Elliptical small 47 15.4 0.92 0.35 0.79 0.28 Medium 28 27.3 1.04 0.32 1.95 0.40 Large 2 14.9 1.10 0.90 2.24 2.17 Bulging - 124 176.6 3.44 1.02 2.27 0.62 concave disc - 64 421.6 1.20 0.43 3.10 1.75 concavo-cylindrical - 143 168.1 2.60 0.82 1.24 0.48 cylindrical Medium 5 15.9 3.65 2.30 1.26 0.54 small 584 201.0 1.25 0.40 0.69 0.37 tooth - 10 28.9 2.62 0.22 1.42 0.62 hexagonal - 2 9.9 2.90 - 2.80 - Mask - 1 96.3 5.46 - 5.50 - Pearl spherical - 238 43.3 0.74 0.31 0.80 0.32 natural (baroque) - 186 - 2.87 0.24 2.78 0.39 Mixed - 321 46.9 1.10 - 0.90 - Olivella - 6,592 486.3 1.8 0.65 0.85 0.22 Oliva? - 1 21.4 6.07 2.70 - - Marginella - 15 2.6 0.98 0.82 0.68 0.50 1 1 6 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y columellas or from the thickest parts of the body of shells (usually distal). this style is perhaps what Brown (1976) refers to as ?large elongate? beads. convexo- cylindrical and large elongate beads might represent a similar concep- tion of bead morphology. the difference lies in the con- straints of the raw material. the convex outline sometimes resembles a lengthened hexagon (Figure 7.15). these have also been referred to as ?spindle- shaped? beads. Convexo- Cylindrical Drop- Shaped Beads these beads are flattened at each end and are signifi- cantly narrower at one end so that the bead assumes a drop shape (448723s04 and 448728s01). there are three such beads, and all of them are broken along the axis. Spherical Beads Length equals diameter for spherical beads (Figure 7.11). they can be somewhat irregular and not precisely round when viewed from the end. this irregularity stems somewhat from the difficulty of working with shell. Larger spherical beads are almost always made from the colu- mella (thickest part) of large univalves, primarily Busycon. smaller spherical beads can be made from other parts of shells, including thick wall sections sometimes found in distal areas of the whelk shells. Beads cluster into three general size categories, as shown in table 7.4. Cylindrical Beads cylindrical beads are nearly straight sided and are longer than they are wide (Figure 7.12). this grouping posed a problem in that in many cases within lots of simi- lar beads, the beads graded from being clearly convex in outline to being cylindrical. in such cases the beads are not separated into smaller groups because it seems that the variation is due more to variation in manufacturing than to a different ?mental template? used by the craftsperson to manufacture the beads. Concavo- Cylindrical Beads these beads are concave in outline, constricted in the center, and flared outward at the ends of the bead (Figure 7.13). spool shaped, they resemble human phalanges, and en masse they look very much like small bones. this re- semblance to bone may not be an accident. the shape is similar to that of bones as presented in the iconography of shell engraving (see Phillips and Brown, 1978:151). Brain and Phillips (1996:361) believe this spool- type shell bead may prove to be a useful chronological marker, as it ap- pears to be linked to late prehistoric and protohistoric contexts. FIGURE 7.11. spherical columella beads (423314). FIGURE 7.10. concave disc beads (423313). N U M B E R 4 9 ? 11 7 Bulging (Compound Cylindrical) Beads these interesting beads are a variation on a convexo- cylindrical theme in that they have constrictions near each end, giving the impression that they are composed of three distinct sections (e.g., 378258s01). Usually, the wider cen- ter section bulges distinctly outward. they are recorded as ?bulging? beads (Figure 7.14).the 124 beads of this type range between approximately 1 and 3.4 cm in length. Flattened Hexagonal Beads a low t/D ratio distinguishes flattened hexagonal beads from the asymmetrical bowed beads of hexagonal shape discussed above. one such bead (378253s01) is shown in the left center of Figure 7.15. there are only two in the collection. they measure 2.9 cm in length by 2.8 cm in width. FIGURE 7.13. concavo- cylindrical ?phalange? beads (378259). FIGURE 7.12. cylindrical beads (423316). 1 1 8 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y FIGURE 7.14. Bulging beads (423327). FIGURE 7.15. hexagonal and asymmetrical convex beads (378253). N U M B E R 4 9 ? 11 9 Convex Disc Beads there are only three of these beads (423169s05, 423329s02, 423311). the curvature indicates that the bead blank was cut from a shell wall. one exhibits red ocher on the surface. Mask Beads only one mask bead is present, but it is well crafted (Figure 7.16). Bead number 448773 is realistically modeled and incised with deep V- shaped lines. the head wears no forelock. the view is frontal, with the eye treated as an oval within a forked- eye pattern. the pupil was created using a drill. teeth are raised, and the cen- ter of the mouth is missing from the surface. the person wears a band around the forehead that has three par- allel lines in the center, with sets of two lines on each side. Features appear to be rendered in craig style. the human face wears a perforated pulley earspool. the bead measures 5.46 cm long by 5.5 cm wide. a 0.66 cm diam- eter hole has been bored from either end. the channel has considerable sediment remaining inside it. this large bead weighs 96.3 g. Shell Tooth Beads these beads are shaped like teeth. they have a pyra- midal shape, with the hole drilled through the wider por- tion of the piece. they are not numerous (n = 10), but they do vary in size. included in this category is one tiny pyramidal pearl drilled through the apex. the shell beads in this group are all fairly large, with dimensions of greater than 1 cm in diameter and 2 cm in length. Gastropod and Bivalve Beads a variety of small freshwater snails are present, but in small numbers. table 7.5 lists these varieties along with their ranges, when known (Figure 7.17a). these des- ignations are based on identifications made prior to my study and recorded on the catalog cards for 378250. For 448889, identifications had been elicited from Dr. Rose- water of nMnh?s Department of invertebrate Zoology, Division of Mollusks, in the mid- 1970s. the small snails Campeloma, Angitrema, and Anculosa were undoubtedly used for beads. Bivalves include one specimen identified by Rosewa- ter as Laevicardium elatum sowerby. this is a cockle that derives from the Pacific, and its presence is interesting be- cause it suggests elaborate economic integration. cockles of the same genus are also found off the coast of Florida. another Pacific shell is a specimen of genus Haliotis, which includes abalones. this shell has well- preserved nacreous luster and appears to be unmodified. Marginella. Marginella sp. is a small ma- rine univalve (Figure 7.17b). Used for beads across north america, they are present but not plentiful in the collec- tion from spiro at the nMnh. whole Marginella shells are worked by grinding down the apex on the side of the shell obliquely to create a hole. the beads can then be strung using this hole and the natural aperture at the distal shell. Olivella. there are several species of Oliv- ella. the smallest, also referred to as ?rice shells,? re- semble small grains of rice (Figure 7.17a). Olivella are small marine univalves. Multiple species have been iden- tified in the materials recovered from the wPa excava- tions (Brown, 1983:149). there are two primary sizes present, representing a minimum of two different species. the tiny ?rice? shells rarely exceed 10 mm in length. a larger variety varies around a length of 1.5 cm. there is FIGURE 7.16. shell mask bead (448773). 1 2 0 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y one example of a larger shell (6 cm in length) that may be a different genus, perhaps Oliva. species designations for the smaller shells in the collection from nMnh are given as Olivella dama, a gulf of california species. Re- analysis of other Olivella beads in the collections from the sam noble oklahoma Museum of natural history have also turned up O. dama (Kozuch, 2002:697). this finding supports the probability that materials reached spiro from considerably farther west. Pearls Pearls are numerous (Figure 7.18), as indicated in table 7.4. in nearly all cases, pearls were not modified TABLE 7.5. Variety and range of gastropod and bivalve shells. Catalog Mollusk Count Range Type number(s) Gastropods Campeloma sp. 5 widespread Freshwater snail 378250, 448889 Angitrema salebrosa conrad 4 tennessee River Freshwater mollusk 378250 Anculosa tintinnabulum Lea. 45 tropical america Freshwater mollusk 378250, 448733 Angitrema verrucosa Raf. 1 tennessee River Freshwater mollusk 378250 Angitrema sp. 18 tennessee River Freshwater mollusk 448733 Prunum sp. 1 caribbean? Marine snail 448889 Total gastropod count 74 Bivalves Donax sp. 3 widespread Marine clam 448889 scallop? 1 Variable Unknown 448885 Haliotis fulgens Philippi 1 Pacific: oregon to Baja abalone 448776 Laevicardium elatum sowerby 1 southern california to Panama cockle 448775 Pecten sp. 2 widespread Freshwater bivalve 448889 Unio sp. 1 widespread Freshwater mussel 448889 Total bivalve count 9 FIGURE 7.17. whole shell beads: a, Olivella beads (378252); b, Marginella beads (378251); c, gastropods (378250). N U M B E R 4 9 ? 12 1 beyond drilling them for suspension. Freshwater pearls derive from bivalves from the Unionidae family. they are subdivided according to shape, with oddly shaped or ba- roque pearls grouped separately from spherical pearls. in some cases, this division was reflected when catalog num- bers had been originally assigned. the largest single pearl is a baroque pearl 2.87 cm by 2.78 cm. others are ex- tremely small, approximately 3 mm in diameter. Composite Bead Artifacts artifact 423331 consists of a modern glass jar con- taining galena in small chunks (<20 mm maximum), the small variety of Olivella shell beads, very small disc beads, quartz fragments, small cylindrical beads, and larger thin disc beads. Bits of dried clay, dirt, organic matter and bone were also present. this mass was sorted into its constitu- ents for study. Lot 448734 contained similar combinations of beads. in another case a single glass jar (423330) con- tained the small disc beads mixed with convexo- cylindrical beads. groupings or mixtures of beads are relevant to re- constructing original context. it is certainly possible that beads so grouped were actually recovered together, espe- cially in the case of 423331, which contained dirt, clay, and bone fragments along with the beads. in several cases beads still adhere to the original prehis- toric string. one set consists of seven small (0.25 cm thick and 0.7 in diameter) elliptical shell beads on an s- twist, two- ply string measuring approximately 5 cm in length (423307). a number of beads nestle among textile frag- ments (423378). trowbridge stated in his catalog that these became detached during his unrolling of textile bundles. stonE bEads stone beads are made either from a black carbonate fluorapatite (phosphate), a black or dark gray sedimentary stone (possibly slate), or a dark brown finely grained clay- stone or siltstone. a few appear to be hematite, judging from the red ocher deposit on the surface. these, however, FIGURE 7.18. Pearl beads: a, small round pearls, strung (423332); b, large baroque pearl (423340). 1 2 2 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y are manufactured as pendants rather than beads. the stone beads assume the form of thick discs, convex- sided cylinders, or spheres. apparently, the phosphate nodules are naturally spherical, occurring as nodules within shale (Brown, 1996:643), and require little modification beyond drilling. the stone is lustrous, shiny, and black. in all cases the bores drilled through stone beads are biconical. Banks (1984:78) refers to the phosphate nodules that are found in the seminole Formation along Flat Rock creek in tulsa, oklahoma, which is likely the source for the phos- phate for beads at spiro. sources for the other stone raw materials are unknown. stone beads are not common (n = 27). types represented in the nMnh collection are listed in table 7.6, and several phosphate beads are shown in Figure 7.19. the elliptical, spherical, and cylindrical beads are all manufactured from phosphate nodules. the disc beads are primarily of sedimentary stone, either siltstone or, in one case, sandstone. only one disc bead appears to be phosphate. an interesting feature is the inclusion of bead blanks. three unfinished convexo- cylindrical bead blanks are of an unclassified lithic material. Undrilled beads were certainly not strung, so they were unlikely to have been ?worn? into a mortuary deposit by the deceased. they must have been placed into the graves as a discrete offer- ing. the half- finished beads imply that bead manufacture was carried out either at spiro or nearby. CoPPEr bEads copper beads are rare. there are two types present in this collection, disc and tubular, and the disc type is prob- ably fraudulent. Disc Type the copper disc beads present in the collection of nMnh and shown in Figure 7.20 do not appear to be TABLE 7.6. Frequency and characteristics of stone beads. a dash (-) indicates data nonexistent or not available; Max., maximum; Min., minimum. Combined Length (cm) Diameter (cm) Style Count weight (g) Mean Max. Min. Mean Max. Min. convexo-cylindrical 7 35.4 1.84 0.75 2.96 1.32 1.09 1.66 Elliptical 3 26.5 1.30 1.69 1.04 1.95 2.66 1.32 spherical 6 68.3 1.65 2.03 1.16 2.11 2.47 1.72 cylindrical 3 13.2 1.59 2.20 1.21 1.04 1.23 1.57 irregular disc 1 1.6 0.34 - - 0.2 - - Disc 7 18.6 0.62 1.12 0.30 1.58 2.35 0.89 Total count and weight 27 163.6 FIGURE 7.19. Phosphate beads (448686). FIGURE 7.20. copper disc beads?apparently fraudulent (448713). N U M B E R 4 9 ? 12 3 prehistoric artifacts because they were manufactured of smelted copper.4 Brown (1976, 1996) writes that beads of this style are not known to occur in Mississippian contexts. there are 35 disc beads from catalog numbers 378266, 423216, and 448713, representing donations from Braecklein, trowbridge, and Meyer. all three collectors could easily have received beads from the same individual. combined weight is 225.5 g. Bead length ranges from 0.40 to 0.75 cm, with diameters between 1.90 and 1.45 cm. Tubular Type there are some clearly prehistoric copper beads in the collection, and these are long and tubular in form, manu- factured from rolled copper sheeting (Figure 7.21). they seem to divide naturally into four sizes. sizes and frequen- cies of beads are shown in table 7.7. they are all from lot 448714. the string may even remain inside the copper tube thanks to the preservative properties of copper. in two cases, the string is well preserved, testifying to how these beads may have been worn. Lot 448714 includes a thick cord on which three tubular copper beads and one very small shell disc bead have been preserved. another artifact in the same lot contains a long bead broken into two frag- ments that cling to the original string. PenDantS there are pendants of stone, shell, and bone. all con- tain some element by which the piece may have been sus- pended, usually consisting of holes or grooves. Pendants may have been placed with the body or included as part of composite ornaments. shEll PEndants shell is by far the most common material for making pendants, as it is for making beads. Most of these are colu- mella pendants, a type made by removing the outer shell wall of large Busycon. a few other types of shell pendant are present as well. Columella these pendants are pictured in the iconography of shell and copper artifacts. they were suspended from necklaces and rested on the wearer?s chest such that the apex of the shell is pointed downward (Figure 7.22). colu- mella pendants may be drilled through distally, or they may be left undrilled. trowbridge?s pendants have all been combined into lot 423305. he numbered his specimens individually, and for this reason, each columella from his collection was analyzed separately. objects in Meyer?s collection do not benefit from separate numbering, and because the pen- dants were numerous, these were grouped into smaller lots for which measurements were averaged. width of the columella and length of the whole columellas are useful for estimating the size of the parent univalve. the people of spiro favored large shells for cups and for making gor- gets (see gorgets in this chapter). cutting the columella away from the shell results in a large hole being left in the fabric of the shell. FIGURE 7.21. small and medium copper beads: a, three beads strung on original cordage (448714); b, loose beads, several with original cordage remaining inside (448714). TABLE 7.7. Frequency and characteristics of tubular copper beads. Length Weight Size (cm) Count (g) small 1.5?2.0 5 4.4 Medium 2.5?3.5 7 6.8 Medium large 3.5?4.5 14 17.0 Large 4.5?6.0 6 11.9 Total count, weight 32 40.1 1 2 4 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y columellas in trowbridge?s collection all appear white and chalky. there is little prehistoric residue present, im- plying that he may have cleaned them. Meyer appears not to have cleaned his specimens because they do appear to be stable and well caked with residue. also, Meyer may have treated his with something to halt deterioration. at any rate, there is a distinct difference between the two col- lections in terms of amount of residue present and condi- tion of the shell. Large lots were subdivided on the basis of the location and extent of damage. this division reflects variation in taphonomic process more than style or use. stylistically, these pendants are all similar. in most cases the distal tip is absent, and therefore, presence or absence of a drilled hole (the only stylistic difference noted by Brown, 1976) is unknown. table 7.8 summarizes dimensions and descrip- tions of the columella pendants. several types of residue occur on the columella beads. one columella (448778) accompanies two fragments of well- preserved twill- weave basketry. other columellas have faint traces of matting or mat impressions, and these combined impressions suggest that such pendants were disposed of in proximity to basketry (or matting) in the mound. Very little copper stains the pendants, so they were not placed in proximity to copper. Sculpted Columella only one of the over 90 columella pendants has been modified beyond drilling a hole in the distal end for sus- pension (Figure 7.23). Pendant 448780s01 has a design consisting of two parallel lines extending down from the apex, following the natural whorls around to the opening, and then following the shoulder to the opposite side of the opening. two more sets of these double lines extend TABLE 7.8. Frequency and characteristics of columella pendants. Dash (-) indicates data are nonexistent or not available. Total Mean length Width Weight Portion present count (cm) (cm) (g) whole pendants with distal tip 16 18.7 5.4 1,659.0 apical fragments 28 12.5 5.4 2,538.4 Distal fragments 1 4.0 - 7.4 nearly complete pendants 34 19.3 6.5 4,635.6 Lateral fragments 5 18.5 6.1 485.0 spines only 9 12.0 3.3 471.5 Total count, weight 93 9,796.9 FIGURE 7.22. cat- man warrior wearing necklace of columella pen- dants and displaying scalps (423245). FIGURE 7.23. sculpted columella pendant (448780s01). N U M B E R 4 9 ? 12 5 around the apex. the artifact has a brown crusty residue on the apex. one spot of brown residue is quite thick. in addition, a fragment of mica adheres to the apex. Other Shell Pendants several other forms of shell pendants were fabricated. one variety is made from a long, cylindrical piece of colu- mella that has been perforated at one end. it differs from columella pendants in that the apex is not present. there are six of these in the collection (Figure 7.24b). Mean di- mensions are length 7.02 cm, diameter 2.7 cm, and hole diameter 0.28 cm. the longest is 9.35 cm long; the short- est is 4.92 cm. total weight amounts to 94.2 g. holes are usually biconical; however, one such pendant has a con- tracting hole, indicating that it was drilled all the way through from one side. the second variety is a shell bauble with a groove carved around one end and no hole. this variety resembles weights or sinkers. the two present in the collection (448886 and 448727) measure 6.72 and 6.78 cm long and 0.96 and 1.18 cm in diameter, respectively. a larger type (Figure 7.25) has three incised proximal grooves (448738) and resembles an artifact shown by Brown (1996: fig. 2- 129a). the third type is a pendant cut from shell and in- cised (448769). it is shaped like a tassel and, if inverted, resembles the shape of a crown mace (Figure 7.26). it is possible that this type was meant to be worn from the ear or earspool because an identical artifact is pictured on engraved shell number 423234 (Phillips and Brown, 1984: pl. 282). a nearly identical artifact from the wPa excavations is pictured in Brown (1976: fig. 77e; 1996: fig. 2- 129). unusual shEll forms there are also some other unusual forms of shell ar- tifact. one is a small, irregularly shaped piece of bivalve wall that has two small perforations along one side. Pre- sumably, it was meant to be strung and worn as an orna- ment, in the same manner as some gorgets are worn. FIGURE 7.24. Pendants: a, bone pendant carved to resemble bear tooth (378264); b, cylindrical columella shell pendant (378263). FIGURE 7.25. sculpted shell pendant (448738). 1 2 6 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y two shell hooks are present in the collection (Figure 7.27). the flat hooks are grooved around one end for tying. another similar artifact is less curved. a similar pendant from another collection is shown in hamilton (1952: pl. 83a). in another artifact, the opening edges of shells were perforated (423302s02). the use of this arti- fact is unknown. stonE PEndants other pendants can be classified according to where the hole is placed. the most common form is that of a worked stone having a hole in one end for stringing or tying. Cannel Coal two pendants are made from this soft material. in both cases the pendant is oval (423167) or elliptical in shape (448683s04) and perforated with a single hole at one end (Figure 7.28). the perforation is incomplete on one of the pendants, with a drill start on one face only. Hematite these small pendants are made of dense, heavy stones that exhibit a powdery reddish residue on the surface. the shape is cylindrical. there are two finished pendants of this type and one that is roughed out but unperforated. these are grouped along with beads of phosphate nodules in catalog number 448686. Other Stone an odd pendant of greenstone is incised on one face (448684). this is an unusual design in that the stone is not simply drilled through the top to provide a point of FIGURE 7.27. shell hooks (447770). FIGURE 7.26. shell pendant (448769). FIGURE 7.28. cannel coal pendant, both sides (423167). N U M B E R 4 9 ? 12 7 attachment; rather it has a carved loop (Figure 7.29). the surface is well polished and exhibits a design of crossed lines. overall, it resembles the back of a beetle, with the wings outlined and the loop serving as the head. another stone pendant is perforated in the center (423166). it has an overall diamond shape and is made of sandstone. this object resembles woodland period gorgets and may be an older object that was interred as part of a reburial. Finally, an unfinished pendant is merely a pebble with the hole for suspension only partially drilled. bonE PEndants there is one broken pendant of bone (see table 8.6). it is drilled through on one end (423341). another pendant (Figure 7.24a) is composed of what appears to be antler (378264). this artifact represents a bear tooth and is iden- tical with other bear tooth pendants shown in hamilton (1952: pl. 60) and in Fundaburk and Foreman (1957). it is 5.6 cm long and well polished. this artifact is in excel- lent condition. CEramiC PEndant the only ceramic pendant is 448931. it is very heavy and stonelike but friable, and magnification shows a gritty paste. the pendant is broken in half medially. a deep groove extends around the top of the pendant. GorGetS gorgets are ornaments that were suspended from the neck to hang upon the wearer?s chest. such ornaments are analogous to the pectorals worn throughout Mesoamerica and manufactured of jade and bone. gorgets are made pri- marily from shell, usually Busycon, although there is also a bone gorget fragment in the collection. shEll gorgEts Brown (1976) classified shell gorgets into morpho- logical types on the basis of attributes of manufacture and design. the custom was to decorate gorgets on the inside surface of the shell wall, which would afford a concave surface for decoration. gorgets were less frequently in- cised with the design on the outer surface, which is quite interesting in light of the tradition of engraved shell cups, all of which were decorated on the outside wall only. the inner surface is of a somewhat different luster and smoothness. Perhaps the decoration itself could be more easily executed on inner surfaces. the inner surface is nat- urally smooth and, for fresh shells, is softer and easier to engrave. (as indicated in my experiments done to simulate engraving shell using stone tools, fresh shell is not as hard as seasoned shell, and the inner surface is softer than the outer.) there are 56 gorget fragments, representing at least 31 gorgets. the gorgets can be classified according to either manufacturing style or iconographic style. Brown (1976, 1996) devised a scheme for classifying gorgets according to manufacturing style. the portion of the shell used (inner or outer surface), the areas decorated, shape (circular or otherwise), and the medium of decoration (e.g., incising) are all indicators of manufacturing style. an adaptation of this method is used in discussing the gorgets in the nMnh collection. a summary of gorgets and masks is given in table 7.9. when Phillips and Brown worked with the shell cup designs, they were able to extrapolate the design styles to include gorgets as well as cups. Phillips and Brown (1978, 1984) refer to two major design schools, Braden and craig. gorgets correspond only to the craig school. the number of fragments in each catalog context represents the total number of fragments, including those that have been refitted and readhered. FIGURE 7.29. stone pendants: a, beetle- shaped greenstone pendant (448684); b, broken stone pendant (423168); c, diamond- shaped pendant (423166). 1 2 8 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y Brain and Phillips (1996) published a separate study of gorgets and plotted design distribution across the south- east. they refer to styles based on subject matter, orienta- tion, and other characteristics. they connect several styles to spiro: Pearce, which has a single bird, interpreted as a turkey; Jackson, which displays birds in swastika orienta- tion; hull, which has human hand motifs; and Philbrook, a birdman theme. they refer to one style that is found only at spiro as hamilton (Brain and Phillips, 1996:55?59). in the hamilton style, two human figures are arranged sym- metrically on either side of a pole and are executed within the parameters associated with craig a styling. Fenestrated Gorgets the best represented style was the fenestrated gorget. these artifacts are usually round (type 1 in Brown, 1976, 1996) but may be other shapes as well (type 3 in Brown, 1976, 1996). Portions of the design have been cut out, re- sulting in an overall lacy effect. cutting shell in this way is difficult and time- consuming; however, the resulting gor- gets are quite impressive. the contrast between the shell and the background would make them show up vividly on the wearer. Brown?s type 1 refers to artifacts with a circu- lar outline. type 3 refers to gorgets cut to conform to the outline of the design on the gorget. in other words, they are often not round. the 17 examples of circular gorgets comprise 37 sepa- rate fragments. Most of these represent the craig a style; only two examples are craig B. Four of these circular gor- gets are decorated with a modified star with concentric circles (Figure 7.30). (see Brown, 1976: fig. 81g for an identical type in the University of arkansas collection.) the style resembles a motif associated with Moundville (Fundaburk and Foreman, 1957). TABLE 7.9. Frequency and characteristics of shell gorgets by style. Style Iconographic style Symbols Count MNI a Weight (g) Fenestrated circular craig a Raccoon, concentric rings 8 6 214.4 craig B Bird, cross, dotted circle 7 2 133.6 craig unclassified humans, greek cross 19 6 b 267.7 Unclassified sun disc 3 3 26.9 other shape craig a and unclassified concentric half circles and rings, forked eye, akron grid 4 3 223.8 solid Engraved inner surface craig a and unclassified Dots, birdman 2 2 123.9 Engraved outer surface, mask craig B Forked eye 3 1 37.0 Perforated circular Unclassified none 1 1 9.2 Unmodified wall section Unclassified none 2 2 175.7 Unclassified craig unclassified none 4 3 147.3 Mask craig B and unclassified human face 3 2 189.9 Total counts, weight 56 31 1,549.4 a Mni = minimum number of individuals. b includes 448742, a gorget not located for this analysis. FIGURE 7.30. circular fenestrated gorget depicting a sun symbol (448740). N U M B E R 4 9 ? 12 9 the Pearce style (Brain and Philips, 1996) is repre- sented by artifacts 423292 and 448765, two fragments that refit. the birdman Philbrook style shows up on ar- tifact 423293. Brain and Phillips (1996:61) point out the sophistication of this design in which the wings hanging down from the birdman?s arms cross in front of the body. Examples of the bilaterally symmetrical hamilton style are 448747, 448748, 448749, 448755, and frag- ments from 448765 (see Figure 7.31). gorget 448742 depicts a man facing a forked pole, a craig c motif (see Figure 7.32). Brain and Phillips classify this as the hous- ton style. they distinguish this style, one that has a similar design to hamilton, for the simple reason that hamilton depictions are craig a, whereas this is craig c. another circular fenestrated gorget depicts a birdlike creature ren- dered in craig B style (Figure 7.33). two of the gorgets in this group were made on the reverse face of engraved shell cup fragments. one is large and highly concave and has a large hole in the center (448761s02.1). the other (448761s01.1) is made on the inner surface of a shell fragment and depicts a human holding tightly to a raccoon staff (a whole raccoon dangles from the top of the staff). the noncircular gorgets include a variety of forms. one artifact is a long, narrow object with a pattern of brickwork (448763) (Figure 7.34a). another is an un- usual gorget made on a Pleuroploca shell (Figure 7.34b). FIGURE 7.31. craig a gorget fragments: a, depicts raccoon pole (448750); b, depicts warrior (448747). FIGURE 7.32. gorget showing man facing forked pole (448742). 1 3 0 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y on the outer face is an amphisbaena- type tongue, prob- ably a remnant of an engraved shell cup design. on the inner surface motifs depict a forked eye and bird?s tail. the cutout area appears to be a part of the inner surface design. although this could be a case of reuse, Phillips and Brown (1978:vi, 1984: pl. 274) discuss the possibility that this was intended as a two sided design. a third artifact appears to be a cutout of a woodpecker?s head and tail (Figure 7.35a). it has a thick copper incrustation on one face (448762). the precise mode of wearing or using such an object is unknown (but see Phillips and Brown, 1984: pl. 338D). another gorget of this type is a reused engraved shell cup (448762s02). a large circle has been removed from the center, probably for making another gorget. Motifs. Motifs seen on these gorgets include raccoon bindings, eye- in- hand motifs, concentric rings, greek cross, forked eye, concentric half circles, chevrons, and bird and human motifs. Residues. among the residues present are brown resinous patches, brown stains, copper stains, and basket impressions. Solid Gorgets solid gorgets are much less common. there are only three, and one of these is more appropriately classified as a mask or maskette (423295). two others are solid pieces decorated on the inner surface (type 2 in Brown, 1976). one is a round object (448757) decorated on the inner surface with lines of dots. the dots were executed through a series of drill starts that are not very deep and are ar- ranged in two rows spaced 2.6 cm apart. the other (423293) shows a serpentine arm with claw and bead brickwork at the wrist, elbow, and shoulder. this is a fragment of a birdman with outstretched arm, feathers, and petaloid decoration (Phillips and Brown, 1984: pl. 148). Perforated Gorgets Perforated gorgets are decorated with numerous cir- cular perforations drilled through the piece. there is only one gorget of this type in the collection. Brown (1976) referred to this variation as type 5. artifact 423294 has two perforations on a circular gorget. the fragment repre- sents only a small portion of the whole. it is possible since this is a small piece that the perforations present were for suspending the object. therefore, it is impossible to tell if this came from a gorget with lots of holes or if these holes would have been the only two. FIGURE 7.34. noncircular fenestrated gorgets: a, pattern of brick- work (448763); b, craig B bird motif on inner surface of Pleurop- loca shell (448746). FIGURE 7.33. craig B style gorget depicting avian subject (423292). N U M B E R 4 9 ? 13 1 Unclassified Gorget one unclassified gorget is engraved on an irregularly shaped piece and depicts a Janus- faced figure whose head is surrounded by a sort of semicircular frame (448759) (Figure 7.35b). Shell Sidewall Gorgets there are two gorgets of this type (referred to as type 9 in Brown, 1976). they are manufactured by taking a section of the outer wall of a univalve and polishing off the edges. the shape is that of the shell wall, with upper and distal body areas retained. one is made from a Cassis shell and bears an inscription that reads ?tally gorget.? it is signed by Braecklein (Figure 7.36). the perimeter of this gorget has been incised with short hatched lines run- ning perpendicular to the edge (378246). on the outside are two impressions or stains that are reminiscent of cord- age. the other gorget of this type is from a small Busycon (423308s01). it exhibits some hatching on the top lip, but that is the only decoration. Mask Gorgets there are three masks. such masks may have been worn over the face, perhaps after death (Phillips and Brown, 1984: pl. 273). as such, they are not gorgets proper but are included here because they are similar in terms of analysis. one is made from the apex of a shell (448743) and might have been worn as a gorget, judging from the holes for suspension (Figure 7.37a). the impression is that of a human face. taking out the tip of the apex formed the gaping mouth. two eyes are drilled all the way through, and an incised circle surrounds these. a long raised nose extends down between the eyes. the surface is very well polished. the mouth has five incised lines extending down from the lower edge. three parallel lines zigzag out and down from each eye. odd random lines also seem to ema- nate from below the eye. Beneath the mouth on the left is a recently incised scribble. this gorget is complete. total measurements are 11.36 ? 9.08 cm. the holes at the top for suspension are 0.45 cm in diameter and biconical in FIGURE 7.36. gorget made from Cassis shell (378246). FIGURE 7.35. gorgets: a, woodpecker head gorget (448762); b, Janus- faced figure (448759). b a 1 3 2 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y shape. this is a very interesting and unique artifact, unlike anything from the wPa excavations. the second mask is a fairly large artifact manufac- tured from the distal part of a Busycon shell (448861s07). a human face is shown, done in the craig B style (Fig- ure 7.37b). the mouth is open and shows teeth that are separated (Phillips and Brown, 1984: pl. 273a). a beaded forelock is shown pulled to the left (when viewing). the nose is a long trapezoid, and the eyes resemble fish. the width of the artifact is 11.45 cm. the third is a smaller mask (width is 7.72 cm), also done in the craig B style (423295). Faint toolmarks are visible around the relief- carved features (Figure 7.37c). the piece consists of three refitted fragments. the eyes are sunken, making the nose stand out. Lips and teeth protrude, and large forked eyes frame the nose. the fig- ure?s teeth, eye decoration, and ear detail are incised. two holes on each side and at least one at the top are pres- ent. the smaller size and plentiful holes imply use on a headdress or other composite artifact (Phillips and Brown, 1984: pl. 273). traces of red ocher are present. bonE gorgEt a single bone gorget fragment is present (423343). this disc was cut from bone and then perforated in the center (Figure 7.38). two circles are inscribed around the perforation. what appears to be a circular arrangement of perforations encloses the central design. Parts of at least five perforations are visible. three more lines are inscribed FIGURE 7.37. Masks: a, whimsical face cut from shell apex (448743); b, craig B style face (448861s07); c, craig B style gorget (423295). N U M B E R 4 9 ? 13 3 around the ring of perforations. the outer edge is scalloped. the artifact has an overall finish in a dark, rich reddish brown, as though it has a patina or was affected by pig- ment. the finished size of the gorget can be estimated at 7 cm in diameter (see table 8.6). the bone appears to be the scapula of a large mammal, e.g., bison, deer, bear, or elk.5 heaDDreSSeS headdresses were composite artifacts that consisted of plumes, pins, headbands, plaques, fastenings, and beads or pendants. they often contained parts using a wide variety of raw materials, including copper, wood, bone, leather, shell, and textiles. Pins Pins were used in fastening portions of headdresses and especially in attaching copper plumes. Pins are known to have been made from copper and from bone, and both of these material types are represented. Copper Pins Prehistoric copper pins have square cross sections and taper gradually from the center to the tips. there are only six fragments of authentic pins (448714). these are broken into segments approximately 5 cm long. the set includes two tips. the lot containing these pin fragments consisted primarily of tubular copper beads, and the pin segments had gone unnoticed among the beads. these copper pins are heavily corroded and highly fragmented (Figure 7.39b). FIGURE 7.38. Bone gorget (423343). FIGURE 7.39. copper pins: a, fraudulent pins manufactured from copper wire (448712); b, authentic cop- per pins and s- shaped pieces (448714s10, s11). 1 3 4 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y six other small curved fragments were also identified from the same lot. these are s- shaped, approximately 1.0 cm wide, and 0.65 cm thick. these are also fragmented, with an average length of about 2.6 cm. the function of the curved segments is unknown, but they could also be seg- ments of pins or other fasteners. the curved pieces are all covered on one side with fragments of a coarse, straight, pale brown fibrous material that J. gardner (unpublished notes on conservation measures taken for spiro collections at national Museum of natural history, smithsonian in- stitution, 1970?1971) apparently interpreted as leather, al- though it also resembles vegetable fiber. it is highly likely that the curved segments were recovered together from the same archaeological context. Perhaps they were used to hold a garment together. Most of the artifacts listed as copper pins are fraudu- lent, a fact mentioned in the inventory of materials ac- cessioned from J. g. Braecklein (national Museum of natural history, 1937). Both Mohrman (1939) and hamilton (1952:51) address the concern over authentic- ity of copper pins. analysis performed on one of the pins showed the artifact to be of smelted copper (caL report on artifact number t1421; smithsonian institution, Mu- seum conservation institute, unpublished). these pins are cigar shaped, with consistent diameter and round cross section. none are broken or bent, a condition suspicious in itself because most of the copper artifacts are quite frag- mentary. they were probably made of copper wire (Figure 7.39a). apparently fraudulent copper pins in the collec- tion include the artifacts listed in table 7.10. these pins were apparently widespread, as they ap- peared in several different accessions. trowbridge?s cata- log does not indicate from whom he received the pins, but they are listed among other materials that he received in 1936. therefore, the perpetrator of the fraud is unknown, but the pins entered the artifact market not long after the commercial excavation of craig Mound (see also Mohrman, 1939). in an accession inventory of materials received from Braecklein in 1937, there is a notation that copper pins were received from the property owners at the mound. Bone Pins Bone pins are present as well (see table 8.6). these are distinguished from awls or other bone tools by a lack of wear on the pointed tips (Figure 7.40a). the copper and bone pins in the collection are nearly all broken, and this may have been part of the ritual process in depositing them in the mortuary feature. artifact 448707 is a complete pin 17.3 cm long that tapers to the distal tip and features a spatulate end. it is heavily copper stained (Figure 7.40b). Lot number 448408s01 consists of 15 fragments of bone that refit into nine pin fragments. seven distal pieces are included. Eight more fragments of tapering pins are pres- ent in 423324 and 379214. these are narrow, straight, and round to elliptical in cross section. several are heavily stained with copper. the most massive of the tapering type is 1.05 cm in diameter (379214). the other type of bone pin or implement is grooved on one or both surfaces (448708s02). these are noticeably wider and have a concavo- convex cross section. six frag- ments refit into five sections. their use is not clear because of the grooves and greater width of the implements. in the nMnh catalog, all of these are referred to as awls. TABLE 7.10. apparently fraudulent copper pins in the collection. Catalog number Count 378268 2 397719 2 423217 4 448712 15 542527 to 542531 5 Total pins 28 FIGURE 7.40. Bone pins: a, fragments (448708); b, complete pin with spatulate end (448707). N U M B E R 4 9 ? 13 5 two artifacts may, indeed, be awls or some kind of piercing implement (Figure 7.41). they have very sharp oblique tips. one has been made from the os penis of a raccoon, Procyon lotor, and the other is a tibia, probably also from a raccoon. the tibia tool has two worn patches along the body of the piece. hEaddrEss PlaquEs and PlatEs Copper sheet copper is known to have been used for mak- ing copper plumes, which were attached to headdresses, and plaques, which comprised the frontal portion of some headdresses. headdress plaques are thin plates with nu- merous perforations for the attachment of bindings or other ornaments. some of the copper foil materials in the nMnh collection could easily be from headdresses. one set of fragments consists of two pieces with an out- line done in a reverse scallop. several other fragments ap- pear to match these, although they do not refit (423221). copper sheet artifact fragments, which usually cannot be differentiated between headdress parts and free- standing plates, are discussed and shown in chapter 8. Brown lists four types of headdress plumes from cop- per that are all designed to mimic feathers. these would have been either lashed to a pin with cordage or other fiber threaded through pairs of holes along a pin track or fastened by the pin using a riveted copper piece (Brown, 1976:311). Plumes often exhibit a dorsal offset near the base. they are often but not always decorated in repousse patterns of parallel lines representing feather markings. in this collection, no pieces can be called plumes unambigu- ously, although there is one fragment that may possibly have come from a plume. catalog number 448716s02 contains a sheet copper artifact with one pair of perfora- tions flanking a groove, presumably for attachment of the pin. the fragment has a repousse design. Bone there is also a small plaque of turtle carapace with several small perforations (448898). it has a parabolic shape and is decorated with parallel incised lines (see table 8.6). although it is by no means certain that this is indeed a fragment from a headdress plaque, it is similar enough to be included in this section (Figure 7.42c). another bone plate consists of two fragments (423221s03). in the collection, this is included in a lot con- taining copper fragments. it is a thin circular disc of bone with a hole in the center. the diameter when complete would have been at least 12 cm. the doughnut- shaped item has small holes around the periphery and around the center hole. on one side it has been reverse painted in a design of concentric circles (see Figure 7.42d). this artifact resembles a bone plaque shown and discussed by Brown (1966:553, fig. 2- 112b) from collections of the gil- crease institute. that plaque has a cross and circle motif painted on the center and concentric rings. it has only a small perforation in the center, not the large hole implied by the artifact in the smithsonian collection. cone ornamentS the final artifacts considered in this chapter are five cone- shaped wooden ornaments (448889). these have a raised rim around the outer wide edge and are hollowed for about one- half of their length. they have become warped and squashed (Figure 7.42b). in all cases a hole has been drilled through the apex of the cone, out through the hollow in the back of the cone. the raised rim is per- forated in four spots equidistant around the perimeter. in all but one of these ornaments there are two depressions, one on either side, about halfway between apex and rim. these are not exactly opposite, but fall in one hemisphere FIGURE 7.41. Bone awls, pins, or piercers: a, sharpened os penis from a raccoon (379212); b, mammal bone tool (379213). 1 3 6 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y of the piece. although there is no residue in these spots, they could represent eyespots in some sort of stylized zoo- morphic form. Dimensions for the cones are summarized as follows: length from the apex to the rim ranges from 6.97 to 7.4 cm, and diameter ranges from 5.5 to 6.43 cm. Mean weight is 29.3 g. the use of these cone artifacts is unclear. Because they have holes in the perimeter, they could have been sewn or tied to some other object. noteS 4. these were analyzed at the smithsonian?s Museum con- servation institute, according to notes on the conservations cards, but the report number is unknown. 5. identification by R. Purdy, Department of Vertebrate Pale- ontology, smithsonian institution, 1975. FIGURE 7.42. other ornamental objects: a, effigy earspools (448895); b, zoomorphic wooden cones (448899); c, turtle carapace plaque (448898); d, bone plate plaque. Plates the 11 plates that follow show photographs of selected engraved shells to illustrate a range of engraving styles. Many fragments not previously published are included. in the images, numbers following the letter ?a? are the assigned catalog numbers for the artifacts shown; number labels beginning ?ant? no longer have relevance and should be disregarded. in the captions, an asterisk (*) indicates an object not in Phillips and Brown (1978, 1984). codes within square brackets (e.g., ?[s07 ]?) are the catalog iden- tifier subnumbers used to identify fragments in table a.1 and appendix B; num- bers in parentheses refer to the plate number(s) in Phillips and Brown (1978, 1984). 1 3 8 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y PLATE 1. Row 1: 423253, craig c cup with snake and talon motifs*; 423343, bone gorget. Row 2: 423238, craig c cup with man holding serpent staff, lateral view*; 423252 (on right), Braden c cup with bird in flight and possible armadillo*. Row 3: 423238, dorsal view; 423253, apical view. N U M B E R 4 9 ? 13 9 PLATE 2. Row 1: 448822, craig c cup with forked pole theme*; 448839, craig c cup with human figure, disembodied heads, and bow, lateral view. Row 2: 448839, dorsal view; 448740, type 1 gorget with sun circle; 448744, type 1 unclassified gorget. Row 3: 448840, craig c cup with pear-shape motif and bands; 448843, Braden a cup lip fragment with serrated border*. 1 4 0 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y PLATE 3. Row 1: 448877, 448880, and 448816, fragments in two views. Row 2: 448877, left, [s07], craig B cup with bilobed arrows motif; right (cluster of three), [s11], [s05], and [s03], sun circles and snake motifs. Row 3: 448877, from left, [s04] Braden a cup with concentric circles (34.5); [s05] craig B with snake man theme (222Da); [s02] unclassified (a2); [s03] Braden (B2c); and two snake fragments. N U M B E R 4 9 ? 14 1 PLATE 4. Row 1: 448860 with refits from 448880, Braden a with human heads and bilobed arrows (12); 448867[s03], unclassified cup with snake, forked eye, and circles motifs. Row 2: 448867 (4 fragments), craig B and unclassified snake and bird themes; 448869 (5 fragments), craig style wing motifs. Row 3: 448869, left (2 fragments), [s03], craig B wing motifs; center (3 fragments), craig feather and plume motifs; far right, [s02], craig a feathers and barred oval. Row 4: 448869, far left, [s01]; 7 fragments to right, craig feather and plume motifs. 1 4 2 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y PLATE 5. 448866. Row 1: left, 2 fragments; center, apical view of craig B fragment with snake (241a); right, 3 fragments, center of which is [s05] craig (218aa). Row 2: left, fragment; center, [s01], apex; right, [s06], Braden a (8aa). Row 3: left, 5 fragments with cross-hatching; center, feather motif; right (2 fragments), [s10], Braden a. Row 4: left, [s11], craig B cup (242); right, [s12], Braden a amphisbaena (28a). N U M B E R 4 9 ? 14 3 PLATE 6. 448858. Row 1: left, [s01], craig B; center upper, [s02]; center lower, [s05], human figure wearing columella pendant; right, 4 frag- ments. Row 2: [s06]; [s07] (2 fragments), craig a; [s09]; birdman theme. Row 3: left, [s10] (2 fragments), craig B; [s11] (3 fragments); birdmen in various styles. 1 4 4 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y PLATE 7. Row 1: 448871, left, [s01] (4 fragments), craig cup with bison; center, [s03] (2 fragments), Braden a*; right, [s02] (2 fragments), Braden B. Row 2: 448871; 448876 (2 fragments); 448877 (2 fragments). Row 3: 448877[s04] (4 fragments), concentric circle motif. Row 4: 448877, left, [s08], craig B cup with bilobed arrows; right, [s07], craig B cup. N U M B E R 4 9 ? 14 5 PLATE 8. Row 1: 448880, left, craig a, 3 distal and body fragments (168B, -c); right, [s29], Braden B, 2 fragments (96a, B-7). Row 2: 448880[s34], 38 fragments, including (B-3J, -h, B4Bb) mostly Braden (most not in Phillips and Brown, 1978, 1984). Row 3: 448880: left, 8 fragments; center, [s40] (2 fragments), Braden a-B miscellaneous (B-4g, -E); right, [s40] (2 fragments including a-sk). Row 4: left, 515786; center, 448881 (not labeled), 2 fragments with basketry impression*; right, 448882, amphisbaena cup with basketry impression*. 1 4 6 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y PLATE 9. 448858. Row 1: left cluster, same as Plate 6, top right; right, [s03], human male figure*. Row 2: with 448880, 2 refitted fragments shown in two views. Row 3: left, distal fragment (278b); center, craig B (195a); right, [s04], Braden a atlatl; far right, small fragment. Row 4: left, [s11], 8 fragments showing human legs (163E craig a is second from left); right, 4 fragments (2 refitted) including human figures and framed figures. N U M B E R 4 9 ? 14 7 PLATE 10. 448866. Row 1: left, [s01], 4 fragments (lower left is 96.13); right, [s09], 2 fragments, craig (198hb). Row 2: from left, [s03]; [s04], 2 fragments; [s05], craig (224b); [s08], Braden (52a). Row 3: center, [s09], small fragment craig piasa. Rows 4, 5: upper and lower left, [s19], cup with spider and rattlesnake, craig B (242); upper and lower right, [s12], Braden a cup (28a). 1 4 8 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y PLATE 11. 448853. Row 1: left, 4 fragments; right, with 448867[s02]. Row 2: from left, [s03], [s04], and [s05], human hands on apical view. Row 3: [s05], dorsal view of human hands. R itual equipment encompasses objects thought to be associated with ceremonial or special circumstances. activities surrounding the use of these materials may be social, religious, or both. such equipment would include pipes, figurines, copper plates, and discoidals. in addi- tion, other objects?raw materials, crystals, pigments, or minerals?might also connote ritual or special activity. this chapter is organized into three sections, the first being devoted to clearly manufactured artifacts such as pipes and figu- rines. the second part deals primarily with worked materials of uncertain func- tion. the third covers unworked materials, including pigments, minerals, and other miscellaneous but interesting items. some of the materials discussed in this chapter might fit more easily into the sections on decoration and ornamentation in chapter 7. however, without clear linkage to parts of costumes or headwear, fragments of copper, wood, bone, and leather can just as reasonably be considered here. manufactureD obJectS PiPEs Pipes are considered to be among the most special objects used in north america. Pipe making is a recognized art, with skilled pipe makers working with carefully selected materials. some of the pipes represented are self- stemmed pipes, meaning they lack a separate stem. this type includes the numerous t- shaped pipes. other pipes are composite artifacts in which the user inserts a smoking tube into the pipe to act as a stem. all of the effigy forms fall into this category. Pipes having shapes other than effigy are listed by style in table 8.1. Effigy pipes are treated separately in table 8.2. T- Shaped Pipes these pipes resemble an upside- down letter t, with the base forming the crossbar and the bowl forming the upright. complete pipes have a bowl that 8 Ritual Equipment and ceremonial Materials April K. Sievert 1 5 0 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y rests on or near the center of a long, cigar- shaped stone base. one side will be drilled for use as the pipestem. the opposite end remains solid (in most cases). the nMnh collection contains complete pipes, pipe fragments, and an unfinished pipe of this type. the pipes are usually made of relatively soft stone, including the siltstone, sandstone, and steatite found in the nMnh collection. Brown (1996:507) also lists he- matite, shale, and limestone among the locally available hard stone materials used for t- shaped pipes. it is possible that siltstone, as identified here, is a material that Brown would refer to as either a sandstone or possibly shale. the ?siltstone? is a very finely grained homogeneous material (based on small grain size, rather than on petrographic analysis). t- shaped pipes are divided here into the three catego- ries employed by Brown (1976, 1996:505): contracting bowl, cylindrical bowl, and chandler. they constitute typical Mississippian artifact types and can be described using three fragment categories: bowl, stem, and tail (stem projection). Differences in the types of pipe usually reflect differences in the morphologies of these three pipe sections. Chandler T- Shaped Pipes these pipes have stems and projections that are un- equal in both cross section and length. they are probably the earliest form (Brown, 1996:505). there is one com- plete chandler pipe in the collection (448657), shown in Figure 8.1. it is made from fine- grained sandstone. in cross section the stem is round and the tail is a flattened ellipse. the surface is polished. Brown residue is caked onto the sides of the bowl, which exhibit shallow ridges TABLE 8.1. characteristics of stone and ceramic pipes by type (except effigy). all measurements are in centimeters. the abbreviation ?t? indicates t-shaped; a dash (-) indicates that data are nonexistent or not available. Catalog Bowl Bore Bowl number Type Element Length Width Height Diameter diameter depth length Material 423156 Elbow complete 5.81 2.92 4.22 2.32 0.71 2.15 2.37 siltstone 379215 Red River stem 6.84 0.76 - - 0.35 - - clay 423154 t cylindrical or Bowl 8.76 4.61 8.7 3.22 1 5.6 - siltstone/ chandler sandstone 448657 t chandler complete 15.28 3.32 8.29 2.38 0.85 5.44 6.75 sandstone 448658 t contracting complete 32.2 3.89 7.57 2.66 0.54 5.35 14.8 siltstone or pipestone 448660s02 t contracting Bowl and base 8.68 4 6.32 2.82 0.58 4.05 - siltstone fragments 423157 t cylindrical complete 25.2 3.5 7.95 2.72 0.84 5.25 10.73 siltstone 448659 t cylindrical complete 20.1 3.6 7.5 3.27 1 4.82 8.78 siltstone 448660s01 t cylindrical Bowl and base 20.7 4.28 9.46 3.48 1.05 6 - siltstone fragments 448660s03 t cylindrical Bowl and base 16.6 4.23 7.82 3.45 0.91 5.95 - siltstone fragments 423158 t cylindrical complete 40.8 3.95 10.1 2.99 1.44 6.41 19.5 siltstone 423159 t cylindrical or complete 29.4 4.06 9.49 5.2 - 5.8 11.7 siltstone contracting (siliceous shale) 423155s01 t unclassified tail 10.68 2.47 1.83 - - - - siltstone 423155s03 t unclassified stem 7.31 1.66 1.67 - 0.68 - 7.31 siltstone 423155s04 t unclassified tail fragment 4.72 3.74 2.16 - - - - siltstone 448661s02 t unclassified Bowl and tail 4.65 2.58 5.35 1.98 0.45 3.99 - clay 448661s03 t unclassified Bowl base and tail 7 3.12 3.33 2.28 0.15 - - clay 448661s04 t unclassified Bowl, half - - - - - 3.89 - clay 448661s05 Unclassified Bowl fragments - - - - - - - siltstone 448662s01 t unclassified tail fragment 4.95 3.25 2.31 - - - - Limestone 448662s02 t unclassified tail 7.9 2.51 2.34 - - - - steatite 423155s02 Unclassified stem 6.59 2.28 2.1 - 0.58 - 5.33 clay 448661s01 Unclassified Bowl only - 3.58 - 2.18 - 3.9 - clay 417938s05 Unclassified Bowl fragment 3.73 3.66 1.22 - - - - sandstone N U M B E R 4 9 ? 15 1 from the drilling process. there is also a bowl of another pipe that could come from either a chandler or cylin- drical bowl pipe (423154). the fragment consists of the entire bowl and a short piece of the tail. Material is a very finely grained siltstone or sandstone. the tail is plano- convex in cross section. the bore extends across the bowl and indents into the tail, suggesting the possibility that the bore was drilled before the bowl. there are toolmarks on the base, and the interior of the bowl is ridged, in a manner similar to the central perforation on pulley- type earspools. there is brown sediment in the bowl, a brown patchy hard and thin residue on the outer surface near the tail, and copper stains in two locations on the outer surface. the flat bottom of this pipe would allow it to sit upright. Cylindrical Bowl T- Shaped Pipes these pipes have been found in harlan and spiro phase contexts (Brown, 1976:227; 1996: fig. 2- 90h). this style of pipe has a cylindrical bowl along with stems and stem projections that are roughly equal in length and cross section (Figure 8.1b,c). the cylindrical bowl types are often found broken. the breakage may be either from intentional ?killing,? from excavation, or from damage incurred during use. Most of the collection of t- shaped pipes in the nMnh falls into this category. Four cylindri- cal bowl pipes are complete. one such pipe is an unfinished blank (423159). the pipe shape is roughed out, and a depression has been cre- ated in the top for the bowl (Figure 8.2a). From the rough, nicked surface of the unfinished pipe, it is clear that the method for creating the blank is pecking. Furthermore, the base shows that a narrow tool was applied using an adzing motion in shaping the base. this artifact correlates with a similar unfinished pipe in Brown (1996: fig. 2- 90h). the largest artifact of this type is a pipe measuring over 40 cm in total length (423158). it has incurred a break near the bowl, but both parts refit exactly, without loss of material at the point of the break (Figure 8.2b). an unusual characteristic of this pipe is that it has yel- low ocher smeared all over the surface. an ashy residue remains in the bottom of the bowl. one of the pipes in this group is incised with a de- sign on either side of the bowl. on pipe 423157 is a motif similar to that present on a pipe from the wPa excava- tions (Brown, 1976:240, fig. 45a; 1996: figs. 2- 89e, 2- 91a). this motif appears to be a circle with two legs or streamers, housed under a parabolic motif (Figures 8.2c, 8.3a, b). the motif is then repeated below, only reversed vertically. the motif may be a stylized hawk eye or ?weep- ing eye? representation (Brown, 1996:508). the design is present on both the right and left sides of the bowl. the bore contains a reddish brown residue. in the bowl the residue is thick, caked, and dark. the bore enters the bowl just slightly off center. Polishing and grinding marks are visible on the surface. the incised design displays a neat V- shaped groove, suggesting it was carved using a chipped- stone tool. the stem is round in cross section. the last complete pipe of this style (448659) has a slightly damaged bowl that sits just slightly toward the tail (Figure 8.1b). the stem and tail are the same shape in cross section. the bowl displays ridges from the boring process, but they are uneven, unlike the bowls of other cy- lindrical bowl pipes. the material appears to be siltstone, and there is sediment remaining in the bowl. FIGURE 8.1. t- shaped pipes: a, chandler pipe (448657); b, cylin- drical bowl pipe (448659); c, cylindrical bowl pipe (448660). 1 5 2 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y two other pipes of this type are broken (448660s01 and 448660s03). in each case, the bowl and portions of both stem and tail are present, but the tips are missing. these both have thick, unanalyzed residues on the inner surfaces of the bowl. Contracting Bowl T- Shaped Pipes with these pipes, the bowl contracts as it moves up from the base. only two such pipes are present in the nMnh collection. one (448658) is incised and was prob- ably painted as well (see Figures 8.3a, 8.4a). the deco- rated pipe has lines incised in a spiral around both the stem and tail and incised chevrons on the bowl. this chev- ron pattern on the bowl is reminiscent of the decoration along the rims of sanders Engraved ceramics. the surface of the pipe appears to have been painted with a reddish pigment. in spots the pigment has flaked off to show plain gray stone beneath. the inner surface of the bore is gray. there are traces of soot in the bowl. the other pipe with a contracting bowl (448660s02) is not decorated. it exhibits copper stains on the outer sur- face and thick brown residue on the inside of the bowl. a shiny film on the outer surface may be a preservative. the cross section on the tail is round; on the stem it is a slightly flattened ellipse. the bore enters the bowl off center. Un- like the other t- shaped pipes, this one appears to be made from steatite. T- Shaped Unclassified Pipes there are also fragments of tails, stems, and bowls that cannot be classified into types. overall, the cross sec- tion of such pipe stems and tails is round to plano- convex, with the majority being round. one tail fragment is lime- stone (448662s01). some residue is present in the bowls FIGURE 8.2. cylindrical bowl t- shaped pipes: a, unfinished pipe blank (423159); b, very long pipe (423158); c, pipe with incised bowl (423157). N U M B E R 4 9 ? 15 3 and bores of nearly all fragments. although in some cases this residue appears to be depositional sediment, in others it appears charred and probably represents the remains of the material smoked. two pipes having short tails are ce- ramic (448661s02, 448661s03). one exhibits a patch of red hair stuck to a broken edge (423155s04). Elbow Pipe one elbow pipe is present (423156). it appears to be made of a fine- grained siltstone and is complete and un- damaged (Figure 8.4b). inside the bore is a reddish resi- due. the bore enters the bowl off center. grinding marks are visible around the bowl. Faint red lettering reads ?c26 13? ?,? interpreted as a price of $13.00. Red River Pipe artifact 379215 is the stem from a single Red River pipe. these pipes are ceramic, with a narrow, straight, polished stem and a bowl that sits squarely on one end. there are two sections that refit for a total 6.84 cm in length. the diameter is 0.76 cm. Bore diameter is 0.35 cm. the paste is a grog- tempered clay that registers 2.5YR 4/4 reddish brown on the Munsell color chart. there are no bowl fragments. Presumably, the stems are manufactured by rolling clay around a narrow reed. FIGURE 8.3. Decorated t- shaped pipes; a, painted and incised contracting bowl pipe (448658); b, incised cylindrical bowl pipe (423157); c, incised pipe. FIGURE 8.4. stone pipes: a, contracting bowl pipe (448658); b, elbow pipe (423156); c, cat- linite effigy pipe (423152). 1 5 4 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y Unclassified Pipes Various portions of other ceramic and lithic pipes, in- cluding bowl fragments and stem tips, are present as well (Figure 8.5). one bowl fragment appears to have a projec- tion, resembling a lug, on the side (448661s05). the rim of the bowl is incised with hatch marks (Figure 8.5f). a shell- tempered ceramic bowl fragment (448661s01) also exhibits hatching around the rim (Figure 8.5a). two stone cone- like fragments included in an artifact lot containing various pipe fragments appear at first glance to be pieces of pipe bores (448662s03). however, closer ex- amination reveals that they have an internal structure con- sisting of rays extending out from a center. they are broken longitudinally along the center axis. these are more likely fossil corals or perhaps some crystalline mineral. the outer surfaces do appear to have been intentionally smoothed. Effigy Pipes there are effigy pipes manufactured to represent both people and animals (and one that crosses this distinction). Dimensions for effigy pipes are shown in table 8.2. all of the effigy pipes are carved of stone, which is ground and polished to a smooth outer finish. in several cases, the bottoms of the pipes are decorated. Because every side of the piece may be visible during use, it comes as no surprise that they often have decorations on all faces, including the bottom. Zoomorphic Effigy Pipes there are three pipes made in the form of animals. two resemble a feline (possibly puma), and the third FIGURE 8.5. ceramic and stone pipe fragments: a, ceramic (448661s01); b, stone (448661s02); c, ceramic pipe, top view (448661s03); d, ceramic pipe, side view (448661s02); e, ceramic pipe bowl fragment (448661s03); f, stone pipe bowl fragment (448661s05). TABLE 8.2. characteristics of effigy pipes by style. all measurements are in centimeters; a dash (-) indi- cates that data are nonexistent or not available. Catalog Bowl Bore Bowl number Style Length Width height diameter diameter Depth 423152 canine 17.70 2.32 4.92 1.76 0.98 2.92 423160 crouching man 8.70 5.05 8.87 - 2.47 - 448663 Pipe holder 10.72 6.66 13.49 2.75 2.37 3.10 448664 Pipe holder 10.20 8.24 14.1 2.99 2.35 2.95 448665 two-headed figure 9.42 8.27 10.00 3.30 3.00 3.80 448666 Feline 14.70 5.82 7.00 3.30 2.47 3.40 448667 Feline 10.40 7.80 8.40 8.90 3.05 4.20 448668 human?bird 18.50 15.00 16.00 4.20 3.30 3.60 448669 crouching woman 18.50 14.00 24.00 4.45 4.00 4.85 N U M B E R 4 9 ? 15 5 represents a canine. the tail of the feline on pipe 448667 (Figure 8.6) curves up and around the bowl, terminating near the head, in a pattern similar to that described by Brown (1976:247; 1996:513) for two puma pipes from the wPa excavations. it appears that the feline crouches over or straddles a human form. the piece is carved on all surfaces, including the bottom (Figure 8.7c). the view of the bottom clearly shows a headless human form in out- line. the person?s head appears to wrap around the front of the pipe to face out below the head of the cat. the bowl is in the form of a round, lipped pot similar to the bowl of 448663, and it sits squarely in the center of the animal?s back. the stem extends out from behind the animal, so the effigy would not face the smoker. spirals are incised onto shoulders and knees. Lines are also incised at elbow and ankle joints. horizontal incised lines decorate each side of the body. the animal clearly has rounded paws with incised claws and was illustrated in hamilton (1952: pl. 17D). this pipe is manufactured from a highly crystalline stone of uncertain composition, perhaps bauxite or sphal- erite, as suggested by notes accompanying the artifact. sediment remains in the crevice around the bowl. the second limestone zoomorphic pipe (448666) is in poorer condition than 448667. Featured is a deteriorated rendition of a quadruped (Figure 8.8). the animal seems FIGURE 8.6. Feline effigy pipe (448667). FIGURE 8.7. Effigy pipe bases showing design detail: a, 448665; b, 448663; c, 448667. FIGURE 8.8. Feline effigy pipe (448666). 1 5 6 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y to be seated on its haunches but raised up slightly on fore- legs. it could represent either a feline or canine stance. the head faces forward but is not raised. the overall slinky pose suggests cat. the stem exits from the animal?s hind- quarters, and the bowl rises from the animal?s back. the surface is badly eroded from chemical decomposition of the limestone. it is very friable, unlike other raw materi- als used for pipes. although it is possible to discern that the surface was once ground smooth, most of the original surface is gone. Dark sooty residue is present in the bowl. Pipe 423152 is of an effigy style that would be irregu- lar for caddoan pipes (Figure 8.4c; hamilton, 1952: pl. 22B). it has the red color and soft texture typically ex- pected from catlinite (most notably deriving from Min- nesota sources). the bowl sits well forward, right behind the head of a canine animal with an open mouth and well- carved tongue. teeth are shown along the mouth open- ing. the canine head has drilled eyes and is also drilled through the ears. the body of the pipe has alternating ex- cised diagonal bands and constricts at the butt. a rattle- snake is incised on the bottom. the surface is coated with a patina, and on the smoking end a lighter shade of red shows through. the style is more often associated with later contexts (griffin, 1952:91), and although there is no reason to doubt J. g. Braecklein?s report concerning its provenience as the craig Mound, it very likely represents a depositional episode much later than that which created the great Mortuary. trowbridge (1958) records that John hobbs excavated the pipe in november 1935. hobb?s wife sold it to Braecklein, who in turn sold it to trowbridge on 11 December 1935. the bowl was incised with concentric rings. it was originally broken and repaired. the reconstructed portion has been painted to resemble the original surface but is not incised. the patched area appears to have been filed with a metal file, leaving toolmarks on the catlinite near the joins. these file marks are also found in spots on the bowl. they appear as minute, perfectly parallel sets of shallow incised lines. this was the first pipe from spiro to find its way to the smithsonian institution. in 1936, Braecklein loaned the piece to the smithsonian for casting, and there are three casts of this pipe in the collection. the pipe was returned to Braecklein and was later acquired by trowbridge. the original was accessioned along with the remainder of the trowbridge collection. this pipe has no prehistoric residue; however, paint and some form of filler were used in reconstructing the bowl. the repaired surface is outlined with a thin yellow line. trowbridge (1958) re- cords that the repair was done by the smithsonian institu- tion at the time they made casts of it in 1936. Human Effigy Pipes six finely worked limestone pipes are present. these include some styles that duplicate those in the collections from the wPa excavations and others that do not. Crouching Man. the first (423160) is a rather small pipe that depicts a man holding his arms crossed in front, with his hands resting on opposite shoul- ders (Figure 8.9). Feet are modeled in two different ways on the two sides. on one side, the foot is tucked underneath the leg, and just the toes and distal ball of the foot show. on the other side the foot is twisted up on the outside of the body and is shown in profile. the individual?s facial fea- tures are distorted in a grimace. Features are modeled in low relief. the pipe is broken, and part of the top of the pipe is missing. this pipe was shown in hamilton (1952: pl. 17a). no culturally deposited residue appears to remain. Crouching Woman. Pipe number 448669 is a very large limestone artifact (Figure 8.10). it depicts a human wearing beaded side locks and a double back hair bun. the figure is squatting. the stem exits at the rear, and the bowl opens at the top, behind the head. the mouths of both the bowl and stem bores are flush with the surface. this large pipe stands over 24 cm high. the head is 8.8 cm wide, 8.0 cm high, and 9.5 cm deep. the figure has two small breasts at the front and probably represents a woman. her hands rest on her knees. the back hair bun is tied around the middle and modeled by a raised band with two horizontal incised lines. the forehead shows a clear case of frontal flattening and deformation. at the ear is an oddly shaped motif done in bas- relief. it resembles FIGURE 8.9. crouching human effigy pipe: a, side view; b, front view (423160). N U M B E R 4 9 ? 15 7 a round- tipped stemmed projectile point with a square notch on the right lateral edge of the motif. Beaded side locks are done in relief down either side of the head. the body is blocky and spare, with the exception of rounded buttocks on the figure?s back. there is some crusty residue in the bowl. copper is present on the base, and ocher can be seen inside a depres- sion on one side. Perhaps this was originally painted. a large black stain on the base suggests that someone set this down on a dark fluid that seeped into the stone. some striae from grinding are visible, and the stone was incised using a sharp fine tool that left V- shaped grooves. abra- sion from use or excavation is visible on the front left lower side. Pipe Holder. two pipe holder effigies are present. one human effigy pipe (448663) of limestone portrays a kneeling figure with the legs folded underneath the body (Figure 8.11). arms are outstretched and hold up a pipe with a large round pipe bowl. the stem issues from the abdomen, so that the pipe faces the smoker. the head is slightly turned to the right side. the modeled bowl is held slightly to the left side so that the bowl is not pre- cisely above the stem. the round bowl has an indented, standing flat rim. a large nose and open mouth are vis- ible on the pipe holder. Ears are shaped realistically. Long hair or a headdress flows down the back of the head and extends to the waist. the original surface is visible on the posterior base, where the buttocks are clearly modeled. the original surface is visible on the outer posterior bowl, an area that may have been protected. Between the but- tocks on the base is an incised triangular form that might represent genitalia (Figure 8.7b). Manufacturing traces are mostly obliterated by weathering, and the limestone is badly leached. sediment and sooty residue line the bowl. the bore is clogged with residue. some brown residue is also present on the posterior face. this pipe was pictured by hamilton (1952: pl. 13) and resembles one from the collection of the gilcrease institute and shown in Brown (1996: fig. 2- 101c). the only difference between the pipe in the nMnh collection and that at the gilcrease is the position of the pipe holder?s legs, which in the case of the latter, are extended. another pipe, 448664, is similar stylistically to 448663, except that the arms encircle an elbow pipe with hands clasped in front (Figure 8.12). as with 448663, the stem extends out from the abdomen such that the smoker faces the figure. the figure is raised up on his lower legs so that there is a space between the legs on the base. Feet FIGURE 8.10. human effigy pipe (448669): a, side view; b, oblique view front; c, oblique view rear. FIGURE 8.11. human effigy pipe, pipe holder style (448663): a, oblique view front; b, side view. 1 5 8 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y are turned inward with the toes pointing toward the spine. the figure wears its hair pulled back into a bun. the head is tilted backward so that it extends beyond the line formed by the spine. Body length is 8 cm. this artifact is weath- ered, and no trace of the original surface remains. any traces of manufacturing have been obliterated. the only residue on the pipe is soot in the very base of the bore. T wo- Headed Figure. Pipe number 448665 is a two- headed human limestone figure (Figure 8.13). the heads are positioned side by side rather than back to back as they are in a pipe from the wPa excavations described by Brown (1996:526, fig. 2- 101a,b). the two- headed ef- figy from the wPa excavation has double torsos, whereas the one from Meyer?s collection at the smithsonian has a single body only. the stem extends out 1 cm from the posterior surface. the bore perforates the figure from the anterior to posterior surface. toolmarks extend down into the bowl. Eyes and mouths are simple slits and are not rendered with much detail. the right head has crude fea- tures incised with the eye, extending into the nose. the ears are crudely defined with an incised curvilinear line. in each case the right ear is clearer. Knobs on the backs of the head are probably hair buns. the left head has been broken off but readhered. the figure kneels with knees together. there is one hand on either side of the bowl, and these are broken away. the right wrist has two parallel lines incised as bracelets. Below the knees there are incised lines running parallel around the leg. Five lines encircle the right leg, and six deep and two shallow lines can be seen on the left leg. the widest point is at the heads. the user of this pipe would face the back of the figure. on the base, the feet are shown by extending the leg carving using an incised line. the heels point up. the buttocks are clearly shaped, and a groove extends onto the base, where it terminates in an incised circle (Figure 8.7a). the right buttock has two incised concentric ring motifs. the left buttock has an incised foot but no other decoration. hamilton (1952: pl. 16a) illustrates this piece sans the left head. there is some smoke residue on upper surface of inner pipe bowl. Human?Bird. artifact 448668 is a large pipe that is part bird and part human (Figure 8.14). the right side shows a crouching eagle with wing and incised tail. its leg rests on its foot, and the talons are clearly incised. the bird leg wears an incised bracelet. the head is a rounded knob incised with an oval design that termi- nates at the chin. the left side depicts a kneeling, crouch- ing human wearing bracelets. the hand rests on the knee, and five digits are clearly incised on the hand. the foot is tucked under the body. one buttock is incised around onto the posterior surface and flanks the stem opening. the opening is flush with the surface on both stem and FIGURE 8.13. two- headed human effigy pipe (448665): a, oblique view front; b, front view. FIGURE 8.12. human effigy pipe, pipe holder style (448664): a, front view; b, side view. N U M B E R 4 9 ? 15 9 bowl. an incised line connects the oval on the head to the bowl. another incised line rings the neck. a human shoulder is clearly modeled. the base has two diagonal incised lines extending from front right to posterior left. the effigy does not face the smoker. the bird wing and leg extend farther forward than the human shoulder and knee. the juxtaposition of bird and human reflects the birdman theme often seen on engraved shells. the bird is clearly a raptorial type. Both V- shaped and U- shaped grooves are visible. hor- izontal toolmarks are present in the bowl. the limestone is similar to that used for 448665. this is a large, heavy pipe and would be somewhat unwieldy to use. copper extends onto the human side of the piece, on the dorsal surface and on the posterior rim of the bowl. Very little residue is vis- ible in the bowl. Red residue (ocher) is present only on the base and is quite faint. the pipe was shown in hamilton (1952: pl. 8). no pipes of this type were recovered in the wPa excavations (Brown, 1996:514). griffin (1952:90) wrote that he did not believe the composite human- eagle effigy pipe shown by hamilton to be authentically made by spiro?s inhabitants. it is unclear if griffin is referring to this pipe or to the pipe showing an eagle crouching over a human (hamilton, 1952: pl. 6). the crouching eagle rep- resentation is not an uncommon one for spiro and is re- corded for the wPa excavations, assuming that griffin did refer to the human- bird composite pipe in this collection, which shares overall characteristics of manufacture with the large crouching woman pipe, 448669. also, given the frequency of human- bird representations in spiro?s artistic representations, a human- bird transformation such as this one seems reasonable from an iconographic perspective. in addition, the copper and other residues are consistent with other stone pieces from the craig Mound. disCoidals gaming stones and related paraphernalia are expected for caddoan sites. the most obvious indicators of games or competitions are ?chunkee? stones, large discoidal arti- facts used in the prehistoric game that involved hurling or bowling a stone disc. Four such stones are present (Figure 8.15). of these, two are miniature and would likely have been too small for use in actual games. one is a large, smooth, biconcave artifact of a red metamorphic stone, possibly quartzite. another is of a smooth, fine- grained black sedimentary stone. one of the miniatures is made of marble, and the other is made of a dark gray sedimentary stone. sizes are shown in table 8.3. three of the four are biconcave in cross section, whereas the remaining one is biplano (448676). the depths of the concavities on each side are variable. none exhibit clear traces of utilization. FIGURE 8.14. oblique views of human- bird effigy pipe (448668): a, left side, the human; b, right side, the bird. 1 6 0 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y chunkee stones are, if utilized, expected to exhibit some damage to the outer perimeter. CoPPEr copper has been discussed to some extent in chap- ter 7 under the topic of headdress parts. the discussion is continued here to include certain other lots of copper and composite copper artifacts. the nMnh collection contains some 1,582 g of copper (excluding pins and beads). in most cases, the copper is highly fragmentary or folded several times. Fragments total 382, and many are quite small (<1 cm maximum dimension). Because cop- per was folded and corroded into masses, it was generally not possible to distinguish among what might have been copper plumes, headdress plaques, and falcon plates that have been reported for spiro elsewhere (hamilton, 1952; Brown, 1996). however, other characteristics, such as presence of perforations or rivets, were noted and may en- able some gross interpretations along these lines. a sum- mary of sheet copper artifacts is shown in table 8.4. For sheets, it is difficult at best to ascertain the char- acter of the parent artifact from squashed and folded fragments. simple style designations such as perforated, embossed, or unclassified are used to refer to types among the sheet artifacts. in subdividing lots having numerous fragments for analysis, size, thickness, and degree of fold- ing and crumpling are also used as criteria. FIGURE 8.15. Discoidals; a, 448678; b, 448679; c, 448680s01; d, 448680s02. TABLE 8.3. Discoidal characteristics and raw materials. Catalog Diameter Length Raw number (cm) (cm) material Notes 448676 12.89 6.3 Quartzite Large but unused 448679 11.3 3.18 sedimentary no depression 448680s01 5.88 2.18 Marble thin in center (0.4 cm) 448860s02 6.58 2.98 sedimentary has a lip N U M B E R 4 9 ? 16 1 Brown (1976:413) and hamilton (1952) discuss cop- per plates and copper- covered wooden plaques. here Brown observes that copper artifacts are made in two different ways. Plates are stiff, self- supporting sheets, whereas plaques are made of a carved wooden backing covered with a thinner copper foil. it seems likely that copper plumes would also be made to be thicker and self- supporting. the plates can be either figural or geometric. From the wPa excavations, well- preserved copper plates were found with twilled basketry, implying that the copper was buried in baskets (Brown, 1996:546). the copper from commercial excavation in the great Mortu- ary was recovered broken, crumpled, and folded (Figure 8.16). this copper was demolished prior to burial. the highly fragmentary and crumpled condition of the copper pieces in the trowbridge and Meyer collections is consis- tent with their recovery in the great Mortuary by the relic hunters. Despite being damaged, the copper still retains basket impression, suggesting it was also deposited in bas- kets or on mats. artifact 423227s01.1 could represent a portion of a hawk plate, although the head present on the plate resem- bles a woodpecker more than a raptorial bird. this artifact has been X- rayed in an attempt to reconstruct more of the design. this attempt was largely unsuccessful. hamilton (1974) records significant patching to plates done with rivets; therefore, rivets should be present among the copper pieces from the nMnh collection. Figure 8.16b shows a piece that has numerous rivets. Rivets would also be useful in providing ways to attach the components of TABLE 8.4. sheet copper artifact fragments. Style Size (cm) Count Weight (g) Comments Embossed square plate 10.4 ? 6.5 1 16.5 448715s07; partial square Embossed fragment <2.0 ? 3.0 176 166.2 448715s01, s02; 448718 423221s04; 423222 <3.0 ? 10.0 30 332.5 448715s03, s04, s06, s09, s10; 448737s01.1 Embossed bird motif 12.0 ? 9.25 6 161.1 423227s01.1; woodpecker 8.0 ? 5.0 14 244.3 448715s04; badly folded and corroded Embossed motif 5.0 ? 5.10 1 4.2 423218s01 nonembossed 16.0 ? 15.0 8 81.4 448716s01 Riveted 10.0 ? 5.6 5 82.3 448715s08 Perforated 14.0 ? 7.3 5 22.5 423221s02 Riveted 6.15 ? 5.9 1 8.1 423221s01 Unclassified <2.0 ? 3.0 118 193.3 423225s01; 378267; 448715s05, s11; 448737s13; 448934s06; 423218s02; 42319 >2.0 ? 3.0 3 29.6 423223; 448717; 423220 composites 12.0 ? 9.25 6 161.1 sheet with pendants, basketry 423227s01.1 3.9 ? 2.57 1 7.6 Fragment with disc shell bead; 423226s01.1 11.1 ? 4.55 1 61.4 Distal end Busycon on sheet; 448737s02 10.0 ? 7.0 6 112 Fragments with fiber and shell beads; 423224s01.1 Total count and weight 382 1,684.1 FIGURE 8.16. copper sheet artifacts: a, crumpled embossed frag- ment (423218); b, fragment with numerous rivets (423221s01). 1 6 2 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y copper artifacts. Brown (1996:547) remarks that rivets would be expected more for objects that would undergo stress, perhaps as a function of being worn during ceremo- nial or dance contexts. therefore, rivets will be expected more for costume components than for discrete plates that were not parts of other artifacts. copper has been instrumental in preserving organic materials. copper artifact 378267 is a lot containing bits of copper plating. one fragment adheres to a fragment of leather 2.1 ? 4.2 cm in size. the leather is thin and brown and very soft. Lot 423224 contains a group of copper fragments, one with a very long frayed bit of basketry fiber adhering to it. there is also copper with basketry covered in two colors of clay, green and brown. the basketry fiber is very friable and splits easily, so the entire piece appears brushy. Many of the other copper fragments display bits of basketry or even cloth that have been preserved by the action of copper salts in reducing bacterial and fungal growth. some copper pieces are undergoing decomposi- tion. in one case, 423220, a residue that resembles soap is present and may be cuprous chloride, but this residue has not been analyzed. shell beads are stuck to the copper in some cases. ar- tifact group 448737 contains copper plates with beads at- tached. convexo- cylindrical, disc, bowed, and columella beads are present. figurinEs Figurines are functionally somewhat different from other decorated objects. their function is for display and representation. they cannot be smoked, worn, or filled. there are five figurines representing humans, and these are listed in table 8.5. Raw materials vary, as do sizes and styles of representation. Wood Brown (1996:531) reports that several wooden human effigies were recovered, and these came primarily from the relic- hunting episode that struck the great Mor- tuary deposit. wooden statuary is thought to have been a standard feature in mortuary contexts. there are three wooden sculptures depicting human males in the collec- tions that Richard Meyer donated to the smithsonian. the figures are seated. one figure (448891) is attached to a tenon (Figure 8.17a), which probably served to anchor the sculpture (hamilton, 1952: pl. 25). the eye spots are excised and probably had inlays at one time. total length including the tenon is 50.5 cm. the buttocks are defined. the head has two slots in the center and right of center (the left side is damaged). a deep incision runs around the head behind the ears. the right ear is realistically rendered. a dent in the abdomen could signify the navel. the hairline is clearly carved as well. the right arm has separated at the shoulder. Residues include glauconite, copper, and a basket fragment that adheres to the face (cheek). although the right side is well preserved, the left side shows evi- dence of fire damage. Perhaps it was burned prior to or as a part of a mortuary ritual. the piece was burned enough to damage the head and destroy the left arm and leg. another figure is represented by a body (448890) and a separate but refitting head (448896). the head (Figure 8.17b) became detached and is a slightly different color than its body; however, even with warping, these clearly refit (and are shown together in hamilton, 1952: pl. 26B). TABLE 8.5. characteristics of human effigy figurines and maskettes. a dash (-) indicates data are non- existent or not available. Width/ Catalog Length diameter Height number (cm) (cm) (cm) Raw material Notes 448890, 448896 11.40 9.97 28.60 Red cedar seated man 448891 9.20 21.50 38.70 Red cedar seated man with tenon 448892 13.00 16.00 30.50 Red cedar ?spiro Man?; painted 448706 7.00 7.35 9.70 galena small seated male 448772 - 10.82 21.00 shell standing male; craig c 423298 4.15 - - shell Maskette 423299 3.95 - - shell Maskette 423300 4.05 - - shell Maskette 423301 3.90 - - shell Maskette N U M B E R 4 9 ? 16 3 the figure is sitting on one knee with the other leg flexed in front and the foot flat on the basal plane. an occipital hair knot is worn. Very vague toolmarks can still be seen. the buttocks are defined by a groove. the head is narrow and somewhat flattened longitudinally. the left side of the face is clearly defined, especially the left ear and eye, as well as the nose. the eye spot is deep and could have held an inset. the right eye is flattened toward the side of the face. the right side is featureless except for part of the eye. in this case it was the right side that became damaged by burning. workmanship and style are quite similar to that of 448891. although it is speculation that they were made by the same artist, it is clear they were engendered by a single genre in human depiction. the third wooden figure (448892) is a well- preserved seated male (Figure 8.18). this is a well- known figu- rine that has been illustrated (hamilton, 1952: pl. 26a; Brown, 1984b:247; Brose et al., 1985:138). the man is modeled realistically. Deeply sunken eyes once held insets that are now missing (they may have been copper or shell). the forehead is high and not flattened (i.e., without cra- nial deformation). the nose is aquiline, and nostrils are shown. the lips protrude around an open mouth. the head is 10.77 cm long and 8.9 cm wide. a long neck ex- tends above a clearly defined clavicle. the width at the shoulder is 13.44 cm. Pectorals are modeled on the chest. the man is nude and devoid of ornament except for an ear decoration. Male genitalia are realistically sculpted. the man sits cross- legged, with his feet tucked underneath his knees and his hands resting on his knees. he wears no footwear, but the toes are visible. the area between his crotch and legs is carved out, leaving a space. the back FIGURE 8.18. wooden figurine, seated ?spiro Man? (448892). FIGURE 8.17. wooden figurines: a, seated man with tenon (448891); b, head only (448896). 1 6 4 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y shows no detail and is without the buttock definition pres- ent on pipes. the sculpture exhibits remnants of adhesive and traces of copper in the eyes. there is red paint residue in the mouth and white paint around the mouth and on the cheeks. traces of copper occur elsewhere on the body. the right ear has a bit of cordage drawn through a drilled hole in the lobe. the only decoration on the base is the excision of the feet. the left earlobe is damaged. the right arm is broken in two places but has been repaired and filled with some sort of filler that has been colored brown to match the object. the front part of the legs has also cracked, and these cracks are filled with an unidentified residue. the left hand is damaged. the most interesting damage is present on the back, where the man was stabbed four times. the perforations run in a line extending from the man?s upper left to the lower right. the holes are square in outline, and depths of the holes are 0.3, 0.75, 0.3, and 0.98 cm. these stab holes could have been used to mount the figure. the square outline of these holes suggests that the damage was done prehistorically with a copper pin, which is customar- ily square in cross section. if done after excavation, the implement used would have to have been a square nail, either cut or hand wrought. the latter seems unlikely. Stone one human figurine was carved from a single lump of galena (Figure 8.19). the figure is seated, with arms resting on legs in a posture similar to that assumed by the wooden figurines. Details are absent. the galena (lead sulfide) has undergone some chemical alteration, and some features originally present may now be obscured. the base of the figure is coated with a thick layer of copper and malachite. Perhaps it rested on a copper deposit while buried. this galena figure is interesting in that galena is a less than trac- table material for doing lapidary art. crude modeling is typical of the piece, but the arms are realistically shaped. Ears are depicted as bumps. the lack of detail may be as much a function of the difficulty in working galena as it is of style, intent, or the skill of the artisan. the buttocks are modeled, and the head is flattened at the forehead. some of this flattening may be from erosion of the original sur- face, or it may represent cranial deformation. this object has not been reported elsewhere, and similar figures have not (to my knowledge) been reported. it is significant that hamilton (1952) did not report this effigy figure among the stone effigies that he listed. it is from the Meyer collec- tion, which hamilton did inventory. Perhaps the figure was included with the 500 pounds of worked galena reported by hamilton (1952:87). galena is known to assume the shape of balls and other geometric shapes, probably as a result of grinding it to obtain white lustrous pigment. the figurine is a rare form. in fact, Brown (1996:643) indicates that galena was not used in making other artifacts. this artifact appears to be an exception. the figure has basketry impression and residue on the anterior side, especially on the figure?s face, knees, and chest. there is little to no basketry on the figure?s back. this lack, along with the heavy layer of copper on the base, suggests that the figure sat upright on copper, facing and touching basketry. Shell one figural shell effigy is present in the collection (Figure 8.20). it differs from other effigies in that the raw material limits the object to two dimensions. this object consists of a human form cut out from shell and incised with decorations depicting accessories. the total height of 21 cm excludes the head, which is not with the collection.6 the front of the figure is the inner shell surface; how- ever, the simple bead grid designs are carried around to the back (outer shell) surface at wrists, knees and feet. the man is shown wearing a belt decorated with a repeated motif of dotted circles in diamonds. a triangular bib ex- tends down onto the chest from the neck. the craig c styling is bold and simple. a striking feature of this object is that it was made using material from a previously engraved cup. the outer FIGURE 8.19. galena figurine (448706): a, front view; b, side view. N U M B E R 4 9 ? 16 5 surface presents a palimpsest of designs. the original Braden B design on the outer surface depicts large faces with bilobed arrow headdresses and was used by Phillips and Brown in refitting the head from the collection of the national Museum of the american indian. the overall large size suggests that the original Busycon from which it was cut was a very large specimen. Maskettes there are four small shell maskettes carved in low re- lief on shell discs (Figure 8.21). they were almost certainly made by the same craftsperson because they are very simi- lar in workmanship and style. human faces are depicted in frontal view, and each person has a beaded forelock. in two cases the forelocks are pulled to the right side of the face, and in the other two they are pulled to the left. two have facial decoration (one each of the right- sided and left- sided forelock wearers). the ones with facial decoration have larger round faces, and the hair on top is detailed using hatched lines. the two smaller- faced people have no decora- tion and less overall detail on their faces. original notes sug- gest that these represent women, whereas the larger faces represent men. all four have pendants hanging from the ear or earspools. the smaller- faced people have smaller ear pendants too. these artifacts are referred to as cameos in the nMnh catalog. it is possible that when new, the differ- ential luster of layers of shell may have been visible. at this time, all surfaces are equally weathered and dulled. all four maskettes have a yellow residue, perhaps ocher, on the face. these are the medallions originally reported by ham- ilton (1952:57?58, pl. 85) as having been found inside a FIGURE 8.20. shell figurine (448872): a, front view; b, back view showing previous engraving. 1 6 6 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y cedar box that the commercial excavators destroyed. the box reputedly had a lid of mica. Brown (1996:531) argues that the maskettes placed in a mica- topped box represent a symbolic burial. hamilton (1952:57) also reported that these came from one of the southern mounds of the craig Mound complex. Animal Effigies animal effigy forms in this collection are rare. one wooden effigy (448910) represents a small bird head. in- sets have been excised for eyes and a crest. a large glob of melted resin rests in the posterior inset. there is no sign that this artifact was attached to anything. that is, there is no body, but the object does not appear to have been broken. overall, the head closely resembles the figure on the wooden effigy bowl handle (448894) except that it is much smaller (see Figure 4.38). Brown (1996:531?532) describes bird effigies that were probably designed to sit atop staffs. the birds he describes are whole birds, not just the head. bonE objECts objects of bone are not plentiful. it is possible that as the relic hunters discarded human bone, they also dis- carded artifacts of modified bone. one artifact (448939) comprises four beaver bones, including innominate, femur, tibia, and fibula (Figure 8.22). the bones are tied together with cordage and may represent a ritual bundle. the pres- ence of beaver bones is interesting because beavers do not otherwise occur in the iconography. several other bone artifacts or implements are pres- ent. they consist of awls, spatulas, and tubes (table 8.6). two artifacts appear to be spatulate scrapers: 448710, which is made from an elk metapodial, and 448709, which is made from an ungulate long bone. artifact 448709 is a long object made from a long bone that has been split (Figure 8.23a). the handle retains the shape of the condyle. all edges are well rounded, and the surface is covered with striations from grinding. the working edge or tip is worn to an asymmetrical shape and well rounded from use. FIGURE 8.21. shell cameo maskettes: a, 423298; b, 423299; c, 423300; d, 423301. N U M B E R 4 9 ? 16 7 FIGURE 8.22. Beaver leg bone bundle tied with cordage (448939). TABLE 8.6. Frequency and characteristics of bone artifacts. a dash (-) indicates data are nonexistent or not available. Catalog Length Width Thickness Weight number Object (cm) (cm) (cm) (g) Count Comments 448939 Bundle 13.0 5.0 4.5 54 4 Beaver bones 423341 Pendant 0.71 1.93 - 0.8 1 Unmodified bone 378264 Pendant 5.60 1.51 1.19 13.2 1 antler tooth 423343 gorget 5.00 3.40 0.42 5.8 1 Mammal scapula 448708s02 awl 17.50 2.13 0.89 46.4 6 troughed fragments 379212 awl 8.1 3.53 0.46 2.4 1 os penis 379213 awl 8.53 1.21 1.10 2.3 1 Long bone 423342 Pin 8.87 0.67 - 3.8 1 Distal fragment 448707 Pin 17.13 0.97 0.51 5.9 1 spatulate 379214 Pin 7.50 0.74 - 37.1 7 Fragments, 1 distal 448708s01 Pin 13.43 0.65 0.60 28.3 15 Fragments, 7 distal 448710 scraper 26.70 4.67 5.1 214.6 1 Metapodia, elk 448709 scraper 33.5 5.10 - 81.5 1 Long bone, ungulate 448711 tube 17.3 2.5 2.1 73.7 1 hollow long bone 448898 Perforated plate 10.06 4.78 0.26 6.5 1 turtle 423221s03 Perforated plate 9.50 3.38 0.1 5.9 2 Round, thin FIGURE 8.23. Bone spatulas: a, ungulate long bone (448709); b, elk bone metapodia spatula with toothed tip (448710). 1 6 8 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y the other spatulate scraper, 448710, is a long bone that has been split to open up half the length (Figure 8.23b). the working edge is 3.4 cm wide, and along this edge are fine teeth cut in with a very narrow tool, perhaps cord and grit. the teeth are precisely parallel, so they could not have been cut using a stone tool. there is a red patchy residue on the outer surface, and the working edge is damaged. the one tube, 448711, is manufactured from a hol- lowed long bone. the exterior is well polished, and it ap- pears to have sediment inside. woodEn objECts wood preservation is excellent, and there are numer- ous fragments of worked wood. some items are complete, others appear to be the wooden portions of composite ar- tifacts, and still others are fragmentary. wooden artifacts in the nMnh collection are listed in table 8.7, and some are shown in Figure 8.24. one artifact type known from spiro is a large, lanceo- late, wooden blade form that was encased in copper foil. the objects in the nMnh include two small wooden frag- ments and one larger distal fragment that are all deeply carved on one face (Figure 8.24a). these fragments could be portions of similar wooden blades. Piece 423391 exhib- its very faint toolmarks and copper residue. the depres- sions carved onto one face mimic flake scars. although only the distal tip is extant, the size of the tip suggests an overall artifact size consistent with the examples from the ohio historical society (Brown, 1976:194). Brown (1996) refers to these artifacts as ceremonial weaponry, carved and copper covered to form large symbolic blades. the nMnh also retains three plaster casts of artifacts that are in the collection of the ohio historical society. there is a large blade- form artifact of wood (448897) that is asymmetrical in shape (Figure 8.24d). it is perfo- rated in several locations, and one face is carved with shal- low, smooth- sided depressions, perhaps representing flake scars (Figure 8.24 hamilton, 1952: pl. 24B). the proximal end of this tool has a rectangle excised into it on each face (offset, not back- to- back). this recessed and perforated area could have served as a location for setting in and TABLE 8.7. Frequency and characteristics of wooden objects in the collection. a dash (-) indicates that data are non- existent or not available. Catalog Length Width Thickness Weight number Object (cm) (cm) (cm) (g) Count Comments 448897 Blade 26.00 5.61 0.83 28.1 1 Distal fragment 423396 charcoal 8.50 2.64 1.45 7.4 7 three are charred wood 423397 Fragment - - - 4.1 20 splinters 448910 Effigy object 5.85 2.65 3.25 15.6 1 Bird head effigy 423392 ornament 6.79 4.16 0.96 4.6 1 Blade 423393 ornament 10.07 2.06 0.50 1.3 1 Fragment of blade 423391 Blade 14.34 4.63 0.83 9 1 Distal fragment of ovate blade; plano-convex cross section 423398 Pole 11.12 10.00 - 456.2 1 Log segment 423400 Pole 279.00 10.00 - - 1 tapering 423399 Pole 309.00 10.00 - - 1 straight 423401 Pole 414.00 10.00 - - 1 tapering 423395 stick 35.00 1.76 0.91 11.4 1 splinter 448893 Vessel 12.40 9.84 12.80 220.9 1 Vessel ornament, puma 448894 Vessel 13.58 8.47 11.24 162.8 1 Bird effigy and bowl rim 448911s02 worked fragment 3.57 2.00 1.13 4.4 17 Fragments 448911s01 worked fragment 6.62 5.18 2.30 55.1 8 Various 448911s05 worked fragment 13.45 6.04 1.55 58.6 1 spatulate artifact 448911s04 worked fragment 18.80 4.00 2.00 229.1 24 stick fragments 423394 cedar 4.00 2.20 1.25 3.4 3 Fragments N U M B E R 4 9 ? 16 9 affixing another facet (now missing) of a composite arti- fact. there is no residue to belie the character of this miss- ing component. the proximal edge has three projections. artifact 448911s05 is a complete spatulate artifact (Figure 8.25). it is thinner at the wider end. Perhaps it once belonged to a macelike form, although the morphol- ogy is not consistent with known wooden maces. some of the most interesting wooden objects are also among the plainest. Lot 448911s04 includes fragments of wood that at first glance seem simply to be narrow thin sticks. there is regularity in form, however, and on closer examination, it is apparent that many of these fragments refit. two fragments are shaped at one end, one exhib- its a mat impression, and others are stained with copper. Dimensions given in table 8.7 are the maximum. Eleven of the fragments are 5?7 cm in length and 1.5?2.5 cm in width. they are regular and do not exhibit branching. it is suspected that several pieces represent parts of similar ar- tifacts or the same artifact and may represent portions of a bow. one flat, wide fragment differs from the others and may be a portion of atlatl or wooden mace or blade. some are slightly charred. smaller, chunky fragments within this lot are less suggestive, although they also appear to have been worked. FIGURE 8.24. wooden artifacts, blade- like forms: a, blade- form fragment (423391); b, (423392); c, 423393; d, asymmetrical blade- form artifact (448897). 1 7 0 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y there are also three long cedar poles. these are all 10 cm in diameter and between 2.5 and 4 m long (table 8.6). they are either from litters or from internal mound supports. two taper to one end, and one is of consistent diameter throughout. a small section of a cut log is also present in the collection. lEathEr Very few pieces of leather are present in the collec- tion. Leather is a highly perishable material that normally would not preserve. the leather present either adheres to copper or is heavily copper stained, suggesting that the copper and copper salts acted as a preservative. artifact 448934s08 is a tanned leather rectangle. cop- per stains one side. the piece is perforated along the edges, presumably for stitching. it exhibits a well- preserved mold of a loosely meshed basket or mat. the leather sat flat on the mat and worked its way down to conform to the structure of the matting. it is possible that the leather was actually sewn to the mat structure. it was originally part of the large twill basket, but their association is unclear. it may be that the basket had a covering of leather. artifact 423378s01 consists of leather fragments ad- hering to corroded copper sheets. Perhaps these repre- sent some sort of composite artifact, such as a headdress plaque with leather facing. Pieces of leather along with fragments of basketry adhere to copper sheet fragments (378267). one piece of leather is 2.1 ? 4.2 cm in size. the leather is thin and brown and soft, unlike rawhide. another fragment (423376s02) is a thin piece of hide from an unidentified animal. Yet another fragment (448920) is a tanned leather rectangle similar to that re- corded for 448934s08. copper fragments adhere to one face. at one time it had been square?at least, there is a perfectly square corner present. it had been sewn. small perforations are present close to the remaining edges at ap- proximately 1.2 cm intervals. Edges are slightly scalloped in appearance, as they would be if the piece had been sewn and stretched in the process. there are two larger round perforations in the left central area. this piece is partially decomposed. there is also a soft reddish brown leather fragment (448916s01) that is 4.0 cm by 3.5 cm in size. other WorKeD obJectS other materials, including several kinds of miner- als, show modification. one intriguing object is a lump of ocher (448924) that has been shaped into a rounded mound. there are wide, shallow grooves running verti- cally along the face, and these appear to be finger trails. there is a pit nearly 1.0 cm deep in the top, suggest- ing that something was embedded into the apex of the object. a few of the galena pieces show signs of working as well. one is a smoothed cube with a depression in one face (448704). the artifact is 5.42 cm long by 5.01 cm wide by 3.75 cm high. it is a nodule with incised or cut cross- hatching on four faces. one corner has the surface shaved off in strips 0.33 cm wide and as much as 3.0 cm long. one face is smooth and ground out, leaving an indenta- tion with wide, shallow walls and circular toolmarks. the different surface treatments are interesting. this could be a piece used in producing pigment, or it could be a ritual object. another is a pyramidal galena object, approxi- mately 2.0 cm in length, with six well- smoothed faces (448705s01). Perhaps this object was worked as a pre- form for a pendant or was used in gaming or another un- known purpose. in addition, a baked clay biconical object is also pres- ent. this object (448928) is a typical representative of a class of objects associated with the Late archaic site of Poverty Point in Louisiana. it suggests some interesting al- ternatives. the object in question could have been traded for, kept, and handed down through successive genera- tions; it could have been recycled from an earlier context; or it may not derive from spiro at all. FIGURE 8.25. wooden spatulate artifact (448911s05). N U M B E R 4 9 ? 17 1 unWorKeD ritual ma terialS Unworked materials include items which may be of ritual importance or the result of ritual behavior but which are essentially unmodified. this category includes paint or pigments and chunks of minerals, such as quartz, galena, or mica. also, any materials that are not manufac- tured artifacts but are, rather, by- products of the activities at spiro are included. PigmEnt Brown (1996:536) lists numerous minerals that were used as pigment, including minerals that could be used to create hues of the colors red, yellow, brown, green, white, and black. Ocher Both the trowbridge and Meyer collections contained ocher. Most of it is fragmentary, yet there are some large lumps present. the combined weight of ocher is 2,682.5 g, including the weight of worked ocher lumps. the weight excludes highly fragmentary and friable ocher contained in two 1- pint jars. the ocher ranges from a dark, almost ma- roon, color to yellow; however, most of it is a strong red. ocher is present as a residue on many artifacts, in- cluding lithic tools and shell cups. Furthermore, the red slip found on sanders ceramics was apparently produced using red ocher. Red paint produced from ocher remains on one of the large stone maces. Glauconite there are 171 fragments of glauconite in one lot in the collection of R. K. Meyer (448923). combined weight is 755.9 g. glauconite, which Brown (1996:644) describes as a hydrous silicate of iron, aluminum, and potash, is a claylike mineral ranging in color from green to bluish green. glauconite was apparently used in painting or for smearing on artifacts. Maces such as 423196 and figures such as 448891 exhibit traces of glauconite on the sur- faces. glauconite is of such a distinctive color that it is unlikely to be mistaken for naturally occurring clay sedi- ment. where it occurs, it is usually overlaid by pale brown clay, interpreted as vestiges of mound sediment. the glau- conite must have been applied to the surface before the ar- tifacts were deposited in the mound. glauconite has been associated with crockett curvilinear incised ceramics. several vessel bases of this type are present and contain thick deposits of this mineral on the interior of the vessels. sources for glauconite include marls located in arkansas as well as gulf coastal areas. minErals Mineral Lumps there are 10 odd pieces of heavily burned stone. a report on analysis done by the conservation analytical Laboratory (smithsonian institution, Museum conserva- tion institute, 1975, unpublished caL report 2018) indi- cates that these are of micaceous sandstone that has been coated with a vegetable gum. Brown (personal communi- cation, 1992) suggested that they may contain manganese. the gum is very likely flammable and would probably burn brightly. these pieces may represent the remains of torch- like objects whose function was to provide light. subse- quent reanalysis in 2010 of fragments from lump 448689 (smithsonian institution, Museum conservation institute, unpublished Mci report 2018.2) did not confirm the origi- nal interpretation. additional analyses will be necessary to determine the possible function of the mineral lumps. De- scriptive statistics are shown in table 8.8. these stones appear to be artificially shaped. several exhibit long, wide grooves that do not appear altogether natural. they all have irregular but smoothly curved sur- faces along with the grooves. Use beyond burning is not indicated. they are heavily burned and thickly coated with a layer of charcoal. the stone matrix is friable from heat damage. Galena Raw galena (lead sulfide) is conspicuous. a large number of small fragments (n = 140, weight = 1,762.1 g) were recovered from a quart jar of mixed Olivella beads TABLE 8.8. statistical summary of burned mineral lump mea- surements (n = 10). Standard Measurement Mean Extremes deviation Length (cm) 11.11 7.60 / 13.02 1.90 width (cm) 5.19 4.10 / 6.95 0.90 thickness (cm) 3.54 2.42 / 4.46 0.63 weight a (g) 186.14 101.8 / 251.1 55.26 a total weight is 1,861.4 g. 1 7 2 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y and small shell beads (presumably found in the same pro- venience). there are other nodules and cubes of various sizes as well, as shown in table 8.9. some of the pieces exhibit signs of working, possi- bly as a part of the production of pigment because the modification appears random, not directed toward the creation of finished galena artifacts. Piece 448703 has red pigment and thickly applied brown clay in crevices. although the catalog card stipulates that there are no signs of use, the object appears to have been reduced in a couple of areas. this is the largest nodule and is natu- rally shaped roughly like a crouching animal. Perhaps this natural shape led to this nodule being used as a ritual object. otherwise, it could have been raw material for obtaining powdered galena. another object, 448702, is worked slightly, with both ends ground using a circular, pestle- type motion. Quartz Brown (1996:644?645) writes that quartz crystals are not uncommon from craig Mound contexts and records at least 17 unmodified crystals from the wPa excavations. trowbridge collected two large quartz crystals (423208, 423211). one is milky and measures 9.92 cm long, 5.18 cm wide, and 4.29 cm high. it weighs 267.4 g. a milk quartz crystal was also recovered during the wPa proj- ect (Brown, 1996:536). that crystal is battered or ground along its sides. Brown (1996:537) writes that quartz oc- casionally shows battering. the other quartz from the trowbridge collection is a long, transparent crystal of fine quality that measures 12.54 cm by 4.22 cm by 3.16 cm and weighs 222.3 g. the larger crystal is a single rock crystal with sparse red residue on its surface. the base has been shaped, and there are fairly even conchoidal fractures around the perimeter of the angled base. still other fragments of quartz are present, but they are miniscule. thirty- two tiny fragments of quartz were mixed with galena, small shell disc and convexo- cylindrical beads, and a large quantity of Olivella shell beads that were separated from lot number 423331. Mica Mica is not plentiful. there is a single vial contain- ing three small fragments, each approximately 1 cm long. these fragments are all elliptical in shape. a few tiny flakes (n = 3) were recovered from the conglomeration of beads and galena that compose 423331. another large piece of mica (448921) is more than 4 cm in diameter. it exhibits one clearly cut edge whereas the other edges are ragged. Brown (1996:543?544) reports on mica recovered from the craig and Brown mound contexts at spiro. one use for mica was in covering cedar boxes reported to have been excavated by the relic hunters but then destroyed. othEr matErials the following materials are not assumed to have cer- emonial or ritual function, although the slag suggests a mortuary event if it resulted from cremation. these mate- rials do, however, provide information regarding the origi- nal archaeological context. Glass Slag there is a piece of frothy vitrified silica with a large glass bubble present in the surface (423210). the glass bubble measures 0.88 ? 1.02 cm in diameter, and the slag is 8.48 cm long, 4.57 cm wide, and 1.65 cm high. weight is 47.7 g. there appears to be bone within this slag as well. Rogers (1980) mentions the recovery of silica froth from another part of the site. according to Brown (per- sonal communication, 1992), crematory remains?vitri- fied bones?are not uncommon. silica froth also occurs from the intense burning of grassy materials (Brown, 1996:645). three other fragments of burned bone and shell are present as well. these are small and exhibit no vitrification (423303). Coprolites there are three pieces of fecal matter that weigh 5 g total. Maximum dimensions are approximately 6 cm TABLE 8.9. Unmodified galena nodule measurement summary (n = 140). all measurements are in cm. total weight is 1,762.1 g. Standard Measurement Mean Extremes deviation Length 5.1 2.4 / 11.0 2.96 width 4.17 2.4 / 7.3 1.84 thickness 3.28 1.5 / 6.69 1.69 N U M B E R 4 9 ? 17 3 long by 0.8 cm in diameter. organic material is clearly preserved in these artifacts. Sediment in some cases, vials of sediment have been removed from some objects at some point during their curation, and these vials can be considered as remnants or residues relating to the original archaeological context. of the six artifacts that fit into this category, one derived from a stone pipe (448669s02), one from a shell gorget (423308s04), one from a copper sheet (448715s05), one from a textile (4233378s05), and two, which contain clay and bits of vegetable matter, from mixtures of beads, mica, and metal fragments (423330s06, 4233301s02). Seed a single large seedpod is present among the wooden fragments of 448911. the plant represented has not been identified. SummarY the magnitude of the collection of ceremonial, spe- cial, or prestige materials present in the collection from nMnh cannot be underestimated. the ritual items de- scribed in this chapter do not constitute the entire ceremo- nial assemblage from the mound deposit by any means. shell cups in particular were undoubtedly used in special or ceremonial contexts but were more easily discussed as containers. of all the materials included in this chapter, pipes, many of which are figural, and the other effigy fig- ures clearly indicate that a life rich in social interaction and spiritual import typified the individuals buried in the great Mortuary. in some ways, the presence of raw materials having ritual significance, such as pigments and minerals, is sur- prising. the artifact miners apparently collected materi- als that were not made into obvious finished artifacts. although it is impossible to say that these raw materials themselves were used in a ritual context, they certainly were deposited in a special context, having been interred in the craig Mound. in addition, the prestige value of ma- terials such as copper, hematite, and galena has a long his- tory in eastern north america, having been included as a major component of grave goods during Middle wood- land copena contexts in the southeast. note 6. Phillips and Brown (1984: pl. 285) remarked that the head was acquired from the Museum of the american indian, heye Foundation, by the smithsonian and that the two pieces are now together. no record of such a transaction could be located, and there is no head present in the collec- tion at the national Museum of natural history. Javier Urcid, Department of Anthropology, Brandeis University, P.O. Box 549110, MS 006, Waltham, Massachusetts 02454, USA (formerly with the Department of Anthropology Repa- triation Office, National Museum of Natural History, Smithsonian Institution, Washington, D.C., USA). t his chapter deals with the skeletal remains from the craig Mound at spiro and discusses only those catalog numbers that contain human bone. the set of remains with number 448937 includes 109 loose teeth that were sorted into the groupings shown in table 9.1.7 these teeth have not yet been studied and documented. given their condi- tion?not only fragmentary but also outside of anatomical context?it is consid- ered that their future study will not greatly affect the conclusions reached here regarding minimum and maximum number of individuals. some of these teeth might actually form part of maxillary and mandibular fragments listed under number 448935. cataloG number 423344 this catalog number includes two right and one left parietal. two of them are antimeric. all the sutures (sagittal, occipital, and a portion of the coronal in the partially assembled cranium) are open both endo- and ectocranially, suggest- ing that both crania were from young adults. although warped and deformed, the incomplete cranium seems small and thus might belong to a female. the isolated parietal is larger and might belong to a male. Both skulls had ossicles in the lambdoid suture. their color and texture are very similar, suggesting that the two came from the same underground burial context. the isolated left parietal has a green stain in the ectocranium, in the area of asterion. there is no evidence of pathology. cataloG number 423345 this catalog number includes the left half of an adult mandible. on the basis of slight to moderate dental wear the individual could have been between 35 and 45 years of age. the mental shape and eminence suggest a male. although the mandibular is eroded, the entire left lower dental arch is present. there are some moderate alveolar tori. traces of soil, especially along the broken edge, 9 human Remains Javier Urcid 1 7 6 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y indicate that the bone was found in an underground burial context. cataloG number 423346 this catalog number includes the complete left hu- merus of an adult individual. specific age is undetermin- able. the occurrence of slight marginal osteophytes in the distal epiphysis suggests an age range between 35 and 45 years. the shaft of the bone looks robust, but the diameter of the head is only 39 mm, well below the mean for females, which is 43 mm. near the distal end of the shaft there is a bony excrescence that is not surrounded by active remodel- ling. it seems to be a very localized and healed osteitis. cataloG number 423347 included in this catalog number is the complete right tibia of an adult individual. specific age cannot be deter- mined. however, the lack of degenerative lesions on the surfaces of articulation suggests a young adult. the bone is gracile and might correspond to a female. some erosion on the epiphyses and cracks on the surface of the bone indicate an underground burial context. there is no evi- dence of pathology. cataloG number 423348 this catalog number includes the complete left in- nominate of a male between 30 and 40 years old. the sexual determination is based on the angle of aperture in the sciatic notch, the lack of a preauricular sulcus, the lack of medial projection of the auricular surface, absence of a ventral arch, and broad medial aspect of the ischiopubic ramus. the age assessment is based on the morphological changes of the pubic symphysis. the acetabulum has slight marginal osteophytes and slight localized porosis in the articulatory surface. there is postmortem damage, caused apparently by the tip of a nail, in the ventral surface of the ilium, very close to the superior border. cataloG number 423349 this catalog number includes two incomplete in- nominates. only one is human. it is from the left side. it belongs to an adult perhaps between 40 and 50 years of age, as suggested by the morphological changes of the au- ricular articulation. the criterion of changes of the pubic symphysis cannot be used since this portion is missing. Female sex can be determined on the basis of an open sci- atic notch, marked preauricular sulcus, and medial pro- jection of the sacroiliac articulation. the bone is eroded and incomplete, with cracks in the surfaces, some exfolia- tion, warping, and bleaching, mainly on the ventral sur- face. these traits indicate partial exposure weathering. the yellow color on the dorsal surface indicates a previ- ous underground burial context. cataloG number 423350 this catalog number includes a very eroded fragment of a right femur. it belongs to an adult individual, but specific age and sex are undeterminable. Erosion, exfolia- tion, and cracks indicate that the fragment was exposed to weathering. Because of its advanced erosion, its previous taphonomic history has been obliterated. the fragment in- cludes portion of the distal shaft and parts (amorphous) of the distal epiphysis. there is no evidence of pathology in what remains of the original surface of the bone. cataloG number 423351 this catalog number includes three vertebrae, two of which are nonhuman. the human vertebra is incomplete, represented only by the right half of the centrum and a portion of the right arch. it is probably a c4 and belongs to an adult. Because of color and texture it cannot be from TABLE 9.1. types and counts of human teeth present in catalog number 448937. Type Count incisors 10 canines and premolars 20 third molars 16 Upper molars 14 Unerupted or partially erupted teeth 2 teeth with cavities 1 Unsorted teeth 18 Undifferentiated molars 22 Unrecognizable tooth fragments 6 Total 109 N U M B E R 4 9 ? 17 7 individual B in number 448936. specific age and sex are undeterminable. cataloG number 448935 this catalog number includes several different sets of bones. two are fragments of human skull. on the basis of differences in color and texture they do not belong to the same individual, but both were undoubtedly in an under- ground burial context. one is an incomplete right parietal from a young adult. the sagittal and lambdoid sutures were both endo- and ectocranially open, except for the en- docranial area of obelion. the coronal suture was already endocranially fused but open in the ectocranium. sex is undeterminable. there is a fine osteoporosis over much of the ectocranial surface. the other fragment corresponds to a right temporal and might belong to a young adult since there are no degenerative lesions on the temporo- mandibular joint. also, both the superior and inferior sphenotemporal sutures are open endo- and ectocranially. the medium size of the mastoid process does not allow a reliable determination of sex. there is a postmortem per- foration just anterior to the origin of the zygomatic bridge, and the hole was produced from the inside to the outside. thus, the temporal bone was already isolated from the skull when the perforation occurred. this might have been produced by an excavating tool, perhaps a small pick. there is no evidence of pathology in this fragment. also included are two incomplete maxillae that match. From the surviving sutures, the person was an adult between 25 and 30 years of age. the dental wear in the surviving teeth is slight. the left fragment corresponds to the anterior portion and includes alveoli of the incisors, the canine, the two premolars, and the medial surface of the first molar. no teeth remain. the central incisor was lost antemortem. the other teeth fell out after death, and the dental pieces might be among the set of loose teeth numbered 448937. the right maxilla includes the canine, first premolar, and the second molar. the first molar was lost antemortem but shortly before death. the sockets for the tooth?s roots are in an initial stage of resorption. the incisors and the third molar were lost postmortem. additionally, there are two incomplete mandibles. the permanent dentition, the slight degree of attrition, and the dental pathology (caries) suggest that both mandi- bles pertain to young or middle- aged adults. on the basis of color differences, none of the mandibles match with the paired maxillae. Using the same criteria, however, one of the mandibles might correspond with the right parietal described above. thus, there could be as many as four dif- ferent adult individuals in this set. cataloG number 448936 this catalog number includes the miscellaneous re- mains of various individuals. in the discussion of each anatomical set, these individuals are distinguished with letters. 1. Four cervical vertebrae (c1 to c4) of an adult in- dividual (a). the color and anatomical fit between them clearly indicates that they belong to the same individual. the lack of degenerative lesions suggests a young adult less than 40 years of age. sex is undeterminable. traces of soil adhered by shellac evinces an underground burial context. 2. cervical vertebra (c3) of an adult individual (B). color and texture are different from the previous set. no pathology is evident. this is apparently a young adult; sex is undeterminable. traces of soil and a crack in the cen- trum indicate an underground burial context. 3. two fragments of ribs. Both are from adult individual(s). Judging from the color and texture, it is fea- sible that one fragment belongs to individual a and the other to individual B. cracks and erosion on both frag- ments indicate an underground burial context. 4. hand bones, including five metacarpals and three phalanges. the metacarpals include one right third (proxi- mal epiphysis unfused), one right fourth, one left fourth, one left fifth that lacks portion of the distal epiphysis, and one metacarpal without distal epiphysis. thus, its posi- tion and side are undeterminable. Because the right and left fourth metacarpals do not match, the incomplete left fifth and the left fourth listed above cannot belong to the same individual, and since the right third metacarpal had its proximal epiphysis unfused, then this set of bones rep- resents at least three individuals. two were adults and the third was less than 15 years of age. there are also three proximal hand phalanges. two are complete and could belong to the same individual. the third is incomplete and seems to belong to another person. 5. Five long bones, including four femora and an ulna. three of the femora are left and another is right. the latter matches one of the left ones. the paired bones are almost complete but eroded in some parts. they have pronounced exfoliation on the anterior surface and gouges that seem to be the result of postmortem damage. Portions of the epiphyses are missing. the head diameter (44 mm) sug- gests a male. there is no evidence of degenerative lesions 1 7 8 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y in what remains of the articulatory surfaces, so presum- ably, the pair belonged to a young adult. the right femur had traces of a red powder adhering to the surface around the lesser trochanter and tiny spots on the femoral head. this substance might be cinnabar. the evidence of the red paint is less conspicuous on the left femur, but a few traces are discernible on the femoral head and in two spots along the posterior surface of the shaft. there is a cut 216 mm long in the head of the right femur. Because of its loca- tion the cut is either a peri- or postmortem cut. the other two left femora show similar surface changes as the pair discussed above, suggesting that all come from the same burial context. the proximal epiphyses of both bones are incomplete, but enough remains to derive the head diam- eters. according to these, which are 42 and 43 mm, the femora could belong to two females, yet in other features of the shafts, they seem to be slightly more robust than the pair described above. only one of these left femora has traces of red powder adhered to the femoral head. the fifth long bone in this set is a left ulna from an adult in- dividual. its relative robusticity and the lack of degenera- tive lesions in the articulations suggest a young adult male. traces of soil are adhered by a layer of shellac, evincing an underground burial context. however, the color of the bone and the lack of surface changes indicate that the ulna probably does not come from the same burial context as the femora, although the differences could be the result of the bone having been found at a deeper level. 6. nine foot bones, all from adult individuals. one is an incomplete left navicular. there are also six metatar- sals, including two complete left first, one complete right first, one complete left third, one complete right fourth, and a fragment that includes only the proximal epiphysis and portion of the shaft of an unidentified metatarsal. the latter might be from a right foot. the right first metatar- sal does not match with either of the two left ones, and thus, the set of bones has remains of at least three adult individuals. on the basis of color similarities, the right first metatarsal, the right fourth, and the left third might belong to the same individual. other foot bones in this set include two first proximal phalanges. on the basis of color and size, one seems to correspond to one of the left first metatarsals, and the other seems to articulate with the right first metatarsal. 7. Fragment of a parietal bone and the shaft of a left fibula. Both are from an immature individual and probably from the same skeleton (c). age and sex are undetermin- able. the shaft fragment is 13 cm long. it is estimated that the complete bone would have been at least 16 cm in length, giving a possible age of 4 to 7 years. a tag on the fibula says that its weight is 6.2 g. SummarY remarKS considering groups of anatomical elements, the fol- lowing are independent counts of minimum number of individuals. skulls at least four adult individuals and one immature person (4?7) years old Mandibles at least three adult individuals Vertebrae at least three adult individuals humerus at least one adult individual Ulna at least one adult individual hand bones at least three adults and one imma- ture individual less than 15 years old but older than the immature person counted in the skulls set innominates at least two adult individuals Femora at least four individuals: three under number 448936 and one under num- ber 423350 tibia at least one adult individual the figures generated by the skulls and the hand bones provide a composite picture of at least six individuals: four adult, a child between 4 and 7 years old, and an adoles- cent less than 15 years old. if no relationship is assumed between all the remains, then a maximum count of 24 in- dividuals results. the archaeological context should allow determining a more accurate number of individuals. this determination would also permit a more accurate sorting of the age and sex assessments that were provided in the descriptions. note 7. osteologist Beverly Byrd, formerly of nMnh?s Department of anthropology Repatriation office, assisted in collecting and organizing the information on human teeth. t he nMnh collection from the craig Mound contains general arti- fact types similar to those found during controlled excavations by the wPa. shell vessels, pipes, and wooden statuary are well represented in collections made by relic hunters. Looters, however, miss things. the Pocola Mining company undoubtedly went for large, obvious, and well- preserved materials. they discarded human bone, ignored small lithic materi- als, probably missed thousands of shell beads, and hastened the destruction of organic or fragile materials, such as textiles, leather, ceramics, or copper sheet artifacts. this concluding chapter assesses the significance of the nMnh collec- tion in light of the problems with provenience and explores its potential both as a research tool and as a major piece of north american cultural heritage. rePreSentatiVeneSS of the collection the nMnh collection of materials from spiro represents a large portion of the materials known from the site, aside from the collections made during the wPa operations. the nMnh collection includes some of the first pieces brought to the attention of professional archaeologists; collector J. g. Braecklein sent ma- terials to the smithsonian for perusal and casting as early as 1936. in the nMnh collection from the site are just fewer than 5,000 artifacts (not including beads). of these, 2,072 are lithic artifacts, including implements, beads, earspools, pipes, and pieces of minerals. ceramic artifacts, including sherds, account for 327 pieces. shell artifacts number 1,160, excluding the over 20,000 shell beads. But just how representative is this sample of artifacts from craig Mound? the relic hunters collected a little bit of every type of material present, but there are notable differences between the wPa assemblages and those in the nMnh. For example, the wPa excavation has relatively more ceramic material, more small lithic material, and proportionately fewer complete shell vessels. Reasons that account for these differences should reflect something about the nature of commercial excavation and collecting. the first possibility addresses the selective nature of artifact collection in which hunters collected only what they saw and what they thought they could 10 conclusion April K. Sievert 1 8 0 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y sell. small projectile points probably escaped notice, as may have other small items. there are very few copper plate pieces in the nMnh collection. the fragility of copper plates may have contributed to them being destroyed by the relic hunters but salvaged by the more careful wPa crews. second, collectors purchase what they personally de- sire. we can expect that the large collections amassed by Meyer and trowbridge reflect the men?s personal tastes. whereas trowbridge may have recognized the value of bundled textiles that may have looked mucky and fragile to others, Meyer apparently had an eye for showy sculp- tural pieces. Differences between the collections from trowbridge and Meyer are remarkable and highlight the serendipitous nature of artifact collecting. For example, of the nine whole or nearly whole pots in the collection, trowbridge had possessed five, which he sold along with only a handful of other sherds. Meyer donated primarily sherds, including many of the slipped sherds from ovaloid bowls, some of which have been refitted to form nearly whole pots. this is interesting in light of the conception (discussed below) that only whole artifacts are of interest to collectors. trowbridge collected 199 pieces of cloth to Meyer?s 29. among wooden artifacts, it was trowbridge who collected four cedar poles and no effigies figures and Meyer, on the other hand, who possessed three human fig- ural effigies, two vessel handles, and 19 wooden earspools. Meyer also donated seven stone effigy pipes, and trow- bridge sold two to the museum. third, collectors and relic hunters are interested in com- plete artifacts, rather than broken pieces. shell cup fragments number 908, including 84 whole or nearly whole decorated and undecorated cups. this compares with 85 complete ves- sels plus 3,695 fragments reported by Brown (1996: table 2- 81) for the wPa material. although several of the com- plete cups in the nMnh collection have been refitted from multiple fragments, the presence of the same number of whole cups to four times the fragments is remarkable. Finally, there could easily also have been real differ- ences in the nature of the deposits removed by the Po- cola Mining company compared to nearby contexts in the mound. the artifact hunters did strike at and remove much of the central portion of the large feature that Brown (1996) refers to as the great Mortuary. there may have been more intact shell in this area, which is thought to have been a protected hollow cavity. SiGnificance of the collection Many of the materials from the great Mortuary fea- ture in the craig Mound are unique. wooden mortuary figures, well- preserved stone effigy pipes, textiles, and ga- lena figurines are rare. For these reasons, the craig Mound burials offer a rare glimpse into late prehistoric lifeways. Preservation of materials interred in the craig Mound was extraordinary, due in part to the large amount of copper artifacts placed in the mound and to the rather imperme- able sediments capping the mortuary features. although not a tomb in the sense evoked by some of the newspaper writing of the 1930s, the possibility of a protected cavity has been supported by Brown (1996). these conditions are in part responsible for the overall character of the col- lection as rare and unique. spiro yielded one of the largest collections of Mississippian textiles known, and the tex- tiles at the nMnh form a large component of this mate- rial. collector harry trowbridge?s work at conserving the textiles that he rescued, in part by controlling tempera- tures, undoubtedly contributed to the fact that so many textile pieces have survived. of course, we have no way of knowing how much material was destroyed during the commercial excavations. the artifact hunters were none too careful as they tun- neled into the center of the great Mortuary. therefore, the material in the collection comes from a context that did not entail the collection of provenience data, and what there is, is anecdotal. therefore, the materials from com- mercial excavations at spiro present two problems. the first is an ethical one in that the materials derive from commercial excavation (looting). wylie (1995) has dis- cussed the issue of looting and the fact that publication of looted materials adds commercial value to the artifacts, something that archaeologists wish to avoid. second, the lack of contextual information makes these materials less than useful for many types of analyses?those dealing with mound structure or burial associations, for example. this problem may actually be less critical for material from craig Mound than for other Mississippian burial contexts. Brown (1996:98?102) points out that the as- semblages of materials within burial units at spiro were in some cases probably already mixed, as a result of a mortu- ary practice that involved collecting and reburying earlier remains with later ones. Even so, the materials in the nMnh collection are not without scientific utility (sievert, 2003). the arti- facts themselves can provide technological information concerning raw materials, manufacturing, and use. the extensive iconography displayed on the artifacts can be used to interpret social and ceremonial life. these data, in turn, can be used in researching questions of subsistence, technology, social interaction, regional integration, and religion. although problematic on a site structural scale, when put into the larger framework of what happened N U M B E R 4 9 ? 18 1 during the Late Mississippian period, the materials have a lot to say. the SPiro collection anD miSSiSSiPPian reSearch throughout the decade of the 1990s, foci for Mis- sissippian research changed, with more emphasis on po- litical economy, the structure of prestige systems, social interaction and integration, and religion. Research on the edges of the Mississippian world, from the perspectives of both geography and time (King and Meyers, 2002), has shown that manifestations of Mississippian society were far- reaching and influential, despite individual polities not always enjoying far- reaching regional influence. although it is not the intent of this work to fully review late pre- historic research in the caddoan area or the southeast, it does seem appropriate to point out the relevance of a collection like this one to ongoing research into Mississip- pian society. Part of what characterizes the late prehistoric period in eastern north america is far- reaching intercultural so- cial interaction. similarities in mound form and iconog- raphy indicate the spread of ideas over a wide area, from wisconsin in the north to Florida in the south and from oklahoma in the west to Virginia in the east. there is ample evidence for interaction in the form of trade, and this becomes apparent when looking at the artifact distri- bution maps developed by Brain and Phillips (1996). some information on trade comes from style; other data can be gathered on the basis of source analyses. Because scien- tific methods for the analysis of materials are constantly improving, the spiro collection will be useful continually. the source analysis done for the obsidian end scraper (catalog number 378273) collected by J. g. Braecklein in 1935 is a case in point. Energy dispersive X- ray fluo- rescence revealed a trace element signature that indicates the material came from the Pachuca obsidian source in Mexico (Barker et al., 2002). this result supports the no- tion that trade relations extended across a wide expanse, in this case to Mesoamerica. Petrographic analyses of lithic materials and chemical analyses of lithic, clay, and other minerals could prove invaluable for interpreting so- cial interaction and integration on a wide scale. Further- more, spiro is itself a mound center located on the edge of the Mississippian world (Payne and scarry, 1998:47). it retains connections to sites in the east but is poised on the edge of the great Plains, perhaps brokering trade to and from the west (Rogers, 1991a, 1991b; schambach, 1993). For this reason, an understanding of spiro is critical to an understanding of trade relations throughout the western half of the Mississippian world. the collection from spiro can inform not only on matters of distant trade but also on social interaction at the site itself. objects that were social in usage, such as pipes and cups (relatively common artifacts in the craig Mound), were used not in solitary situations but in con- texts involving more than one person. Pipe ceremonies such as the calumet of the historic period were ceremonies that extended social relations (hall, 1997). the materials from the craig Mound?the wealth? indicate individuals of importance, perhaps a hereditary chiefly elite. Much of the research about the structure of political influence during the Mississippian (Barker and Pauketat, 1992) has been focused on cahokia (Pauketat, 1993, 1994; Pauketat and Emerson, 1997; Milner, 1998). spiro, a much smaller site with a smaller population yet possessing a remarkably appointed mortuary feature, makes an excellent counterpoint to the Middle Mississip- pian manifestation farther north. Finally, archaeologists can expect more focus on the study of iconography and its meaning. as research be- comes increasingly emic, the iconography found at spiro and other sites throughout the southeast is taking on added dimension. it is likely that we are only beginning to appreciate the magnitude of the knowledge coded in the symbols and motifs of the Mississippian world. appendix a: catalog summary of artifacts in the national Museum of natural history spiro collection analYtical ProceDure anD Da ta liSt the artifacts in the spiro collection fit into one or more of seven raw ma- terial categories: lithic, ceramic, shell, metal, textile, bone, and wood. Listed below are some of the variables used for recording different kinds of objects and the methods employed for taking measurements. CEramiCs Color. Paste colors were recorded using the Munsell color system for the outer surface, the inner surface, and the core. in cases where the colors ranged among adjacent levels in tone or value, the hue was recorded along with the name of general color family. For black, it is difficult to assign a hue, so black was often referred to simply as black. Wall T hickness. For rim sherds, wall thickness was measured consistently approximately 1 cm below the rim. For body sherds, which can vary widely, one measurement was taken at what appeared to be an average thickness. For base sherds, thickness was measured near the center of the base. Size. For whole vessels, diameter and height were taken. For nons- pherical vessels (i.e., elliptical bowls), length and width were noted. in the data- base, diameter and width occupy the same data field. in some cases single sherds were measured to indicate size, especially if it represented a large part of a vessel. in cases of partial pots, those dimensions that could be measured or estimated were recorded. 1 8 4 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y a separate subnumber was given to each separate grouping created to record multiple styles or elements within each catalog number. shEll CuPs a separate data form was created for each fragment. in cases where fragments adhered to other fragments, the single fragment comprising the largest portion of the ves- sel was chosen, and the data form was made on the basis of that catalog number. in cases where the largest frag- ment derived from a lot with many artifacts and there was another fragment of the same cup that was assigned a unique catalog number, the data form was based on the unique fragment, thereby reducing the number of separate data forms created. as with other artifact categories, information from catalog cards was recorded along with information taken directly from the artifacts themselves. conservation cards kept with the collection included a compendium of ideas about and interpretations of motifs recognized in the shell engravings. Joan s. gardner?s notes (unpublished) on motifs represented were recorded, along with personal interpretations. Basic measurements that were taken include length, width, and height of shells that were complete enough to suggest overall dimensions for the complete vessel or shell. Length was measured as the distance from apex to distal tip, width was the maximum distance across the open- ing, and height was recorded as the maximum height of the piece when resting on the shell opening. Dimensions for smaller fragments were also taken, but these were re- corded as maximum or minimum dimensions, especially for lots containing more than one fragment. weights of fragments and vessels were recorded. in general, shell fragments were treated similarly to potsherds. Refitting is an aid to analysis in that it allows for the interpretation of more complete, discrete artifacts. in cases for which there is more than one fragment of a shell but they do not refit exactly, fragments will still be counted together under one data entry. For example, par- ent fragment 448816 has been refit to a small fragment of 448880 and an apical fragment from 448877. in addition, two other fragments, one each from 448877 and 448880, refit each other but do not exactly refit to 488816. Be- cause they are considered by Phillips and Brown (1978, 1984) to have derived from the same shell, they will be counted as one shell. Unless fragments are treated in this way, the redundancy of design among segments of the same vessel will inflate the counting of the incidence of motif occurrence. here Phillips and Brown (1978, 1984) were consulted in order to match portions of single shells. the term ?parent? is used to refer to the largest portion of shell or the one having a unique catalog number, and it will be this parent fragment that will be given a data form. other fragments included under the parent were listed in the section devoted to my comments. Descriptions of design use an orientation with the dis- tal end down, the apex up, and the shell lying on the aper- ture. Right and left designations used in descriptions refer to this orientation. what Phillips and Brown (1978) refer to as spire is referred to in the present study as apex. inner always refers to the inner surface of the shell. For incomplete vessels, width refers to minimum width in order to give a width measurement that would identify the piece but would not be presumed to consti- tute the actual width of the vessel. For fragments, maxi- mum width is the maximum width of the largest fragment within a lot. Maximum length is the length of the longest fragment in a lot. Fragments were measured along the axis of the original shell, with length always referring to the shell axial length. therefore, a piece might be wider than it is long. Measurements could not be taken for all fragments. however, maximum or minimum widths are sometimes given to indicate the general size of the arti- facts. these maximum and minimum dimensions give an idea of the range of fragment sizes represented. these are not to be interpreted as vessel dimensions because such fragments rarely are complete enough to allow for esti- mating vessel length or width. Fragment counts are the total number of fragments including components of pieces which have been glued together (necessary because objects often comprise several nMnh catalog numbers). weights were always taken, but in the case of combined fragment groupings, weights were added together. all engraved shell cups and gorgets were examined and, if needed, conserved by Joan gardner between 1974 and 1976. in 1992 the shells were in stable condition for the most part. gardner described and recorded the condition of the shells in 1992. shell condition can be ranked as follows. ? Poor condition implies a badly eroded, pitted, and porous surface. chalking and flaking occur over 75% of the surface area. ? Fair condition implies eroded surfaces of flaky, chalking, and porous areas accounting for 50% of the surface. ? Adequate condition implies some fairly deterio- rated spots affecting between 25% and 50% of the surface area. ? Good condition presents very solid surfaces over- all, with few chalking or flaky areas. N U M B E R 4 9 ? 18 5 ? Excellent condition requires no flaking or chalky areas. the history of the engraved shells prior to gardner?s treatment remains largely unknown. gardner noted excess polyvinyl acetate (PVa), which had in many cases yellowed on the specimens. she removed excess PVa with acetone and reapplied a thinner coat. this thick coat of PVa may have been applied by the collectors. when Meyer?s collec- tion reached the smithsonian in the late 1960s, Elizabeth gibson was the conservator who examined and treated the collection. given how yellowed gardner found the shells, it seems unlikely that the PVa applied at the time of accession would have altered that drastically in the 6- to 15- year interval between trowbridge?s and Meyer?s dona- tions and 1974 when gardner examined the collection. the collectors could easily have treated objects them- selves, leaving 20 to 30 years for the PVa to become yel- lowed. Furthermore, we have no way of knowing exactly what type of conservative the collectors might have used, and therefore, they may have applied other sealants such as shellac or varnish, which would be expected to become yellow. Joan gardner reported treating three unengraved shells from the trowbridge accession but stipulates only a specific two that she treated. tExtilEs Measurements are not exact, but are approximations. Rather than risk damaging fragments, textile fragments were measured on top of the glass or Plexiglas that covers them. in cases where there are several fragments, the larg- est in the group was measured. the dimensions therefore represent maximum dimensions. weights were not taken for textiles because of the fragility of the artifacts. in most cases thickness was not measured. width of the primary yarns or fibers used in both textiles and basketry was sam- pled. in the case of basket fragments, the length of the longest single strip present was measured. the textile materials have undergone two major con- servation efforts during the past 20 years. Joan gardner worked with the collection in the mid- 1970s. at that time she unfurled several of the bundles, cleaned and applied preservatives to fragments that had already been un- folded, and prepared new mounts for others. at the time she finished, there were still objects stored in trowbridge?s original boxes. in the early 1990s, Lynn arden, contract conservator, prepared new box mounts: sturdy acid- free paperboard boxes with glass insets in the lids. the objects were placed in the mounts on foam that was covered with a smooth synthetic fabric. the artifacts are held in place by the lid, which places gentle pressure on the artifact, keeping it from shifting within the mount. condition of the textiles varies. some are in excellent condition, meaning that they retain the integrity of the original textile and that color and weave are clearly vis- ible. others are highly fragmentary. still others remain as bundles. there has been some difficulty in maintaining the dis- crete nature of many of these textiles. For example, cata- log numbers 423373 and 423372 contain many different fragments of twined tapestry done in multiple colors. number 423373 represents pieces in trowbridge?s box labeled 2719. Unfortunately, over the years the integrity of his designations has been compromised, and in some cases, the linkage between trowbridge?s number and the correct fragment is unclear. some bundles were unfolded during the 1970s, and the resulting fragments were some- times placed in Plexiglas and sometimes not. baskEts Discrete fragments were counted within groupings when possible. small fragments were grouped into lots. ProjECtilE Points in his study of the artifacts from the craig Mound, Brown (1976, 1996) used a variety of attributes to charac- terize the small projectile points and then created a key for point identification based on these attributes. Many of the same measurements were used in the present study. EarsPools the following measurements were taken for pulley- shaped earspools. 1. inner flange diameter is the diameter of the flange that forms the back of the spool. this is usually smaller than the front or outer flange. 2. outer flange diameter is the diameter of the outer flange, the one meant to be seen. it is the outer flange that is often decorated. 3. thickness is the maximum thickness of the entire spool. 4. Diameter of perforation can be used to separate the pulley ring forms from the pulleys having a narrow central bore. 5. wall thickness is average flange thickness, mea- sured 1 cm from the edge of the outer flange. 1 8 6 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y 6. Pulley width is the core diameter measured be- tween the flanges. bEads shells that were cemented together were counted sep- arately, so in some cases, the number in the lot will be higher than the total number of discrete pieces. Lot count is a minimum number of beads rather than a piece count. Beads were so numerous that measurements on indi- vidual shell beads were in most cases not taken. in many cases, individual lots of beads number into the thousands. Rather than take individual measurements, groups within each morphological bead class were sorted by size, and representative measurements were then taken to arrive at an approximate mean for the group. For large lots, count was estimated by dividing the total weight by the weight of 100 beads, then multiplying by 100. it is estimated that the total count will be off by no more than 10% by using this method. For beads of generally rounded shape, the width measurement was consistently taken at the widest point. Beads that were rounded but not truly round in cross section were also measured at the widest point. PiPEs Bore diameters were measured approximately 1 cm into the bore. Bowl diameter is the inside diameter of the bowl. in the case of unfinished pipe 423159, the bowl diameter represents the outer diameter. the width of the bowl is measured approximately 4 cm up from the base. this is generally the widest portion of the t- shaped pipe. note on cataloG SuPPlement there are 1464 separate analytical forms covering over 630 catalog numbers listed in the nMnh catalog. these forms, along with history forms for each catalog number, are contained in an extended catalog supple- ment kept in the Department of anthropology collec- tions management records. the catalog is the primary data source for this collection and can be utilized as a collection guide and analytical tool for studying the col- lection. there is one history form for each nMnh cata- log number, along with one or more data forms. a single template was used to record historical information. al- ternate data collection templates were used in order to accommodate different material types and manufactured goods. the history form contains historical information about the pieces, including extra conservation notes and descriptions made by Joan gardner. the data forms pres- ent information collected by the author of the present study. invEntory of data forms table a.1 presents a summary of the collection listed by catalog number. this summary serves as a quick refer- ence source and as an inventory of data forms. Data forms generated in the database are listed by catalog number fol- lowed by subnumbers and abbreviations indicating ma- terial types. also included are object name (which may differ from the object name recorded in the nMnh elec- tronic database), style (usually abbreviated), count, raw material, and template code. TABLE A.1. summary of artifacts in the spiro collection. Material type codes are as follows: t, textile; s, shell; L, lithic; c, ceramic; M, metal; B, bone; w, wood; o, other. other abbreviations are s Point, small projectile point; L Point, large projectile point; Lg Biface, large biface; frag, fragment; convexo-cylin, convexo-cylindrical; concavo-cylin, concavo-cylindrical; Undesig., undesignated; Micac., micaceous. Catalog Subnumber Material Raw material number (if any) type Object Style Count (if known) 378240 s01 t cordage Braid 1 378240 s02 t cloth spaced weft twining 1 378240 s03 t cordage wrapped warp 1 378240 s04 t cordage wrapped warp barrel-shaped 2 378240 s05 t cordage Braid 2 378240 s06 t sample slide 2 378240 s07 t sample slide 8 N U M B E R 4 9 ? 18 7 Catalog Subnumber Material Raw material number (if any) type Object Style Count (if known) (continued) 378242 s cup Unengraved 1 shell 378243 s01 s cup Unengraved 1 shell 378243 s02 s cup Unengraved 1 shell 378243 s03 s cup Unengraved 2 shell 378243 s04 s cup Unengraved 1 shell 378244 s Pendant columella 2 shell 378245 s worked shell Unengraved 1 shell 378246 s gorget type 9 1 Univalve 378247 s01 s cup Unengraved 1 cassis 378247 s02 s cup Unengraved 1 cassis 378248 s cup Unengraved 1 Univalve 378249 s Pendant Unengraved 1 shell 378250 s01 s Bead gastropod 3 snail 378250 s02 s Bead gastropod 1 Freshwater snail 378250 s03 s Bead gastropod 2 Freshwater snail 378250 s04 s Bead gastropod 1 Freshwater snail 378251 s Bead Marginella 10 Univalve 378252 s Bead Olivella 45 shell 378253 s01 s Bead hexagonal 2 shell 378253 s02 s Bead convexo-bowed 4 378253 s03 s Bead convexo-bowed 1 378254 s01 s Bead cylindrical 1 shell 378254 s02 s Bead cylindrical 1 shell 378254 s03 s Bead convexo-cylin 6 shell 378254 s04 s Bead spherical 1 shell 378254 s05 s Bead convexo-cylin-columella 1 shell 378254 s06 s Bead convexo-cylin-columella 1 shell 378255 s01 s Bead Disc 1 378255 s02 s Bead Disc 1 378255 s03 s Bead Elliptical 2 378255 s04 s Bead Elliptical 1 Pearl 378255 s05 s Bead Elliptical 1 shell 378256 s01 s Bead cylindrical 15 shell 378256 s02 s Bead Elliptical 16 shell 378256 s03 s Bead Disc 440 shell 378256 s04 s Bead Disc 278 shell 378256 s05 s Bead Disc 23 shell 378256 s06 s Bead Disc 92 shell 378256 s07 s Bead Elliptical 30 shell 378256 s08 s Bead spherical 20 shell 378257 s01 s Bead convexo-cylin 61 shell 378257 s02 s Bead convexo-cylin 398 shell 378257 s03 s Bead convexo-cylin 10 shell 378257 s04 s Bead convexo-cylin 5 shell 378257 s05 s Bead convexo-cylin 1 shell 378257 s06 s Bead spherical 5 shell 378258 s01 s Bead Bulging 1 shell 378258 s02 s Bead Bulging 4 shell 378259 s Bead concavo-cylin 4 shell 378260 s Bead spherical pearl 104 Pearl 378261 L Bead cylindrical 1 Phosphate 378262 s gorget type 1 1 shell 378263 s Pendant carved 1 378264 B Pendant tooth 1 antler 378265 s worked frag inset 1 378266 M Bead Fraudulent-disc 4 copper 378267 M, t, o sheet Unclassified 11 copper 378268 M Pin Fraudulent 2 copper 378269 L Earspool Unperforated 2 Limestone 378270 L cube Unmodified 1 galena 378271 s01 L s Point Unclassified 1 chert 1 8 8 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y Catalog Subnumber Material Raw material number (if any) type Object Style Count (if known) 378271 s02 L s Point ashley 1 chert 378271 s03 L s Point Fresno 1 chert 378271 s04 L s Point Massard 1 chert 378271 s05 L s Point Unclassified 1 chert 378272 L L Point Lange 1 chert 378273 L scraper Uniface 1 obsidian 379212 B awl 1 os penis 379213 B awl 1 Long bone 379214 B Pin tapering 7 Bone 379215 c Pipe Red River 2 clay 379216 s Bead concavo-cylin 1 379217 s01 s Bead Disc 5 shell 379217 s02 s Bead convexo-cylin 4 shell 379217 s03 s Bead concavo-cylin 1 shell 379218 s01 s Bead Disc 1 siltstone 379218 s02 L Bead Disc 1 Phosphate 379218 s03 L Bead Disc 1 sandstone 379218 s04 L Bead Disc 1 sedimentary 379218 s05 L Bead Disc 1 sedimentary 379218 s06 L Bead convexo-cylin 1 Phosphate sphericalate 379218 s07 L Bead Disc 1 sedimentary 379218 s08 L Bead Disc 1 sedimentary 386185 s01 t, M Basket twill 1 wicker 386186 t Basketry twill 1 386187 s01 t cloth tapestry 1 386188 s01 t, o cordage strand 2 hair 386188 s02 t, o cloth wrapped warp 3 Feather 386188 s03 t, o cordage tapestry 1 hair 386188 s04 t, o sample slide 1 Feather 397718 s01 s Bead convexo-cylin 106 397718 s02 s Bead Disc 105 397719 M Pin Fraudulent 2 copper 407190 s shell cast 1 Plaster 417938 s01 c sherd williams Plain 10 clay 417938 s02 c sherd Leflore Plain 5 clay 417938 s03 c sherd sanders Plain 1 clay 417938 s04 c sherd woodward Plain 3 clay 417938 s05 L Pipe Unclassified 1 sandstone 417939 s01 L Lg Biface ovoid a 1 chert 417939 s02 L L Point gary 1 chert 417940 L Mano Pitted 1 sandstone 417941 s Bead Disc 1 shell 423144 c Vessel smithport Plain 1 clay 423145 c Vessel Redware-coles creek variation 1 clay 423146 c Vessel sanders Plain 1 clay 423147 c Vessel crockett curvilinear incised 1 clay 423148 c Vessel white Engraved 1 clay 423149 L Vessel Effigy 1 Marble 423150 L Vessel Effigy 1 Marble 423151 s01 c sherd sanders Plain 1 clay 423151 s02 c sherd sanders Plain 1 clay 423151 s03 c sherd williams Plain 1 clay 423151 s04 c sherd sanders Plain 1 clay 423151 s05 c sherd Undesig. Plain grog 1 clay 423151 s06 c sherd williams Plain 1 clay 423151 s07 c sherd southwestern 6 clay 423151 s08 c sherd Unclassified 3 clay 423151 s09 c sherd Polychrome 1 clay 423151 s10 c sherd glazed 1 clay 423152 L Pipe Effigy-irregular 1 catlinite N U M B E R 4 9 ? 18 9 Catalog Subnumber Material Raw material number (if any) type Object Style Count (if known) (continued) 423154 L Pipe t-shaped, cylindrical bowl or 1 siltstone/ chandler sand stone 423155 s01 L Pipe t unclassified 1 siltstone 423155 s02 c Pipe Unclassified 1 clay 423155 s03 L Pipe t-shaped, unclassified 1 siltstone 423155 s04 L Pipe t-shaped, unclassified 1 siltstone 423156 L Pipe Elbow 1 siltstone 423157 L Pipe t-shaped, cylindrical bowl 1 siltstone 423158 L Pipe t-shaped, cylindrical bowl 3 siltstone 423159 L Pipe t-shaped, cylindrical/contracting 1 siltstone 423160 L Pipe Effigy 1 Limestone 423161 L Earspool Perforated 1 sandstone 423162 L Earspool Perforated 1 sandstone 423163 L Earspool Perforated 1 sandstone 423164 L Earspool Perforated 1 Limestone 423165 L Earspool Divided flange 1 Limestone 423166 L Pendant center perforated 1 siltstone 423167 L Pendant Perforated at end 1 coal 423168 L Pendant Perforated at end 1 chert 423169 s01 L Bead cylindrical 1 Phosphate 423169 s02 s Bead convexo-cylin 1 shell 423169 s03 s Bead convexo-cylin 1 shell 423169 s04 s Bead convexo-cylin 1 shell 423169 s05 L Bead convexo Disc 1 Phosphate 423170 s01 L Lg Biface celt form 1 novaculite 423170 s02 L Lg Biface celt form 1 novaculite 423170 s03 L Lg Biface celt form 1 chert 423171 L Lg Biface Fusi-elliptical or Ramey 1 Quartzite 423172 s01 L Lg Biface sword form, Duck River 1 chert 423173 L Lg Biface celt 1 chert 423174 s01 L Lg Biface sword form, general 1 chert 423174 s02 L Lg Biface hoe 1 chert 423174 s03 L Lg Biface sword form, general 1 chert 423174 s04 L Lg Biface sword form, Duck River 1 chert 423174 s05 L Lg Biface sword form, Duck River 1 chert 423174 s06 L Lg Biface Unclassified 1 chert 423174 s07 L Lg Biface sword form, Duck River 1 chert 423174 s08 L Lg Biface sword form, Duck River 1 chert 423175 s01 L L Point Lange 1 chert 423175 s02 L Lg Biface Kay (like) 1 chert 423176 L L Point Lange 1 chert 423177 L Lg Biface Unclassified 1 chert 423178 L Lg Biface ovoid a 1 chert 423179 L Lg Biface ovoid irregular 1 novaculite 423180 L Lg Biface Preform 1 novaculite 423181 s01 L s Point agee 1 novaculite 423181 s02 L s Point Massard 1 novaculite 423182 s01 L s Point agee 1 novaculite 423182 s02 L s Point agee 1 novaculite 423182 s03 L s Point agee 1 novaculite 423183 s01 L s Point Massard 1 chert 423183 s02 L s Point Reed 1 chert 423183 s03 L s Point Massard 1 calcedony 423184 L s Point sequoyah 1 novaculite 423185 s01 L s Point agee 1 chert 423185 s02 L s Point Morris 1 chert 423186 L s Point agee 1 novaculite 423187 L s Point ashley chocoville 1 chert 423188 L s Point agee 1 Quartz crystal 423189 L L Point Johnson 1 chert 423190 s01 L L Point Unclassified side notch 1 chert 1 9 0 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y Catalog Subnumber Material Raw material number (if any) type Object Style Count (if known) 423190 s02 L L Point Expanding B 1 chert 423191 s01 L Drill Expanding 1 novaculite 423191 s02 L s Point Uniface 1 chert 423191 s03 L Lg Biface sword form 1 siltstone 423191 s04 L Uniface Projection 1 Quartz crystal 423192 s01 L s Point agee 1 novaculite 423192 s02 L s Point agee 1 novaculite 423192 s03 L s Point agee 1 chert 423192 s04 L s Point agee 1 chert 423192 s05 L s Point agee 1 Unknown 423192 s06 L s Point agee 1 chert 423192 s07 L s Point agee 1 chert 423192 s08 L s Point agee 1 Jasper 423192 s09 L s Point agee 1 chert 423192 s10 L s Point scallorn 1 chert 423192 s11 L s Point scallorn 1 novaculite 423192 s12 L s Point scallorn 1 chert 423192 s13 L s Point Unclassified corner notch 1 chert 423192 s14 L s Point Unclassified corner notch 1 chert 423192 s15 L s Point Unclassified corner notch 1 chert 423192 s16 L s Point Maud 1 chert 423192 s17 L s Point ashley 1 chert 423192 s18 L s Point ashley 1 Jasper 423192 s19 L s Point hayes 1 Jasper 423192 s20 L s Point hayes 1 Jasper (banded) 423192 s21 L s Point hayes 1 chert 423192 s22 L s Point hayes 1 Jasper 423192 s23 L s Point alba 1 chert 423192 s24 L s Point Perdiz 1 chert 423192 s25 L s Point Perdiz 1 chert 423192 s26 L s Point Perdiz 1 chert 423192 s27 L s Point Perdiz 1 chert 423192 s28 L s Point Perdiz 1 chert 423192 s29 L s Point Perdiz 1 chert 423192 s30 L s Point Bonham tahlequah 1 chert 423192 s31 L s Point alba 1 chert 423192 s32 L s Point hayes 1 chert 423192 s33 L s Point hayes 1 Jasper 423192 s34 L s Point ashley 1 novaculite 423192 s35 L s Point hayes 1 chert 423192 s36 L s Point Martindale 1 chert 423192 s37 L s Point haskell 1 chert 423192 s38 L s Point washita Peno 1 chert 423192 s39 L s Point scallorn 1 novaculite 423192 s40 L s Point Reed/Keota variety 1 novaculite/ chert 423192 s41 L s Point Massard 1 chert 423192 s42 L s Point Massard 1 chert 423192 s43 L s Point scallorn 1 chert 423192 s44 L s Point Reed/Keota variety 1 novaculite 423192 s45 L s Point Reed/Keota variety 1 chert 423192 s46 L s Point Reed 1 chert 423192 s47 L s Point Reed 1 chert 423192 s48 L s Point Reed 1 chert 423192 s49 L s Point Keota 1 chert 423192 s50 L s Point Reed 1 chert 423192 s51 L s Point Reed/Keota variety 1 chert 423192 s52 L s Point Reed 1 novaculite 423192 s53 L s Point Reed 1 novaculite 423192 s54 L s Point scallorn 1 chert 423192 s55 L s Point Reed/Keota variety 1 chert N U M B E R 4 9 ? 19 1 Catalog Subnumber Material Raw material number (if any) type Object Style Count (if known) (continued) 423192 s56 L s Point Keota 1 chert 423192 s57 L s Point Reed 1 chert 423192 s58 L s Point Reed 1 chert 423192 s59 L s Point scallorn 1 chert 423192 s60 L s Point scallorn 1 chert 423192 s61 L s Point collier 1 novaculite 423192 s62 L s Point Keota 1 chert 423192 s63 L s Point scallorn 1 chert 423192 s64 L s Point Reed/Keota variety 1 novaculite 423192 s65 L s Point Reed/Keota variety 1 chert 423192 s66 L s Point Reed/Keota variety 1 novaculite 423192 s67 L s Point Reed/Keota variety 1 novaculite 423192 s68 L s Point Keota 1 chert 423192 s69 L s Point Reed/Keota variety 1 chert 423192 s70 L s Point alba 1 chert 423192 s71 L s Point Reed/Keota variety 1 chert 423193 s01 L Drill Unclassified 1 chert 423193 s02 L Drill Expanding 1 chert 423194 L celt shouldered spade 1 chert 423195 L Mace type 5 1 chert 423196 L Mace Unclassified 1 chert 423197 L Mace type 2 1 chert 423198 L Mace type 1 1 chert 423200 L celt Unclassified 1 sedimentary 423201 L celt Elongate 1 sedimentary 423202 L abrader Polishing stone 1 hematite 423203 L abrader Polishing stone 1 Quartzite 423204 L Mano Pitted 1 Quartzite 423205 L abrader sharpener 1 sandstone 423206 L abrader shaft abrader 1 igneous? 423207 s01 L galena nodule 1 galena 423207 s02 L galena nodule 1 galena 423207 s03 L galena nodule 1 galena 423207 s04 L galena worked 1 galena 423208 L Quartz Unmodified 1 Quartz crystal 423209 s01 o Pigment worked a ocher 423209 s02 o Pigment chunk 1 ocher 423210 L slag Unmodified 1 ash/bone/glass 423211 L Quartz crystal 1 Quartz crystal 423212 L Mineral lump 1 Micac. sandstone 423213 L Pigment Powder 1 ocher 423214 L Pigment Unmodified 15 glauconite 423215 L Mica worked 3 Mica 423216 M Bead Fraudulent-disc 20 copper 423217 M Pin Fraudulent 4 copper 423218 s01 M sheet Plume? 1 copper 423218 s02 M sheet Unclassified 9 copper 423219 M, t sheet Unclassified 3 copper 423220 M sheet Unclassified 1 copper 423221 s01 M sheet Perforated 1 copper 423221 s02 M sheet Perforated 5 copper 423221 s03 B sheet Perforated 2 Bone 423221 s04 M sheet Embossed 12 copper 423222 M sheet Embossed 70 copper 423223 M, t sheet 1 copper 423224 s01.1 M, t, s sheet composites 6 copper 423224 s01.2 M, t, s Basket twill 5 wicker 423225 s01 M, t sheet Unclassified 10 copper 423225 s02 M, t sheet Unclassified 4 copper 423226 s01.1 M composite Disc 1 copper 423226 s01.2 s Bead Disc 1 shell 1 9 2 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y Catalog Subnumber Material Raw material number (if any) type Object Style Count (if known) 423227 s01.1 M, t, s, w composite Unclassified 6 copper 423227 s01.2 s Bead columella 3 shell 423227 s01.3 t Mat twill 1 wicker 423227 s01.4 t cordage 2 ply Z-twist 1 Fiber 423228 s01 s cup Unengraved 1 Univalve 423228 s02 s cup Unengraved 1 Univalve 423228 s03 s cup Unengraved 1 Univalve 423228 s04 s cup Unengraved 1 Univalve 423228 s05 s cup Unengraved 1 Univalve 423228 s06 s cup Unengraved 1 Univalve 423228 s07 s cup Unengraved 1 Univalve 423228 s08 s cup Unengraved 1 Univalve 423228 s09 s cup Unengraved 1 Univalve 423228 s10 s cup Unengraved 1 Univalve 423228 s11 s cup Unengraved 1 Univalve 423228 s12 s cup Unengraved 1 Univalve 423228 s13 s cup Unengraved 1 Univalve 423228 s14 s cup Unengraved 1 Univalve 423228 s15 s cup Unengraved 1 Univalve 423228 s16 s cup Unengraved 1 Univalve 423228 s17 s cup Unengraved 1 Univalve 423228 s18 s cup Unengraved 1 Univalve 423228 s19 s cup Unengraved 1 Univalve 423228 s20 s cup Unengraved 1 Univalve 423228 s21 s cup Unengraved 1 Univalve 423228 s22 s cup Unengraved 1 Univalve 423228 s23 s cup Unengraved 1 Univalve 423228 s24 s cup Unengraved 1 Univalve 423228 s25 s cup Unengraved 1 Univalve 423228 s26 s cup Unengraved 1 Univalve 423228 s27 s cup Unengraved 1 Univalve 423228 s28 s cup Unengraved 1 Univalve 423228 s29 s cup Unengraved 1 Univalve 423228 s30 s cup Unengraved 1 Univalve 423228 s31 s cup Unengraved 1 Univalve 423229 s01 s cup Unengraved 4 Univalve 423229 s02 s cup Unengraved 1 shell 423230 s01 s cup Unengraved 1 shell 423230 s02 s cup Unengraved 1 shell 423231 s Pendant Drop 1 shell 423232 s01 s, t cup Unengraved 1 Univalve 423232 s02 s, t cup Unengraved 1 Univalve 423232 s03 s cup Unengraved-sculptured 1 Univalve 423232 s04 s, t cup Unengraved 1 Univalve 423233 s01.1 s, t cup Unengraved 1 Univalve 423233 s01.2 s, t cordage 2-ply Z-twist 1 423234 s cup craig c 1 Univalve 423235 s cup Braden B 1 Univalve 423236 s cup craig B 1 Univalve 423237 s cup Braden c 1 Univalve 423238 s cup craig c 1 Univalve 423239 s cup craig c 1 Univalve 423240 s cup Braden c 1 Univalve 423241 s cup craig c 1 Univalve 423242 s cup Braden c 1 Univalve 423243 s cup craig c 1 Univalve 423244 s cup Braden c 1 Univalve 423245 s cup craig c 1 Univalve 423246 s cup craig c 1 Univalve 423247 s cup craig c 1 Univalve 423248 s cup craig B 1 Univalve N U M B E R 4 9 ? 19 3 Catalog Subnumber Material Raw material number (if any) type Object Style Count (if known) (continued) 423249 s cup Braden B 1 Univalve 423250 s cup Braden c 1 Univalve 423251 s cup craig B 2 Univalve 423252 s cup Braden c 1 Univalve 423253 s cup craig c 1 Univalve 423254 s cup craig B 1 Univalve 423255 s01 s cup craig B 1 Univalve 423256 s cup craig c 1 Univalve 423257 s cup Braden B 2 Univalve 423258 s cup craig B 2 Univalve 423259 s cup Braden c 2 Univalve 423260 s cup Braden B 1 Univalve 423261 s cup craig B 1 Univalve 423262 s cup craig c 1 Univalve 423263 s cup Braden B 1 Univalve 423264 s cup craig B 2 Univalve 423266 s cup Braden B 1 shell 423267 s cup craig c 1 Univalve 423268 s cup Unclassified 1 Univalve 423269 s cup Braden c 1 Univalve 423270 s cup craig B 1 Univalve 423271 s cup craig B 1 Univalve 423273 s cup craig c 1 Univalve 423274 s cup Braden a 1 Univalve 423276 s cup craig B 1 Univalve 423277 s cup Braden B 1 Univalve 423279 s cup craig a 5 Univalve 423281 s cup Braden B 2 Univalve 423282 s cup craig c 1 Univalve 423283 s cup craig c 1 Univalve 423284 s cup craig B 1 Univalve 423285 s cup Unclassified 1 Univalve 423287 s cup craig B 2 Univalve 423288 s cup craig a 1 Univalve 423289 s cup craig B 1 Univalve 423290 s cup craig B 2 Univalve 423291 s cup Braden B 2 Univalve 423292 s gorget craig B, type 1 2 Univalve 423293 s gorget craig a, type 2 1 Univalve 423294 s gorget type 5 1 Univalve 423295 s gorget-mask type 7 3 shell 423296 s cup craig a 3 Univalve 423298 s01 s cameo Maskette 1 shell 423299 s cameo Maskette 1 shell 423300 s cameo Maskette 1 shell 423301 s cameo Maskette 1 shell 423302 s01 s worked shell Unengraved 1 Univalve 423302 s02 s object Unmodified 1 shell 423303 s, B Bead Disc 3 shell 423304 s Barnacle Unmodified 1 shell 423305 s01 s Pendant columella 1 shell 423305 s02 s Pendant columella 1 shell 423305 s03 s Pendant columella 1 shell 423305 s04 s Pendant columella 1 shell 423305 s05 s Pendant columella 1 shell 423305 s06 s Pendant columella 1 shell 423305 s07 s Pendant columella 1 shell 423305 s08 s Bead Disc 6 shell 423305 s09 s Pendant columella 1 shell 423305 s10 s Pendant columella 1 shell 423305 s11 s Pendant columella 1 shell 1 9 4 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y Catalog Subnumber Material Raw material number (if any) type Object Style Count (if known) 423305 s12 s Pendant columella 1 shell 423305 s13 s Pendant columella 1 shell 423305 s14 s Pendant columella 1 shell 423305 s15 s Pendant columella 1 shell 423305 s16 s Pendant columella 1 shell 423305 s17 s Pendant columella 1 shell 423305 s18 s Pendant columella 1 shell 423305 s19 s Pendant columella 1 shell 423305 s20 s Pendant columella 1 shell 423305 s21 s Pendant columella 1 shell 423305 s22 s Pendant columella 1 shell 423306 s01 s, M Bead spherical 1 shell 423306 s02 s Bead Disc 1 shell 423307 s01.1 s, t Bead Elliptical 7 shell 423307 s01.2 t cordage 2-ply s-twist 1 Fiber 423308 s01 s gorget type 9 1 Univalve 423308 s02 s Bead Disc 31 shell 423308 s03 s Bead Baroque 1 Pearl 423308 s04 s sediment 1 Unknown 423309 s01 s Bead Disc 31 shell 423309 s02 s Bead Disc 1,020 shell 423310 s Bead Disc 2 shell 423311 s Bead convexo disc 2 shell 423312 s Disk Disc 1 shell 423313 s Bead concavo disc 64 shell 423314 s01 s Bead spherical-columella 23 shell 423314 s02 s Bead spherical-columella 48 shell 423315 s Bead spherical 1 shell 423316 s01 s Bead cylindrical 478 shell 423316 s02 s Bead cylindrical 36 shell 423317 s01 s Bead convexo-cylin 81 shell 423317 s02 s Bead Disc 13 shell 423318 s01 s Bead concavo 117 shell 423318 s02 s Bead convexo-cylin 5 shell 423319 s Bead concavo 6 shell 423320 s Bead convexo-cylin 16 shell 423321 s Bead convexo-cowed 53 shell 423322 s Bead convexo-cylin-columella 8 shell 423323 s01 s Bead convexo-cylin-sculptured 1 Univalve 423323 s02 s Bead convexo-cylin-columella 2 shell 423324 s01 s Bead convexo-cylin 42 shell 423324 s02 s Bead Disc 42 shell 423324 s03 s Bead olivella 1 shell 423325 s01 s Bead Disc 52 shell 423325 s02 s Bead convexo-bowed 2 shell 423325 s03 s Bead convexo-cylin-columella 1 Univalve 423325 s04 s Bead convexo-cylin 17 shell 423325 s05 s Bead concavo 1 shell 423325 s06 s Bead spherical-columella 2 Univalve 423325 s07 s Bead Disc 6 shell 423325 s08 s Bead spherical 12 shell 423325 s09 s Bead Olivella 2 shell 423326 s01 s Bead concavo 4 shell 423326 s02 s Bead convexo-bowed 6 shell 423326 s03 s Bead convexo-cylin-columella 1 Univalve 423326 s04 s Bead spherical-columella 2 Univalve 423326 s05 s Bead spherical 7 shell 423326 s06 s Bead convexo-cylin-columella 9 Univalve 423326 s07 s Bead convexo-cylin-columella 55 Univalve 423327 s Bead Bulging 108 shell 423328 s Bead Bulging-columella 1 Univalve N U M B E R 4 9 ? 19 5 Catalog Subnumber Material Raw material number (if any) type Object Style Count (if known) (continued) 423329 s01 s Bead tooth 3 shell 423329 s02 s Bead convexo disc 1 shell 423329 s03 s Bead Elliptical 3 shell 423329 s04 s Bead spherical 1 shell 423330 s01 s Bead Disc 5,715 shell 423330 s02 s Bead convexo-cylin 1,545 shell 423330 s03 s Bead spherical 28 shell 423330 s04 s Bead spherical pearl 2 Pearl 423330 s05 s Bead Olivella 4 shell 423330 s06 B, w, o sediment 30 Bone 423330 s07 s Bead Marginella 1 shell 423330 s08 s Bead Elliptical 5 shell 423331 s01 s Bead olivella 2,750 shell 423331 s02 o sediment 1 soil/bone/ wood 423331 s03 L galena nodule 1,400 galena 423331 s04 s Bead Disc 980 shell 423331 s05 s Bead Disc-thin 96 shell 423331 s06 s Bead convexo-cylin 4 shell 423331 s07 s Bead cylindrical 55 shell 423331 s08 L Mica Unmodified 3 Mica 423331 s09 L Quartz Unmodified 32 Quartz 423332 s Bead spherical pearl 94 Pearl 423333 s Bead spherical pearl 8 Pearl 423334 s Bead spherical pearl 99 Pearl 423335 s Bead Baroque 10 Pearl 423336 s Bead spherical pearl 5 Pearl 423337 s01 s Bead spherical pearl 18 Pearl 423337 s02 s Bead convexo-cylin-sculptured 2 Pearl 423337 s03 s Bead tooth 1 Pearl 423338 s Bead Baroque 63 Pearl 423339 s Bead Baroque 118 Pearl 423340 s Bead Baroque 1 Pearl 423341 B Pendant Drop 1 Bird 423342 B Pin tapering 1 Bone 423343 B gorget irregular 1 scapula 423344? 423351 b 423352 s01 t cloth spaced weft twining 1 hair 423352 s02 t cordage Fiber 1 hair 423352 s03 t cordage Braid-tassel 1 hair 423352 s04 t cordage strand 1 423353 s01 t cloth mantle spaced weft twining 1 hair 423353 s02 t sample slide 1 423354 s01 t cloth mantle spaced weft twining 1 hair 423354 s02 t cordage strand 1 hair 423354 s03 t sample slide 1 423355 t cloth mantle spaced weft twining 1 hair 423356 s01 t cloth spaced weft twining 1 hair 423356 s02 t cloth spaced weft twining 1 hair 423356 s03 t cordage strand 1 hair 423357 s01 t cloth spaced weft twining 1 hair 423357 s02 t cordage strand 1 hair 423358 s01 t cloth simple weave 1 hair 423358 s02 t cordage strand 1 hair 423358 s03 t cordage strand 1 hair 423358 s04 t sample 2 423359 t cloth spaced weft twining 1 hair 423360 t cloth bundle spaced weft twining 1 hair 423360 s02 t sample slide 1 423361 t cloth bundle spaced weft twining 1 hair 1 9 6 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y Catalog Subnumber Material Raw material number (if any) type Object Style Count (if known) 423362 t cloth bundle spaced weft twining 1 423363 s01 t cloth bundle spaced weft twining 1 423363 s02 t sample slide 1 423364 t cloth bundle spaced weft twining 2 hair 423365 t cloth bundle wrapped warp 6 Feather 423366 t Bundle spaced weft twining 5 423367 s01 t cloth bundle wrapped warp 1 Feather 423367 s02 t cloth bundle wrapped warp 1 Feather 423367 s03 t cloth bundle spaced weft twining 1 hair 423367 s04 t cloth bundle wrapped warp 1 Feather 423367 s05 t sample strand 1 423367 s06 t cloth Unclassified 2 423367 s07 t cloth spaced weft twining 1 423368 s01 t cloth spaced weft/wrapped warp 6 423368 s02 t cloth wrapped warp 6 Feather 423369 s01 t cloth wrapped warp 8 Feather 423369 s02 t cloth spaced weft twining 6 423369 s03 t cloth spaced weft twining 7 Vegetal 423369 s04 t cloth spaced weft twining 2 hair 423369 s05 t cordage Braid-tassel 2 hair? 423369 s06 t cloth wrapped warp 2 Feather 423370 s01 t cloth wrapped warp 2 Feather 423370 s02 t, o cordage, cloth wrapped warp, braid 8 Feather 423371 s01 t cloth wrapped warp 8 Feather 423371 s02 t cloth wrapped warp 9 Feather 423371 s03 t cloth wrapped warp 8 Feather 423372 s01 t cloth tapestry 1 hair 423372 s02 t cloth tapestry 2 Feather 423372 s03 t cordage wrapped basketry 1 hair 423372 s04 t cloth tapestry 2 423372 s05 t cloth tapestry, wrapped warp 9 hair 423372 s06 t cordage wrapped basketry 3 hair 423372 s07 t cloth wrapped warp 1 Feather 423372 s08 t cloth tapestry 2 hair 423372 s09 t cloth spaced weft twining 1 hair 423372 s10 t cloth wrapped warp 1 Feather 423372 s11 t cloth tapestry 2 hair 423372 s12 t cloth tapestry 1 hair 423372 s13 t cloth tapestry 1 hair 423372 s14 t sample 1 hair 423373 s01 t cloth tapestry 1 hair 423373 s02 t cloth tapestry 1 423373 s03 t cloth tapestry 5 hair 423373 s04 t cloth tapestry 3 hair 423373 s05 t cloth tapestry 2 hair 423373 s06 t cordage tapestry, wrapped warp 1 423373 s07 t cloth tapestry 4 hair 423373 s08 t cloth tapestry 3 hair 423373 s09 t sample slide 1 423373 s10 t cloth tapestry 4 hair 423373 s11 t cloth, cordage tapestry, braid 5 hair 423373 s12 t cloth tapestry 2 hair 423373 s13 t cloth tapestry 1 hair 423373 s14 t cloth tapestry 1 hair 423373 s15 t cloth tapestry 1 hair 423373 s16 t cloth tapestry 1 hair 423373 s17 t cloth tapestry 2 hair 423373 s18 t cloth tapestry 2 hair 423373 s19 t cloth Braid, tapestry 1 hair 423373 s20 t sample slide 1 423373 s21 t cloth tapestry 1 hair N U M B E R 4 9 ? 19 7 Catalog Subnumber Material Raw material number (if any) type Object Style Count (if known) (continued) 423373 s22 t textile wrapped warp barrel-shaped 1 hair 423373 s23 t cloth wrapped warp 3 Feather 423373 s24 t cloth tapestry 1 hair 423373 s25 t cloth tapestry 2 hair 423373 s26 t cloth tapestry 1 hair 423373 s27 t cloth tapestry 1 hair 423374 s01 t cloth wrapped warp 10 Feather 423374 s02 t cordage Braid 3 hair? 423375 t cordage strand 2 423376 s01 t cordage 2-ply Z-twist 5 423376 s02 o hide 1 423377 s01 t cordage strand 12 423377 s02 t sample slide 2 423378 s01 t, s cloth wrapped warp, tapestry 9 hair 423378 s02 t cloth wrapped warp, braid 3 hair 423378 s03 t cordage strand 1 423378 s04 s Bead spherical pearl 4 Pearl 423378 s05 s Powder Unmodified 1 Lime 423378 s06 s cup Unclassified 1 shell 423378 s07 s Bead convexo-cylin 13 shell 423378 s08 s Bead Disc a shell 423378 s09 s Bead convexo-cylin a 423378 s10 s Bead convexo-bowed 10 423378 s11 M, o hide Leather 3 copper 423379 t cloth tapestry 1 Vegetal 423380 t cloth spaced weft twining 6 hair? 423381 t, w cordage 2-ply Z?twist 1 423382 t cordage-rope Braid 14 Possibly hair 423383 t cordage Loose fibers 1 423384 t, M cordage Braid 2 hair 423385 t, M Basketry Matting 3 423386 t Basketry twill 1 423387 t Basketry twill 1 423388 t Basketry twill 3 423389 t Basketry coil 1 423390 t Basketry coil 1 423391 w Point ceremonial 1 423392 w ornament worked 1 cedar 423393 w ornament worked 1 423394 w, M stick Variety 3 cedar 423395 w stick Unmodified 1 cedar 423396 w charcoal Unmodified 7 cedar 423397 w stick Unmodified 20 cedar 423398 w Pole Unmodified 1 cedar 423399 w Pole st 1 cedar 423400 w Pole taper 1 cedar 423401 w Pole taper 1 cedar 448641 s01 c Vessel sanders Plain?old town Red 6 clay 448641 s02 c Vessel sanders Plain?old town Red 11 clay 448641 s03 c Vessel sanders Plain?old town Red 3 clay 448641 s04 c Vessel sanders Plain?old town Red 4 clay 448641 s05 c sherd sanders Plain?old town Red 2 clay 448641 s06 c sherd sanders Plain?old town Red 6 clay 448641 s07 c sherd sanders Plain 9 clay 448641 s08 c sherd sanders Plain 8 clay 448641 s09 c sherd sanders Plain 39 clay 448641 s10 c sherd sanders Plain 28 clay 448641 s11 c sherd sanders Plain 1 clay 448641 s12 c sherd sanders Engraved 1 clay 448641 s13 c sherd sanders Engraved 1 clay 448641 s14 c sherd sanders Engraved 1 clay 1 9 8 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y Catalog Subnumber Material Raw material number (if any) type Object Style Count (if known) 448641 s15 c sherd sanders Engraved 2 clay 448641 s16 c sherd clement Redware/sanders 2 448641 s17 c sherd sanders Plain 2 clay 448641 s18 c sherd sanders Plain 1 clay 448641 s19 c sherd sanders Plain 1 clay 448641 s20 c sherd sanders Plain 3 clay 448641 s21 c sherd Poteau Plain 2 clay 448641 s22 c sherd Undesig. grog tempered 2 clay 448641 s23 c sherd Poteau Plain 7 clay 448641 s24 c sherd Poteau Plain 1 clay 448641 s25 c sherd Poteau Plain 4 clay 448641 s26 c sherd Poteau Plain 1 clay 448641 s27 c sherd Poteau Plain 1 clay 448641 s28 c sherd sanders Plain-old town Red 4 clay 448642 s01 c sherd Leflore Plain 8 clay 448642 s02 c sherd Undesignated grit 1 clay 448642 s03 c sherd Undesig. plain 2 clay 448642 s04 c sherd Leflore Plain 2 clay 448642 s05 c sherd Undesig. grog 3 clay 448642 s06 c sherd Leflore Plain 1 clay 448642 s07 c sherd Leflore Plain 2 clay 448642 s08 c sherd williams (?) 5 clay 448642 s09 c sherd Undesig. plain grog 1 clay 448642 s10 c sherd williams Plain 1 clay 448642 s11 c sherd Undesig. grog plain 2 clay 448642 s12 c sherd Undesig. shell tempered 2 clay 448642 s13 c sherd Undesig. shell tempered 2 clay 448642 s14 c sherd Poteau Plain 3 clay 448642 s15 c sherd hatchel Engraved 1 clay 448642 s16 c sherd haley Engraved 1 clay 448642 s17 c sherd Pennington Punctate 2 clay 448642 s18 c sherd crockett curvilinear incised 1 clay 448642 s19 c sherd Friendship Engraved 1 clay 448642 s20 c sherd Undesig. engraved 1 clay 448642 s21 c sherd Undesig. incised shell 1 clay 448642 s22 c sherd Undesig. plain 5 clay 448643 s01 c sherd Poteau Plain 10 clay 448643 s02 c sherd Poteau Plain 2 clay 448643 s03 c sherd Poteau Plain 1 clay 448643 s04 c sherd Undesig. grog tempered 5 clay 448643 s05 c sherd woodward Plain 10 clay 448643 s06 c sherd woodward Plain 2 clay 448643 s07 c sherd woodward Plain 1 clay 448643 s08 c sherd Effigy 1 clay 448644 s01 c sherd crockett curvilinear incised 1 clay 448644 s02 c sherd crockett curvilinear incised 1 clay 448644 s03 c sherd Unclassified 1 clay 448645 s01 c sherd sanders Plain 2 clay 448645 s02 c sherd crockett curvilinear incised 2 clay 448645 s03 c sherd Leflore Plain 1 clay 448645 s04 c sherd sanders Engraved 1 clay 448645 s05 c sherd Unclassified 1 clay 448645 s06 c sherd williams Plain 1 clay 448645 s07 c sherd Undesig. slipped incised 1 clay 448645 s08 c sherd williams Plain 1 clay 448646 s01 c sherd Undesig. grit tempered 11 448646 s02 c sherd Undesig. slipped grog 3 448646 s03 c sherd Undesig. incised grog 2 448646 s04 c sherd Poteau Plain 1 448646 s05 c sherd Undesig. shell tempered 2 448647 c sherd Red on polished 2 clay N U M B E R 4 9 ? 19 9 Catalog Subnumber Material Raw material number (if any) type Object Style Count (if known) (continued) 448648 c Vessel williams Plain? 2 448649 c sherd Effigy 1 clay 448650 s01 L s Point agee 1 novaculite 448650 s02 L s Point Rockwall 1 chert 448650 s03 L s Point Keota 1 chert 448650 s04 L s Point sequoyah 1 chert 448650 s05 L s Point sequoyah 1 chert 448650 s06 L s Point sequoyah 1 chert 448650 s07 L s Point haskell 1 chert 448650 s08 L s Point Morris 1 chert 448650 s09 L s Point Morris 1 Unclassified 448650 s10 L s Point agee 1 chert 448650 s11 L s Point agee 1 chert 448650 s12 L s Point agee 1 novaculite 448650 s13 L s Point agee 1 novaculite 448650 s14 L s Point hayes short 1 novaculite 448650 s15 L s Point Keota 1 chert 448650 s16 L s Point Unclassified side notch 1 chert 448650 s17 L s Point Unclassified 1 chert 448650 s18 L s Point Young 1 chert 448651 s01 L s Point Edwards 1 chert 448651 s02 L s Point Massard 1 chert 448651 s03 L s Point washita Peno 1 chert 448651 s04 L s Point Keota 1 chert 448651 s05 L s Point collier 1 chert 448651 s06 L s Point ashley 1 novaculite 448651 s07 L s Point Massard 1 chert 448651 s08 L s Point alba 1 novaculite 448651 s09 L s Point Bonham tahlequah? 1 chert 448651 s10 L s Point Massard 1 chert 448651 s11 L s Point washita chaffee 1 chert 448651 s12 L s Point haskell 1 chert 448651 s13 L s Point Reed 1 chert 448651 s14 L s Point washita garvin 1 chert 448651 s15 L s Point scallorn 1 chert 448651 s16 L s Point agee 1 chert 448651 s17 L s Point Massard 1 chert 448651 s18 L s Point haskell 1 chert 448651 s19 L s Point hayes 1 chert 448651 s20 L s Point coryell 1 chert 448651 s21 L s Point Massard 1 chert 448651 s22 L s Point nodena 1 chert 448651 s23 L s Point Keota 1 chert 448651 s24 L s Point alba 1 chert 448651 s25 L s Point alba 1 Jasper 448651 s26 L s Point Fresno 1 chert 448651 s27 L s Point Reed 1 chert 448651 s28 L s Point agee 1 novaculite 448651 s29 L s Point collier 1 chert 448651 s30 L s Point scallorn 1 novaculite 448651 s31 L s Point Unclassified 1 chert 448651 s32 L s Point alba 1 novaculite 448651 s33 L s Point Bonham 1 chert 448651 s34 L s Point washita garvin 1 chert 448651 s35 L s Point hayes short 1 chert 448651 s36 L s Point agee 1 chert 448651 s37 L s Point hayes 1 chert 448651 s38 L s Point homan 1 novaculite 448651 s39 L s Point alba 1 novaculite 448651 s40 L s Point Reed 1 chert 448651 s41 L s Point Unclassified 1 chert 2 0 0 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y Catalog Subnumber Material Raw material number (if any) type Object Style Count (if known) 448651 s42 L s Point Rockwall 1 chert 448651 s43 L s Point alba 1 chert 448651 s44 L s Point scallorn 1 chert 448651 s45 L s Point Keota 1 chert 448651 s46 L s Point Unclassified 1 chert 448651 s47 L s Point coryell 1 chert 448651 s48 L s Point Massard 1 chert 448651 s49 L s Point agee 1 novaculite 448651 s50 L s Point Unclassified corner notch 1 chert 448651 s51 L s Point Unclassified 1 chert 448651 s52 L s Point washita 1 novaculite 448651 s53 L s Point Keota 1 novaculite 448651 s54 L s Point Unclassified 1 chert 448652 s01 L L Point gary 1 chert 448652 s02 L L Point Lange 1 chert 448652 s03 L L Point gary 1 chert 448653 L L Point Eccentric 1 chert 448654 s01 L Lg Biface Unclassified 1 chert 448654 s02 L Lg Biface Preform 1 chert 448654 s03 L Lg Biface Preform 1 novaculite 448654 s04 L Lg Biface Preform 1 chert 448654 s05 L Lg Biface Preform 1 chert 448654 s06 L s Point nodena 1 chert 448655 L Drill nonhaft 1 chert 448656 s01 L celt shouldered spade 1 siltstone 448656 s02 L celt shouldered spade 1 siltstone 448657 L Pipe t-shaped, chandler 1 sandstone 448658 L Pipe t-shaped, contracting 1 siltstone/pipest 448659 L Pipe t-shaped, cylindrical bowl 1 siltstone 448660 s01 L Pipe t-shaped, cylindrical bowl 1 siltstone 448660 s02 L Pipe t-shaped, contracting 1 siltstone 448660 s03 L Pipe t-shaped, cylindrical bowl 1 siltstone 448661 s01 c Pipe Unclassified 1 clay 448661 s02 c Pipe t-shaped, unclassified 1 clay 448661 s03 c Pipe t-shaped, unclassified 1 clay 448661 s04 c Pipe t-shaped, unclassified 1 clay 448661 s05 L Pipe t-shaped, unclassified 2 siltstone 448661 s06 L sherd Poteau Plain 1 clay 448662 s01 L Pipe t-shaped, unclassified 1 Limestone 448662 s02 L Pipe t-shaped, unclassified 1 steatite 448662 s03 L cone object Unclassified 2 Quartz 448663 L Pipe Effigy 1 Limestone 448664 L Pipe Effigy 1 Limestone 448665 L Pipe Effigy 1 Limestone 448666 L Pipe Effigy 1 Limestone 448667 L Pipe Effigy 1 sphalerite 448668 L Pipe Effigy 1 Limestone 448669 s01 L Pipe Effigy 1 Limestone 448669 s02 L sediment none 1 clay-silt 448670 L celt Elongate 1 sedimentary 448671 L celt Elongate 1 sedimentary 448672 L celt Elongate 1 greenstone 448673 L celt spatulate 1 greenstone 448674 L celt Perforated spatulate 1 Marble 448675 L celt spatulate 1 coal 448676 s01 L celt Elongate, small 1 Jasper 448676 s02 L celt Elongate 1 igneous 448677 s01 L atlatl weight Boatstone 1 greenstone 448677 s02 L atlatl weight Boatstone 1 greenstone 448677 s03 L atlatl weight Boatstone 1 greenstone 448677 s04 L atlatl weight Boatstone 1 greenstone N U M B E R 4 9 ? 20 1 Catalog Subnumber Material Raw material number (if any) type Object Style Count (if known) (continued) 448678 L Discoidal chunky 1 Quartzite 448679 L Discoidal chunky 1 sedimentary 448680 s01 L Discoidal chunky 1 Marble 448680 s02 L Discoidal chunky 1 sedimentary 448681 L Mano Pitted 1 sandstone 448682 L hammerstone 3 Quartz 448683 s01 L Pebble worked 1 Quartz 448683 s02 L Pebble worked 1 Unknown 448683 s03 L Pebble worked 1 Unknown 448683 s04 L Pendant Perforated end 1 coal 448683 s05 L Pebble worked 1 stone 448684 L Pendant Perforated end 1 greenstone 448685 L Pendant Perforated end 1 sandstone 448686 s01 L Pendant Perforated end 1 Phosphate 448686 s02 L Pendant Perforated end 1 Phosphate 448686 s03 s Bead convexo-cylin 1 shell 448686 s04 L Bead convexo-cylin blank 1 Unclassified 448686 s05 L Bead convexo-cylin blank 1 Unclassified 448686 s06 L Bead convexo-cylin blank 1 Unclassified 448686 s07 L Bead Elliptical 1 Phosphate 448686 s08 L Bead cylindrical 1 Phosphate 448686 s09 L Bead Elliptical 1 Phosphate 448686 s10 L Bead convexo-cylin 1 Phosphate 448686 s11 L Bead spherical 1 Phosphate 448686 s12 L Bead Elliptical 1 Phosphate 448686 s13 L Bead spherical 1 Phosphate 448686 s14 L Bead spherical 1 Phosphate 448686 s15 L Bead spherical 1 Phosphate 448686 s16 L Bead spherical 1 Phosphate 448686 s17 L Bead spherical 1 Phosphate 448686 s18 L Bead convexo-cylin 1 Phosphate 448687 L Bead Disc 1 sedimentary 448688 s01 L Flake secondary 1 chert 448688 s02 L Flake Blank 1 chert 448688 s03 L Flake secondary 1 chert 448688 s04 L Flake Blank 1 chert 448689 s01 L Mineral lump 1 Micac. sandstone 448689 s02 L Mineral lump 1 Micac. sandstone 448689 s03 L Mineral lump 1 Micac. sandstone 448689 s04 L Mineral lump 1 Micac. sandstone 448689 s05 L Mineral lump 1 Micac. sandstone 448689 s06 L Mineral lump 1 Micac. sandstone 448689 s07 L Mineral lump 1 Micac. sandstone 448689 s08 L Mineral lump 1 Micac. sandstone 448689 s09 L Mineral lump 1 Micac. sandstone 448690 s01 L Lg Biface Unclassified 1 Unknown 448690 s02 L celt Unclassified 1 coal 448690 s03 L celt Unclassified 1 coal 448690 s04 L celt Unclassified 1 coal 448690 s05 L celt spatulate 3 coal 448690 s06 L celt spatulate 2 coal 448691 s02 L Earspool Perforated 1 sandstone 448692 L Earspool Perforated 1 sandstone 448693 L Earspool Perforated 1 sandstone 448694 L Earspool Perforated 1 Limestone 448695 s01 L Earspool Divided flange 1 Limestone 448695 s02 L Earspool Divided flange 1 sandstone 448696 c Earspool Perforated 1 clay 448697 L Earspool Perforated 1 Limestone 448698 s01 L Earspool Perforated 1 silt/sandstone 448698 s02 L Earspool Perforated 1 silt/sandstone 2 0 2 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y Catalog Subnumber Material Raw material number (if any) type Object Style Count (if known) 448698 s03 L Earspool Perforated 1 silt/sandstone 448698 s04 L Earspool Perforated 1 Limestone 448698 s05 L Earspool Perforated 1 siltstone 448698 s06 L Earspool Perforated 1 sandstone 448698 s07 L Earspool Perforated 1 siltstone 448698 s08 L, M Earspool Perforated 1 Limestone 448698 s09 L Earspool Perforated 1 Limestone 448698 s10 L Earspool Perforated 1 Limestone 448698 s11 L Earspool Perforated 1 Limestone 448698 s12 L Earspool Perforated 1 Limestone 448698 s13 L Earspool Perforated 1 Limestone 448698 s14 L Earspool Perforated 1 Limestone 448698 s15 L Earspool Perforated 1 Limestone 448698 s16 L Earspool Perforated 1 Limestone 448698 s17 L Earspool Perforated 1 Limestone 448698 s18 L Earspool Perforated 1 Limestone 448698 s19 L Earspool Perforated 1 Limestone 448699 s01 L Earspool Perforated 1 Limestone 448699 s02 L Earspool nesting 1 Limestone 448699 s03 L Earspool Divided flange 11 Limestone 448699 s04 L, M Earspool Unperforated 1 Limestone 448699 s05 L, M Earspool Unperforated 56 Limestone 448699 s06 L Earspool Perforated 2 sandstone 448699 s07 L Earspool Unclassified 8 siltstone 448699 s08 M Earspool Unclassified 5 copper 448699 s09 L Earspool Divided flange 1 Limestone 448699 s10 L Earspool Divided flange 1 Limestone 448699 s11 L Earspool Divided flange 1 Limestone 448699 s12 L Earspool Unclassified 50 Limestone 448699 s13 L Earspool Divided flange 2 siltstone 448699 s14 c sherd woodward 1 ceramic 448700 L, M Earspool Unclassified 69 copper 448702 L galena worked 1 galena 448703 L galena worked 1 galena 448704 L galena worked cube 1 galena 448705 L galena worked 1 galena 448705 s02 L galena nodule 8 galena 448706 s01.1 L Figure Effigy 1 galena 448706 s01.2 t Basket impression Unclassified 1 Reed or cane 448707 B Pin spatulate 1 Bone 448708 s01 B Pin tapering 15 Bone 448708 s02 B awl troughed 6 Bone 448709 B scraper Unclassified 1 Long bone 448710 B scraper Unclassified 1 Metapodial 448711 B tube hollow 1 Long bone 448712 M Pin Fraudulent 15 copper 448713 M Bead Fraudulent-disc 11 copper 448714 s01 M Bead tubular small 5 copper 448714 s02 M, t Bead tubular medium 4 copper 448714 s03 M, t Bead tubular medium large 12 copper 448714 s04 M Bead tubular large 4 copper 448714 s05 M, t Bead tubular large 2 copper 448714 s06 M, t Bead tubular medium large 2 copper 448714 s06.2 t cordage 2-ply s-twist 1 Fiber 448714 s07.1 M, t, s Bead tubular medium 3 copper 448714 s07.2 t cordage 2-strand 3?ply Z-twist 1 Fiber 448714 s08 t cordage 3-ply s-twist 1 Fiber 448714 s09 t cordage 2-ply Z-twist 1 Fiber 448714 s10 M Pin tapering 6 copper 448714 s11 M Pin curved 11 copper 448715 s01 M sheet Embossed 5 copper N U M B E R 4 9 ? 20 3 Catalog Subnumber Material Raw material number (if any) type Object Style Count (if known) (continued) 448715 s02 M sheet Embossed 12 copper 448715 s03 M sheets Embossed 16 copper 448715 s04 M sheets Embossed 14 copper 448715 s05 M, w sheets Unclassified 1 copper 448715 s06 M sheet Embossed 7 copper 448715 s07 M, w sheet Embossed 1 copper 448715 s08 M sheet nonembossed 5 copper 448715 s09 M, w sheet Embossed 1 copper 448715 s10 M sheet Embossed 3 copper 448715 s11 M, w sheet Unclassified 10 copper 448716 s01 M sheet nonembossed 8 copper 448716 s02 M sheet Plume? 2 copper 448717 M sheet 1 copper 448718 M sheet Embossed 77 copper 448719 s01 s Bead Disc 119 shell 448719 s02 s Bead Elliptical 3 shell 448720 s Bead Disc 76 shell 448721 s01 s Bead Disc 1,400 shell 448721 s02 s Bead Disc-thin 65 shell 448721 s03 s Bead Disc-thin-sculptured 1 shell 448721 s04 L, o Bead crinoid and seed 2 Fossil 448722 s01 s Bead Variety pearl 321 Pearl 448722 s02 s Bead Baroque 1 Pearl 448723 s01 s Bead convexo-bowed 85 shell 448723 s02 s Bead cylindrical 3 shell 448723 s03 s Bead convexo-cylin 5 shell 448723 s04 s Bead convexo-cylin 1 shell 448724 s01 s Bead convexo-cylin-columella 17 shell 448724 s02 s Bead convexo-bowed-large 2 shell 448724 s03 s Bead convexo-bowed 1 shell 448724 s04 s Bead columella 3 shell 448725 s01 s Bead spherical-columella 80 shell 448725 s02 s Bead Baroque 1 Pearl 448725 s03 s Bead Elliptical 2 shell 448726 s, M Bead spherical 3 shell 448727 s Pendant grooved drop 1 shell 448728 s01 s Bead convexo-cylin 2 shell 448728 s02 s Bead convexo-cylin 50 shell 448728 s03 s Bead Disc 7 shell 448728 s04 s Bead Baroque 1 Pearl 448728 s05 s Bead Elliptical 7 shell 448728 s06 s Bead spherical 17 shell 448729 s Bead convexo-cylin 154 shell 448730 s01 s Bead Bulging-columella 1 shell 448730 s02 s Bead Bulging 9 shell 448731 s Bead concavo 9 shell 448732 s Bead tooth 6 shell 448733 s01 s Bead gastropod 44 Univalve 448733 s02 s Bead Marginella 3 Univalve 448733 s04 s Bead gastropod 18 Univalve 448734 s Bead Olivella 2,790 shell 448735 s Bead Flat circular 2 shell 448736 s Pendant gorget-like 1 shell 448737 s01.1 M, s, t sheet Embossed 3 copper 448737 s01.2 s Bead convexo-cylin-columella 2 shell 448737 s02 M, s, t worked shell Unclassified 1 shell 448737 s03.1 M, s, t Bead convexo-cylin-columella 1 shell 448737 s03.2 t cordage 2-ply s-twist 1 Fiber 448737 s04 M, s Bead convexo-cylin-columella 1 shell 448737 s05 M, s Bead spherical columella 2 shell 448737 s06 M, s Bead convexo-cylin 1 shell 2 0 4 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y Catalog Subnumber Material Raw material number (if any) type Object Style Count (if known) 448737 s07 M, s Bead convexo-cylin-columella 1 shell 448737 s08 M, s, t Bead convexo-cylin 1 shell 448737 s09 M s Bead convexo-cylin 3 shell 448737 s10 M s Bead convexo-cylin 1 shell 448737 s11 M s Bead convexo-cylin-columella 1 shell 448737 s12 M, s, t Bead spherical columella 1 shell 448737 s13 M, t sheet Unclassified 14 copper 448738 s Pendant Drop 1 shell 448739 s Pendant Disc 1 shell 448740 s gorget type 1 1 shell 448741 s gorget type 1, craig a 1 shell 448743 s gorget Mask 1 shell 448744 s gorget type 1, Unclassified 1 shell 448745 s worked shell Unclassified 1 shell 448746 s gorget craig B 1 shell 448747 s gorget type 1, craig a 1 shell 448748 s gorget type 1 1 shell 448749 s gorget type 1, craig a 2 shell 448750 s gorget type 1, craig a 2 shell 448752 s gorget type 1, craig a 1 shell 448753 s gorget type 1, craig 1 shell 448755 s gorget type 1, craig 2 shell 448757 s gorget type 2, Unclassified 1 shell 448758 s gorget style 1, craig a 1 shell 448759 s gorget type ?, craig 1 shell 448760 s gorget type 1, craig B 5 shell 448761 s01.1 s gorget type 1, craig 2 shell 448761 s01.2 s cUP Braden a 2 shell 448761 s02.1 s gorget type 3, craig a 2 shell 448761 s02.2 s cup craig a 2 shell 448762 s, t, M gorget type 3, craig 1 shell 448763 s gorget type 3, craig 1 shell 448764 s worked shell Plain 2 shell 448765 s gorget type 1, craig 12 shell 448766 s Pendant irregular 1 shell 448768 s01 s Bead olivella 1 shell 448768 s02 s Bead convexo-cylin-sculptured 1 Univalve 448769 s Pendant irregular 1 shell 448770 s worked shell hook 2 shell 448771 s01 s Disk Plain 1 shell 448771 s02 s Disk Plain 1 shell 448772 s Figurine craig c 1 shell 448773 s Bead Mask 1 shell 448774 s cup Unengraved 1 Univalve 448775 s worked shell worked 1 shell 448776 s worked shell 1 Bivalve 448777 s Pendant Drop-columella 4 Univalve 448778 s01.1 s, t Pendant columella 1 Univalve 448778 s01.2 t, s Basketry twill 2 wicker, cane 448779 s01 s Pendant columella 4 Univalve 448779 s02 s Pendant columella 4 Univalve 448780 s01 s, L Pendant columella-sculptured 1 Univalve 448780 s02 s Bead spherical pearl 1 Pearl 448781 s01 s Pendant columella 5 shell 448781 s02 s Pendant columella 5 shell 448781 s03 s Pendant columella 13 shell 448781 s04 s Pendant columella 5 shell 448781 s05 s Pendant columella 6 shell 448781 s06 s Pendant columella 1 shell 448781 s07 s Pendant columella 5 shell 448781 s08 s Pendant columella 10 shell N U M B E R 4 9 ? 20 5 Catalog Subnumber Material Raw material number (if any) type Object Style Count (if known) (continued) 448781 s09 s Pendant columella 6 shell 448781 s10 s Pendant columella 1 shell 448782 s00 s cup Unengraved 1 Univalve 448782 s01 s cup Unengraved 1 shell 448782 s02 s cup Unengraved 8 shell 448782 s03 s cup Unengraved 14 shell 448782 s04 s cup Unengraved 5 shell 448782 s05 s cup Unengraved 26 shell 448782 s06 s cup Unengraved 1 shell 448782 s07 s cup Unengraved 1 shell 448782 s08 s cup Unengraved 1 shell 448782 s09 s cup Unengraved 1 shell 448782 s10 s cup Unengraved 1 shell 448782 s11 s cup Unengraved 1 shell 448782 s12 s cup Unengraved 1 shell 448782 s13 s cup Unengraved 1 shell 448782 s14 s cup Unengraved 1 shell 448782 s15 s cup Unengraved 1 shell 448782 s16 s cup Unengraved 1 shell 448782 s17 s cup Unclassified 1 shell 448782 s18 s cup Unengraved 1 shell 448783 s cup Braden B 1 Univalve 448784 s cup craig c 1 Univalve 448785 s cup Braden B 2 Univalve 448786 s cup Braden c 1 Univalve 448787 s cup craig B 1 Univalve 448788 s cup Braden B 1 Univalve 448789 s cup Braden B 1 Univalve 448790 s cup Braden B 1 Univalve 448791 s cup Braden a 3 Univalve 448792 s cup Braden c 1 Univalve 448793 s cup Braden c 1 Univalve 448794 s cup craig c 1 Univalve 448795 s cup craig c 1 Univalve 448796 s cup craig c 1 Univalve 448797 s cup craig c 1 Univalve 448798 s cup craig a 1 Univalve 448799 s cup craig c 1 Univalve 448800 s cup craig c 1 Univalve 448801 s cup craig B 1 Univalve 448802 s cup craig c 1 Univalve 448803 s cup Braden B 2 Univalve 448804 s cup craig c 1 Univalve 448805 s cup Braden B 1 Univalve 448806 s cup craig B 6 Univalve 448807 s cup Braden B 8 Univalve 448808 s cup craig B 1 Univalve 448809 s cup Braden c 1 Univalve 448810 s cup Braden B 1 Univalve 448811 s cup Braden B 1 Univalve 448812 s cup craig c 2 Univalve 448813 s cup craig c 1 Univalve 448814 s cup craig B 2 Univalve 448815 s cup craig a 2 Univalve 448816 s cup Braden c 5 Univalve 448817 s cup craig B 3 Univalve 448818 s cup craig B 3 Univalve 448819 s cup Braden a 2 Univalve 448820 s cup Braden B 2 Univalve 448821 s cup craig B 5 shell 448822 s cup craig c 1 Univalve 2 0 6 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y Catalog Subnumber Material Raw material number (if any) type Object Style Count (if known) 448823 s cup Unclassified 1 Univalve 448824 s cup Unclassified 1 Univalve 448825 s cup craig a 3 Univalve 448826 s cup craig B 3 Univalve 448827 s cup Braden a 3 Univalve 448828 s cup Braden c 3 Univalve 448829 s cup Braden a 2 Univalve 448830 s cup Braden c 2 Univalve 448831 s cup craig B 5 Univalve 448832 s cup craig c 4 Univalve 448833 s cup Braden a 2 Univalve 448834 s cup craig c 3 Univalve 448835 s cup craig c 4 Univalve 448836 s cup craig 4 Univalve 448838 s cup Braden a 2 Univalve 448839 s cup craig c 2 Univalve 448840 s cup craig B 4 Univalve 448843 s cup Braden a 1 Univalve 448844 s cup Braden a 1 Univalve 448845 s cup craig B 1 Univalve 448847 s cup Braden a 2 Univalve 448848 s cup craig B 6 Univalve 448849 s cup Braden a 4 Univalve 448850 s01 s cup Braden B 2 Univalve 448850 s02 s cup craig 1 Univalve 448851 s cup craig a 5 Univalve 448852 s cup craig B 4 Univalve 448853 s01 s cup craig B 3 Univalve 448853 s02 s cup craig B 2 Univalve 448853 s03 s cup craig B 2 Univalve 448853 s04 s cup craig B 3 Univalve 448853 s05 s cup craig c 2 Univalve 448854 s cup Braden c 3 Univalve 448855 s cup Braden B 1 Univalve 448856 s cup Braden B 2 Univalve 448857 s cup Braden B 1 Univalve 448858 s01 s cup craig B 3 Univalve 448858 s02 s cup craig c 1 Univalve 448858 s03 s cup craig c 1 Univalve 448858 s04 s cup Braden a 2 Univalve 448858 s05 s cup craig B 1 Univalve 448858 s06 s cup craig a 2 Univalve 448858 s07 s cup craig a 2 Univalve 448858 s08 s cup craig B 1 Univalve 448858 s09 s cup craig B 1 Univalve 448858 s10 s cup craig B 3 Univalve 448858 s11 s cup Various 16 Univalve 448858 s12 s cup craig a 3 Univalve 448859 s cup craig a 1 Univalve 448860 s cup Braden a 4 Univalve 448861 s01 s cup craig B 4 Univalve 448861 s02 s cup Braden c 1 Univalve 448861 s03 s cup craig B 2 Univalve 448861 s04 s cup craig B 1 Univalve 448861 s05 s cup Braden a 1 Univalve 448861 s06 s cup craig a 2 Univalve 448861 s07 s gorget-mask craig B 2 Univalve 448862 s cup craig a 6 Univalve 448863 s cup Braden a 1 Univalve 448864 s cup Braden c 1 Univalve 448865 s cup Braden a 6 Univalve N U M B E R 4 9 ? 20 7 Catalog Subnumber Material Raw material number (if any) type Object Style Count (if known) (continued) 448866 s01 s cup Braden 6 Univalve 448866 s02 s cup Braden a 3 Univalve 448866 s03 s cup craig B 2 Univalve 448866 s04 s cup Braden a 2 Univalve 448866 s05 s cup craig B 4 Univalve 448866 s06 s cup Braden a 3 Univalve 448866 s07 s cup craig B 3 Univalve 448866 s08 s cup Braden a 1 Univalve 448866 s09 s cup craig B 2 Univalve 448866 s10 s cup Braden a 2 Univalve 448866 s11 s cup craig B 7 Univalve 448866 s12 s cup Braden a 4 Univalve 448867 s01 s cup craig B 2 Univalve 448867 s02 s cup Unclassified 4 Univalve 448867 s03 s cup Unclassified 5 Univalve 448868 s cup Braden B 5 Univalve 448869 s01 s cup craig various 16 Univalve 448869 s02 s cup craig a 2 Univalve 448869 s03 s cup craig B 2 Univalve 448870 s cup Braden B 2 Univalve 448871 s01 s cup craig a 5 Univalve 448871 s02 s cup Braden B 2 Univalve 448871 s03 s cup Braden a 2 Univalve 448872 s cup craig B 4 Univalve 448873 s cup craig B 1 Univalve 448874 s01 s cup craig B 2 Univalve 448874 s02 s cup Unclassified 1 Univalve 448874 s03 s cup craig a 3 Univalve 448875 s01 s cup craig a 2 Univalve 448875 s02 s cup craig a 2 Univalve 448875 s03 s cup Braden B 3 Univalve 448875 s04 s cup craig c 2 Univalve 448876 s01 s cup Braden a 2 Univalve 448876 s02 s cup Braden a 16 Univalve 448876 s03 s cup craig a 1 Univalve 448876 s04 s cup Braden a 10 Univalve 448877 s01 s cup craig c 3 Univalve 448877 s02 s cup Unclassified 2 Univalve 448877 s03 s cup Braden 3 Univalve 448877 s04 s cup Braden a 2 Univalve 448877 s05 s cup craig B 2 Univalve 448877 s06 s cup Braden a 11 Univalve 448877 s07 s cup craig B 2 Univalve 448877 s08 s cup craig B 2 Univalve 448878 s01 s cup craig a 6 Univalve 448878 s02 s cup craig a 4 Univalve 448878 s03 s cup Braden B 3 Univalve 448879 s01 s cup craig B 2 Univalve 448879 s02 s cup craig B 6 Univalve 448880 s00 s cup craig c 2 Univalve 448880 s01 s cup Braden a 1 Univalve 448880 s02 s cup Braden a 1 Univalve 448880 s03 s cup Braden a 2 Univalve 448880 s04 s cup craig B 2 Univalve 448880 s05 s cup craig a 7 Univalve 448880 s06 s cup Braden a 1 Univalve 448880 s07 s cup craig a 2 Univalve 448880 s08 s cup Braden a 9 Univalve 448880 s09 s cup Braden a 1 Univalve 448880 s10 s cup Braden a 1 Univalve 448880 s11 s cup Unclassified 2 Univalve 2 0 8 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y Catalog Subnumber Material Raw material number (if any) type Object Style Count (if known) 448880 s12 s cup Braden c 1 Univalve 448880 s13 s cup Braden B 3 Univalve 448880 s14 s cup craig B 2 Univalve 448880 s15 s cup Braden B 2 Univalve 448880 s16 s cup craig c 1 Univalve 448880 s17 s cup craig c 1 Univalve 448880 s18 s cup craig c 1 Univalve 448880 s19 s cup craig c 3 Univalve 448880 s20 s cup craig a 4 Univalve 448880 s21 s cup craig a 2 Univalve 448880 s22 s cup craig B 4 Univalve 448880 s23 s cup craig B 6 Univalve 448880 s24 s cup craig B 2 Univalve 448880 s25 s cup craig a 3 Univalve 448880 s26 s cup craig a 6 Univalve 448880 s27 s cup Braden c 1 Univalve 448880 s28 s cup Braden B 1 Univalve 448880 s29 s cup Braden a and B 3 Univalve 448880 s30 s cup craig a 4 Univalve 448880 s31 s cup Braden a 1 Univalve 448880 s32 s cup Braden B 2 Univalve 448880 s33 s cup Braden B 1 Univalve 448880 s34 s cup Unclassified 39 Univalve 448880 s35 s cup Unclassified 11 Univalve 448880 s36 s cup Unclassified 14 Univalve 448880 s37 s cup craig B 2 Univalve 448880 s38 s cup Unclassified 9 Univalve 448880 s39 s cup craig a 1 Univalve 448880 s40 s cup Unclassified 4 Univalve 448880 s41 s cup craig B 4 Univalve 448880 s42 s cup Unclassified 10 Univalve 448880 s43 s cup Unclassified 7 Univalve 448880 s44 s worked shell scales 3 scales 448881 s cup Unclassified 3 Univalve 448882 s cup Braden a 1 Univalve 448883 s01 s cup Unengraved 4 Univalve 448883 s02 s cup Unengraved 2 Univalve 448883 s03 s cup Unengraved 15 Univalve 448883 s04 s cup Unengraved 20 Univalve 448883 s05 s cup Unengraved 10 Univalve 448883 s06 s cup Unengraved 12 Univalve 448883 s07 s cup Unengraved 1 Univalve 448884 s worked shell Unengraved 1 Univalve 448885 s01 s worked shell Unengraved 5 448885 s02 s worked shell Unengraved 1 448885 s03 s Pendant Unengraved 2 Univalve 448885 s04 s worked shell Unclassified 6 448885 s05 s worked shell Unmodified 1 Bivalve 448886 s Pendant grooved drop 1 448887 s01 s worked shell columella 7 shell 448887 s02 s Pendant columella 3 shell 448887 s03 s worked shell Unclassified 1 Univalve 448888 s worked shell Unengraved 1 Univalve 448889 s01 s worked shell Unmodified 3 Univalve 448889 s02 s Bead Marginella 1 Univalve 448889 s03 s shell Unmodified 1 Univalve 448889 s04 s worked shell Unmodified 1 Univalve 448889 s05 s worked shell Unmodified 2 Univalve 448889 s06 s shell Unmodified 2 Univalve 448890 w Figure Effigy 1 Red cedar 448891 w Figure Effigy 1 Red cedar N U M B E R 4 9 ? 20 9 Catalog Subnumber Material Raw material number (if any) type Object Style Count (if known) (continued) 448892 w Figure Effigy 1 cedar 448893 w Vessel Effigy 1 cedar 448894 w, s Vessel Effigy 1 cedar 448895 s01 w Earspool Effigy 1 cedar 448895 s02 w Earspool Effigy 1 cedar 448897 w Blade ceremonial 1 cedar 448898 B ornament Perforated plate 1 turtle 448899 s01 w Earspool cone 1 cedar 448899 s02 w Earspool cone 1 cedar 448899 s03 w Earspool cone 1 cedar 448899 s04 w Earspool cone 1 cedar 448899 s05 w Earspool cone 1 cedar 448900 s01 w, s Earspool composite 1 cedar 448900 s02 w Earspool central boss 1 448901 w, M Earspool worked 1 cedar 448902 w Earspool Unperforated 1 448903 w Earspool Unperforated 1 448904 w Earspool Unperforated 1 448905 s01 w, s Earspool Unperforated 1 448905 s02 w Earspool Unperforated 1 448905 s03 w, s Earspool composite 1 cedar 448905 s04 w, s Earspool central boss 1 cedar 448905 s05 w, s Earspool Unperforated 1 cedar 448906 s01 s Disk Plain 4 shell 448906 s02 s Disk Perforated disc 2 448906 s03 s Disk Plain large 2 shell 448907 s Disk Perforated disc 2 448908 s Disk Engraved 1 shell 448909 w Earspool Perforated 6 cedar 448910 w object Effigy 1 448911 s01 w worked frag Unknown 8 448911 s02 w worked frag Unknown 17 448911 s03 o Pod Unmodified 1 448911 s04 w worked frag stick 24 cedar 448911 s05 w worked frag Mace or atlatl 1 448912 t Basketry twill matting 1 cane 448913 s01 t Basketry spaced weft twining 1 Vegetal 448913 s02 t sample slide 1 448914 t Basketry Loose fibers 1 448915 t cloth tapestry 4 hair? 448916 s01 t cloth tapestry 3 hair? 448916 s02 t cloth spaced weft twining, 4 hair wrapped warp 448916 s03 t cloth spaced weft twining 17 hair 448916 s04 t hide Leather 1 448917 s01 t cordage Braid, wrapped warp 10 Vegetal barrel-shaped wrapped warp 448917 s02 s Bead convexo-cylin 7 shell 448918 t cordage Loose fibers 1 448919 t cordage, fur 2-ply s-twist 1 Vegetal 448920 o hide Leather 1 hide 448921 L Mica Flake 1 Mica 448922 t Basketry twill 2 cane 448923 o clay worked 171 glauconite 448924 o Pigment worked 1 ocher 448925 o Pigment Raw 2 ocher 448926 s01 o clay chert unknown 1 clay 448926 s02 o clay worked 1 clay 448927 s01.1 o Pigment Unmodified 1 ocher 448927 s01.2 t Basket twill 1 Reed/cane 448927 s02 o Pigment 1 ocher 2 1 0 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y Catalog Subnumber Material Raw material number (if any) type Object Style Count (if known) 448927 s03 o Pigment worked 1 ocher 448927 s04 o Pigment 1 ocher 448927 s05 o Pigment Raw 1 ocher 448928 c object Poverty point object 1 clay 448929 L L Point gary 1 chert 448930 c sherd Undesig. 1 clay 448931 c Pendant grooved 1 clay 448932 s01 L tube 1 stone 448932 s02 L Bead irregular disc 1 Unknown 448933 o coprolite 3 organic 448934 s01.1 t Basketry twill 1 cane 448934 s01.2 t cloth Braid 1 hair 448934 s01.3 s Bead Disc 500 shell 448934 s02 s Bead Disc 1,650 shell 448934 s03 s Bead spherical 3 shell 448934 s04 s cup Unengraved 4 shell 448934 s05 t Basketry Unclassified 1 cane 448934 s06 M sheet Unclassified 60 copper 448934 s07 t Basketry Unmodified 12 cane 448934 s08 t Leather tanned 1 hide 448935? 448937 b 448939 s01.1 B, t Bone object Unclassified 4 Beaver 448939 s01.2 t cordage 2-ply Z-twist 1 Vegetal 515780 s01 c sherd woodward Plain 3 clay 515780 s02 c sherd williams Plain 1 clay 515780 s03 c sherd Poteau Plain 1 clay 515781 L s Point Morris 1 chert 515782 s01 L Flake Bipolar 1 chert 515782 s02 L scraper End 1 chert 515782 s03 L scraper End 1 chert 515782 s04 L scraper End 1 chert 515783 L s Point washita 1 chert 515784 s01 s Bead spherical 9 shell 515784 s02 s Bead convexo-cylin 99 shell 515785 s01 s Bead olivella 7 shell 515785 s02 s Bead Disc 142 shell 515786 s cup craig 1 shell 515787 s cup Unengraved 37 542527 M Pin Fraudulent 1 copper 542528 M Pin Fraudulent 1 copper 542529 M Pin Fraudulent 1 copper 542530 M Pin Fraudulent 1 copper 542531 M Pin Fraudulent 1 copper 542532 L Earspool Perforated 1 Limestone 542533 s Bead spherical-columella 21 shell 542534 s Bead convexo-bowed 32 shell 542535 s Bead spherical 14 shell 542536 s Bead Olivella 990 shell 542537 s01 s Bead convexo-cylin 42 shell 542537 s02 s Bead spherical pearl 2 Pearl 542537 s03 s Bead Olivella 2 shell 542538 s cup Unengraved 1 Univalve a no number available. b human remains?see chapter 9. appendix B: concordance of References Many of the artifacts in the spiro collection have been illustrated and dis- cussed in other publications. table B.1 provides a list of artifact catalog numbers along with notes and references to the other major published sources. 2 1 2 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y TABLE B.1. Reference summary of the national Museum of natural history?s spiro collection by catalog number. (naa = national anthropological archives.) Catalog Subnumber number (if any) Object References and notes 377378 Mace crowned maces discussed by Brown (1976:166?172), designated as type 1 (figs. 30a, 31a,b) hamilton?s (1952) pl. 41 is the same artifact 378240 Fragment Rogers et al. (2002) 378244 Pendant Fundaburk and Foreman (1957) depict whelk picks, but these are not made solely of the columella 378261 Bead Brown (1976:377) gives types for similar beads 378264 Pendant see Burnett (1945: pls. LXXXi, LXXXii, LXXXV) for similar pendants and use on headband; see Brown (1976) 378273 End scraper Barker et al. (2002) 386187 cloth Rogers et al. (2002) 386188 cloth Rogers et al. (2002) 411901 worked shell Fundaburk and Foreman (1957: pl. 25) 411904 worked shell illustrated (Duffield, 1964: pl. Vii2) 411905 worked shell Fundaburk and Foreman (1957: pl. 27?) 423145 Vessel hamilton (1952: pl. 32a) 423147 Vessel hamilton (1952: pls. 32a, 33c) 423148 Vessel hamilton (1952: pls. 32a, 33B) 423149 Vessel hamilton (1952: pl. 31), partially restored 423150 Vessel hamilton (1952: pl. 31B); Fundaburk and Foreman (1957: pl. 89) 423152 Pipe hamilton (1952: pl. 22B), restored 423160 Pipe hamilton (1952: pl. 17a) 423161 Earspool hamilton (1952: pl. 80, upper left?) 423162 Earspool hamilton (1952, pl. 81) 423163 Earspool hamilton (1952: pl. 81) 423195 Mace hamilton (1952: pl. 42); griffin (1952:93?94) notes resemblance to Duck River focus, tennessee 423196 Mace hamilton (1952: pl. 44a) 423197 Mace hamilton (1952: pl. 44a); Fundaburk and Foreman (1957: pl. 92) 423198 Mace hamilton (1952: pl. 41); Fundaburk and Foreman (1957: pl. 92) 423227 Pendant compare with hopewell bird in Moorehead (1897:259, fig. XLii) 423234 cup hamilton (1952: pl. 106); Phillips and Brown (1984: pl. 282) 423235 cup hamilton (1952: pl. 117a); Phillips and Brown (1978: pl. 81), Braden style 423236 cup hamilton (1952: pl. 108a); Phillips and Brown (1984: pl. 232) 423237 cup hamilton (1952: pl. 131B); Phillips and Brown (1978:pl. 116), style Braden c 423238 cup hamilton (1952: pls. 92B, 105); not in Phillips and Brown (1978, 1984) 423239 cup hamilton (1952: pl. 115B); Phillips and Brown (1984: pl. 329) 423240 cup Duffield (1964: pl. XXXiii); Phillips and Brown (1978: pl. 114), Braden c; hamilton (1952: pl. 129B) 423241 cup hamilton (1952: pl. 94a); Phillips and Brown (1984: pl. 318) 423242 cup hamilton (1952: pl. 129a); Phillips and Brown (1978: pl. 113), Braden c 423243 cup hamilton (1952: pl. 94B); Phillips and Brown (1984: pl. 320) 423244 cup hamilton (1952: pl. 128B); Fundaburk and Foreman (1957: pl. 28); Phillips and Brown (1978: pl. 121) 423245 cup hamilton (1952: pl. 118); Fundaburk and Foreman (1957: pl. 25); also see clements (1945: pl. XLiii) for similar depiction; Phillips and Brown (1984: pl. 305) 423246 cup hamilton (1952: pl. 119a); Fundaburk and Foreman (1957: pl. 25); Phillips and Brown (1984: pl. 278a) 423247 cup hamilton (1952: pl. 92a); Fundaburk and Foreman (1957: pls. 24, 27); Phillips and Brown (1984: pl. 315) 423248 cup hamilton (1952: pl. 122a); Phillips and Brown (1984: pl. 214a) 423249 cup Phillips and Brown (1984: pl. 79); hamilton (1952: 115a); Phillips and Brown (1978: pl. 79) 423250 cup hamilton (1952: pl. 136a); Phillips and Brown (1978, pl. 1222) 423251 cup hamilton (1952: pl. 134D); Phillips and Brown (1984: pl. 246) 423252 cup Bird in flight, chevron, and an animal that may be an armadillo; not in Phillips and Brown (1978, 1984); assigned Braden c based on execution (continued) N U M B E R 4 9 ? 21 3 Catalog Subnumber number (if any) Object References and notes 423253 cup nearly identical to Phillips and Brown (1984: pl. 327); not in Phillips and Brown (1978, 1984) but might fit center portion shown in their (1984) pl. 333c 423254 cup hamilton (1952: pl. 136B); Phillips and Brown (1984: pl. 264) 423255 cup hamilton (1952: pl. 108B); Phillips and Brown (1984: pl. 230) 423256 cup hamilton (1952: pl. 95a); Phillips and Brown (1984: pl. 312) 423257 cup Duffield (1964) plate XXii5 could be a piece of this or similar shell; hamilton (1952: pl. 103c); Phillips and Brown (1978: pl. 89b) 423258 cup hamilton (1952: pl. 122B); Phillips and Brown (1984: pl. 213a) 423259 cup hamilton (1952: pl. 120B); Phillips and Brown (1978: pl. 101) 423260 cup hamilton (1952: pl. 114B); Phillips and Brown (1978: pl. 76), Braden style 423261 cup hamilton (1952: pl. 132c); Phillips and Brown (1984: pl. 234.1) 423262 cup Phillips and Brown (1984: pl. 308.1) 423263 cup Phillips and Brown (1984: pl. B-7L) 423264 cup hamilton (1952: pl. 125a); Phillips and Brown (1984: pl. 233) 423265 cup hamilton (1952: pl. 103B); Phillips and Brown (1978: pl. 86a); see also 448870 423266 cup hamilton (1952: pl. 128a); Fundaburk and Foreman (1957: pl. 28); Phillips and Brown (1978: pl. 64) 423267 cup Phillips and Brown (1984: pl. 316a) 423268 cup hamilton (1952: pl. 130c); Phillips and Brown (1984: pl. a-1cb) 423269 cup hamilton (1952: pl. 117B); Phillips and Brown (1978: pl. 108) 423270 cup hamilton (1952: pl. 116a); Phillips and Brown (1984: pl. 240a) 423271 cup Burnett (1945: pls. Li and Lii); Phillips and Brown (1984: pl. 234a); hamilton (1952: pl. 132B) 423272 cup Phillips and Brown (1978: pl. 89a) 423273 cup hamilton (1952: pl. 122c); Phillips and Brown (1984: pl. 295a), craig style 423274 cup Phillips and Brown (1978: pl. 45a) 423275 cup Phillips and Brown (1984: pl. 225c) 423276 cup hamilton (1952: 134c); Phillips and Brown (1984: pl. 247a) 423277 cup hamilton (1952: 138E); Duffield (1964: pl. XXiX4), probably same shell; similar to fragment shown in Phillips and Brown (1984: pl. a-1F) 423278 cup Phillips and Brown (1984: pl. 187c); Duffield (1964: pl. XXXii1); hamilton (1952:75, pl. 124a); 423279 cup hamilton (1952: 116B); Phillips and Brown (1984: pl. 167c) 423280 cup hamilton (1952: 138c); Phillips and Brown (1984: pl. 257a); see also 448852 423281 cup hamilton (1952: 125D); Phillips and Brown (1978: pl. 53a), Braden B style; Phillips and Brown (1978) show 448880F as being from same shell 423282 cup hamilton (1952: pl. 125c); Phillips and Brown (1984: pl. 304c) 423283 cup Phillips and Brown (1984: pl. 287gb); hamilton (1952: pl. 126a) 423284 cup hamilton (1952: pl. 125B); Phillips and Brown (1984: pl. 215a); Duffield (1964: pl. Xii) has pictured a fragment that might fit this one 423285 cup Phillips and Brown (1984: pl. a-2B) 423286 cup Phillips and Brown (1984: pl. 174a) 423287 cup Phillips and Brown (1984: pl. 263c); hamilton (1952: pl. 138a) 423288 conch Phillips and Brown (1984: pl. 178.1B) 423289 cup hamilton (1952: pl. 134B); Phillips and Brown (1984: pl. 237b) 423290 cup Phillips and Brown (1984: pl. 252b); hamilton (1952: pl. 138D) 423291 cup Phillips and Brown (1984: pl. B-80) 423292 gorget hamilton (1952: pl. 91D); Phillips and Brown (1984: pl. 275Ba); Brain and Phillips (1996:446) 423293 gorget Phillips and Brown (1984: pl. 148ab); Brain and Phillips (1996:445) 423295 gorget hamilton (1952: pl. 91B); Phillips and Brown (1984: pl. 273c) 423296 cup hamilton (1952: pl. 139a); Phillips and Brown (1984: pls. 159h, 163D) 423297 cup hamilton (1952: pl. 139B); Phillips and Brown (1984: pl. 159e); see also 423296 423298 cameo hamilton (1952: pl. 85); Fundaburk and Foreman (1957: pl. 158); Phillips and Brown (1984: pl. 152Ba) 423299 cameo hamilton (1952: pl. 85); Fundaburk and Foreman (1958); Phillips and Brown (1984: pl. 152B) 423300 cameo hamilton (1952: pl. 85) 423301 cameo hamilton (1952: pl. 85) 423305 Pendant hamilton (1952: pl. 83B) 423309 Bead type illustration hamilton (1952: pl. 66) 2 1 4 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y Catalog Subnumber number (if any) Object References and notes 423310 Bead hamilton (1952: pl. 66) 423312 Disc hamilton (1952: pl. 66) 423313 Bead hamilton (1952: pl. 66) 423314 Bead hamilton (1952: pl. 66) 423315 Bead hamilton (1952: pl. 66) 423317 Bead hamilton (1952: pl. 66) 423318 Bead hamilton (1952: pl. 66) 423324 Bead hamilton (1952: pl. 66) 423329 Bead hamilton (1952: pl. 66) 423352 Mantle Pictured in hamilton (1952: pl. 146); lower also shown in trowbridge notebook in naa (wedel correspondence file) 423353 Mantle hamilton (1952: pl. 145); also shown in trowbridge notebook in naa (wedel correspondence file) 423354 Mantle hamilton (1952: pl. 146); also shown in trowbridge notebook in naa (wedel correspondence file) 423355 Mantle hamilton (1952: pl. 147); Fundaburk and Foreman (1957: pl. 133); also shown in trowbridge notebook in naa (wedel correspondence file) 423356 Mantle hamilton (1952: pl. 145); also shown in trowbridge notebook in naa (wedel correspondence file); Rogers et al. (2002) 423357 Mantle hamilton (1952: pl. 148, lower); also shown in trowbridge notebook in naa (wedel correspondence file) 423358 Mantle hamilton (1952: pl. 148); also shown in trowbridge notebook in naa (wedel correspondence file) 423359 Bundle hamilton (1952: pl. 149, lower); also shown in trowbridge notebook in naa (wedel correspondence file) 423360 Bundle hamilton (1952: pl. 149); also in trowbridge notebook in naa (wedel correspondence file) 423361 Bundle hamilton (1952: pl. 149); Rogers et al. (2002) 423362 Bundle hamilton (1952: pl. 149); Rogers et al. (2002) 423365 Bundle whitford (1941); Rogers et al. (2002) 423366 Bundle Rogers et al. (2002) 423367 Fragments Rogers et al. (2002) 423368 Fragments whitford (1941); Rogers et al. (2002) 423369 Fragments whitford (1941); Rogers et al. (2002) 423370 Fragments hamilton (1952: pl. 143); Rogers et al. (2002) 423371 Fragments Rogers et al. (2002) 423372 Fragments Rogers et al. (2002) 423373 Fragments Rogers et al. (2002) 423374 Fragments Rogers et al. (2002) 423375 cord Rogers et al. (2002) 423377 cord Rogers et al. (2002) 423378 s02 cloth Rogers et al. (2002) 423378 s06 cup incised pattern too fragmentary to infer style, has curvilinear pattern, perhaps of the ceramic mimicry styles of Braden 423379 cloth hamilton (1952: pl. 152) 423380 cloth Rogers et al. (2002) 423386 Matting whitford (1941) 423388 Matting hamilton (1952: pl. 62) 423389 Basketry hamilton (1952: pl. 62); whitford (1941) 423390 Basketry hamilton (1952: pl. 62) 448663 Pipe hamilton (1952: pl. 13, left); griffin (1952) states this pipe is Middle Mississippian and associates it with pipes from gahagan (coles creek and gibson aspects); spiro is the western margin of the pipe?s distribution (griffin, 1952:92) 448664 Pipe hamilton (1952: pl. 13, right); griffin (1952) states this pipe is Middle Mississippian and associates it with pipes from gahagan (coles creek and gibson aspects); spiro is the western margin of the pipe?s distribution (griffin, 1952:92) 448665 Pipe hamilton (1952: pl. 16a) 448667 Pipe hamilton (1952: pl. 17D); discussed by griffin (1952:91?92) 448668 Pipe hamilton (1952:34, pl. 8) 448670 celt hamilton (1952: pl. 24a); described therein as bone spatula 448673 celt hamilton (1952: pl. 52, bottom) N U M B E R 4 9 ? 21 5 (continued) Catalog Subnumber number (if any) Object References and notes 448674 celt hamilton (1952: pl. 52); Brain and Phillips (1996:378) 448677 Boatstone hamilton (1952:47, pl. 59) 448678 Discoidal hamilton (1952:47, pl. 58) 448679 Discoidal hamilton (1952:47) 448689 Mineral lumps caL Report 2018; Mci Report 2018.2 (smithsonian institution, Museum conservation institute, unpublished reports, 1975 and 2010) 448711 tube see Fundaburk and Foreman (1957) 448713 Bead hamilton (1952: pl. 69) 448740 Pendant same type as 448741 448741 Pendant Phillips and Brown (1984: pl. 152ib); also see Brown (1976: fig. 81g) for identical type in University of arkansas collection 448742 Pendant hamilton (1952: pl. 88D); Phillips and Brown (1984: pl. 336a) 448743 gorget not pictured elsewhere 448744 gorget not in Phillips and Brown (1978, 1984); fenestrated 448746 gorget Phillips and Brown (1984: pl. 274a) 448747 gorget Phillips and Brown (1984: pl. 138a) 448748 gorget Phillips and Brown (1978: pl. 23cb); Brain and Phillips (1996:447) 448749 gorget Phillips and Brown (1984: pl. 134a); Brain and Phillips (1996:453) 448750 gorget Phillips and Brown (1984: pl. 131D); Brain and Phillips (1996:454) 448751 gorget Phillips and Brown (1984: pl. 131D); Brain and Phillips (1996:454); see also 448750 448752 gorget Phillips and Brown (1984: pl. 144a); Brain and Phillips (1996:453) 448753 gorget Phillips and Brown (1978: pl. 23a) 448754 gorget Phillips and Brown (1984: pl. 128d); see also 448761s011 and s012 448755 gorget Phillips and Brown (1984: pl. 131B); Brain and Phillips (1996:453) 448756 gorget not located 448758 gorget Phillips and Brown (1984: pl. 125); Brain and Phillips (1996:453) 448759 gorget see watson (1950); Phillips and Brown (1984: pl. 152E) 448760 gorget Phillips and Brown (1984: pl. 275a,b); Brain and Phillips (1996:453) 448761 s011 gorget Phillips and Brown (1984: pl. 128c); Brain and Phillips (1996:452) 448761 s012 cup see Phillips and Brown (1984: pl. 128) 448761 s021 gorget Phillips and Brown (1984: pl. 146); Brain and Phillips (1996:455) 448761 s022 cup Phillips and Brown (1984: pl. 146) 448762 gorget Phillips and Brown (1984: pl. 338D) 448763 gorget Phillips and Brown (1984: pl. 152K) 448765 gorget Phillips and Brown (1984: pls. 131B,c, 134B [see also 448749], 148, 151E, 151n, 273F, 275ab [see also 448760], 275Bb, 277ca?b); see Brain and Phillips (1996:451) for hand and eye fragments; Brain and Phillips (1996:453?454) 448769 Pendant see Duffield (1964: pl. Xiii1?3) for ear pendants; Burnett (1945); clements (1945: pl. LViii d); hamilton (1952: pl. 106) (423234) shows man wearing one; Phillips and Brown (1984: pl. 338E) 448770 hook pendant hamilton (1952: pl. 83) 448772 Effigy Phillips and Brown (1984: pl. 285b) 448773 Effigy hamilton (1952: pl. 66, top right) 448782 cup Phillips and Brown (1978:30, fig. 11) 448783 cup Phillips and Brown (1978: pl. 83) 448784 cup hamilton (1952: pl. 115B); Burnett (1945: pl. LViii); Phillips and Brown (1984: pl. 331) (incorrectly numbered as 448819 in Phillips and Brown) 448785 cup Phillips and Brown (1978: pl. 62), where it is referred to as 448880, not 448785 448786 cup Phillips and Brown (1978: pl. 115) 448787 cup Phillips and Brown (1984: pl. 200) 448788 cup Phillips and Brown (1978: pl. 80) 448789 cup Phillips and Brown (1978: pl. 77) 448790 cup hamilton (1952: pl. 114a); Phillips and Brown (1978: pl. 57) 448791 cup Phillips and Brown (1978: pl. 24) 448792 cup Phillips and Brown (1978: pl. 122) 448793 cup Phillips and Brown (1978: pl. 119) 448794 cup similar to hamilton (1952: pl. 95a); Phillips and Brown (1984: pl. 317) 448795 cup Phillips and Brown (1984: pl. 322) 448796 cup Phillips and Brown (1984: pl. 296) 448797 cup Phillips and Brown (1984: pl. 330) 448798 cup Phillips and Brown (1984: pl. 165.1) 2 1 6 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y Catalog Subnumber number (if any) Object References and notes 448799 cup Phillips and Brown (1984: pl. 321) 448800 cup Phillips and Brown (1984: pl. 299) 448801 cup Phillips and Brown (1984: pl. 238a) 448802 cup see 448812 similar design; Phillips and Brown (1984: pl. 300) 448803 cup Phillips and Brown (1978: pl. 99a) 448804 cup Phillips and Brown (1984: pl. 259) 448805 cup Phillips and Brown (1978: pl. 100) 448806 cup Phillips and Brown (1984: pl. 260a) 448807 cup Phillips and Brown (1978: pl. 47), Braden style 448808 cup Phillips and Brown (1984: pl. 235) 448809 cup Phillips and Brown (1978: pl. 106) 448810 cup Duffield (1964, pl. XXXiii6); similar to Phillips and Brown (1978: pl. 67) 448811 cup hamilton (1952: pl. 121B); Phillips and Brown (1978: pl. 58) 448812 cup Phillips and Brown (1984: pl. 301) 448813 cup Phillips and Brown (1984: pl. 328) 448814 cup Phillips and Brown (1984: pl. 188) 448815 cup Phillips and Brown (1984: pl. 266a,B) 448816 cup Phillips and Brown (1984: pl. 122.1b) 448817 cup hamilton (1952: 116a); see 423270 for other horned snakes; Phillips and Brown (1984: pl. 241B) 448818 cup Duffield (1964, pl. XXXiii); Phillips and Brown (1984: pl. 269a,b) 448819 cup Phillips and Brown (1978: pl. 11a) 448820 cup Phillips and Brown (1978: pl. 55c), Braden B style, joins to 448871 (Phillips and Brown, 1978: pl. 55b) 448821 cup Phillips and Brown (1984: pl. 225a) 448822 cup not believed to be in Phillips and Brown (1978, 1984); paired figures facing a pole similar to Phillips and Brown (1984: pl. 317) 448823 cup not believed to be in Phillips and Brown (1984), compare with their pl. 283.1 448824 cup not believed to be in Phillips and Brown (1984) 448825 cup Pictured in Duffield (1964)?; Phillips and Brown (1984: pl. 169a) 448826 cup Phillips and Brown (1984: pl. 205a) 448827 cup hamilton (1952: pl. 137a); Phillips and Brown (1978: pl. 46) 448828 cup Phillips and Brown (1978: pl. 122.3a) 448829 cup Phillips and Brown (1978: pl. 2b) 448830 cup Phillips and Brown (1978: pl. 110a,c) 448831 cup hamilton (1952:75, pl. 124a) Phillips and Brown (1984: pl. 187a); Duffield (1964: pl. XXXii) 448832 cup Phillips and Brown (1984: pl. 310.1) 448833 cup Phillips and Brown (1984: pl. B-3Ra,b) 448834 cup Phillips and Brown (1984: pl. 297a) 448835 cup similar figures on 423234; see clements (1945: pl. XXXVi); hamilton (1952: pl. 119B); Phillips and Brown (1984: pl. 283a) 448836 cup Phillips and Brown (1984: pl. 207Ba) 448837 cup Phillips and Brown (1978: pl. 2a); see also 449929 448838 cup Phillips and Brown (1978: pl. 25a,c) 448839 cup Phillips and Brown (1984: pl. 286b); compare with hamilton (1952: pl. 126a); heye (catalog number 18/9118); and see Burnett (1945: pls. XXXiV and XL) 448840 cup Phillips and Brown (1984: pl. 267) 448841 cup Phillips and Brown (1984: pl. 263a); compare to hamilton (1952: pl. 138a); see 423287 448842 cup hamilton (1952: pl. 116B); Phillips and Brown (1984: pl. 167a,b,d) 448843 cup hamilton (1952: pl. 137a,B); not in Phillips and Brown (1984) 448844 cup Phillips and Brown (1984: pl. B-1g); see Phillips and Brown (1978: pl. 51) for rest of vessel 448845 cup hamilton (1952: pl. 104B); Burnett (1945: pls. XVii, XViii); Phillips and Brown (1984: pl. 219g) 448846 worked shell hamilton (1952: 138B); Phillips and Brown (1984: pl. 159a) 448847 cup hamilton (1952: pl. 123); Phillips and Brown (1978: pl. 1) 448848 cup Phillips and Brown (1984: pls. 226Db, 227a) 448849 cup Phillips and Brown (1984: pl. 31B), Braden a 448850 s01 cup Phillips and Brown?s (1978) pl. 59aa,ac along with fragment from 448874 is part of same shell and shown again in Phillips and Brown (1984: pl. B-2Bd; these were given subnumber 448850s01 N U M B E R 4 9 ? 21 7 (continued) Catalog Subnumber number (if any) Object References and notes 448850 s02 cup Phillips and Brown (1984: pl. B-12h) 448851 cup Phillips and Brown (1984: pl. 153a,b); hamilton (1952: pl. 127B) shows refitting piece of same shell 448852 cup hamilton (1952: pl. 138c); Phillips and Brown (1984: pl. 257b,c) 448853 s01 cup Phillips and Brown (1984: pl. 251a,b) 448853 s02 cup Phillips and Brown (1984: pl. 2571b) 448853 s03 cup Phillips and Brown (1984: pl. 258b) 448853 s04 cup one could be from vessel in Phillips and Brown (1984: pl. 252) 448853 s05 cup Phillips and Brown (1984: pl. 324) 448854 cup Phillips and Brown (1978: pl. 104); Phillips and Brown (1984: pl. B-8g) 448855 cup Phillips and Brown (1978: pl. 95) 448856 cup Duffield (1964, pl. XViii1?3); Phillips and Brown (1978, 1984) 448857 cup hamilton (1952: pl. 109) for comparable figure; Phillips and Brown (1978: pl. 70) 448858 s01?33 cup Phillips and Brown?s plates (1978: pl. 3a; 1984: pls. 168d, 195a) show bridge between 448880 and 1 fragment 448858a similar to pl. XXVii of clements (1945); also see Fundaburk and Foreman (1957: pl. 54) 448858 s01 cup Phillips and Brown (1984: pl. 197a), 2 refitted fragments 448858; 1 matching fragment from 448861 (Phillips and Brown, 1984: pl. 197b) 448858 s02 cup Phillips and Brown (1984: pl. 311); 2 men facing each other and holding snake staffs 448858 s03 cup not in Phillips and Brown (1978, 1984); birdman 448858 s04 cup called the atlatl cup in Phillips and Brown (1978: pl. 9a) 448858 s05 cup Possibly Phillips and Brown (1984: pl. 190) 448858 s06 cup Possibly Phillips and Brown (1984: pl. 158) 448858 s07 cup Possibly Phillips and Brown (1984: pl. 155a), fragment 2, one facing right, arms out in Bilbao stance 448858 s08 cup intertwined snake men theme, Phillips and Brown (1984: pl. 195a) 448858 s09 cup Phillips and Brown (1984: pl. 217b) man with long snake head and neck 448858 s10 cup Phillips and Brown (1984: pl. 210ac,ca,cb) 448858 s11 cup 16 fragments; shown in Phillips and Brown (1984: pls. 163E,F, 198B, 278b) 448858 s12 cup Phillips and Brown (1984: pl. 168.1a); see their note 448859 cup Phillips and Brown (1984: pl. 163a) 448860 cup hamilton (1952: pl. 123B); Phillips and Brown (1978: pl. 12a) 448861 s01?04 cup Phillips and Brown (1978: pl. 21); see also 448865-2 fragments in Phillips and Brown (1984: pls. 155B, 169c, 210D, 273a, 297b, B-8g; B-8K) 448861 s01 cup Phillips and Brown (1984: pl. 220b) 448861 s02 cup Phillips and Brown (1984: pl. 8-K), rows of heads facing left 448861 s03 cup Possible serpent man; has hair on end and lidded eye as on 448852 448861 s04 cup shows snout with teeth; small fragment, Phillips and Brown (1984: pl. 226a) 448861 s05 cup Phillips and Brown (1978: pl. 22g) 448861 s06 cup Phillips and Brown (1984: pl. 156); includes fragment of 448858 448862 cup hamilton (1952: pl. 126B); Phillips and Brown (1984: pl. 161b,c); see also 448858 448863 cup Phillips and Brown (1978: pl. 22D), Braden a 448864 cup Phillips and Brown (1978: pl. 103), Braden c 448865 cup hamilton (1952: pl. 127a); Phillips and Brown (1978: pl. 21a,h [448861], 21d,e [448865, 448880]) 448866 s01?12 cup Phillips and Brown (1978: pls. 25, 110c); Phillips and Brown (1984: pls. 171cb, 198hb, 241a, B-3a, B-3L); hamilton (1952: pl. 111); University of arkansas collection shell with snake-jaguar-wing combination is probable match; see Phillips and Brown (1984: pls. B8, B22, B52) for further matches; see also 448838 448866 s01 cup Various serpent motifs, mostly in Braden B style, cross-hatching; on one apical fragment the snake wraps around the apex and has excised bits and an underbelly of trapezoid with center rectangles; not pictured in Phillips and Brown (1984); twined snakes mostly; one is Phillips and Brown (1978: pl. 96.1) 448866 s02 cup Phillips and Brown (1984: pl. 26) include 2 fragments, 448880 and 448866 448866 s03 cup Phillips and Brown (1984: pl. 204c) 448866 s04 cup Phillips and Brown (1978: pl. 22h) 448866 s05 cup three fragments from one vessel; Phillips and Brown (1984: pl. 224b); largest fragment (448866) shows anus of animal surrounded by a nimbus with hair marks; 2 more smaller fragments of 448880 go with this vessel; 1 edge fragment is from another vessel in Phillips and Brown (1984: pl. 218aa) 2 1 8 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y Catalog Subnumber number (if any) Object References and notes 448866 s06 cup snake with dotted circle, modified trilobite on snake body; Phillips and Brown (1978: pl. 8aa,B), 1 large fragment 448866, 1 small fragment 448848, another small fragment 448876 included under parent 448866s06 448866 s07 cup Phillips and Brown (1984); smaller 4 body fragments are 448878 448866 s08 cup Phillips and Brown (1978: pl. 52a) 448866 s09 cup Larger apical fragment, Phillips and Brown (1978, 1984: pl. 198hb), small fragments numbered; 448866-B resembles exactly the representation in Phillips and Brown (1984: pl. 229); serpentine piasa in swastika arrangement, grouped because of motif similarity 448866 s10 cup Phillips and Brown (1978: pls. 30D [small], 35aa [larger]) 448866 s11 cup Phillips and Brown (1984: pls. 242, B-7i); 6 fragments are 448866 and 448880 448866 s12 cup two large fragments 488866, 2 small fragments 448880; Phillips and Brown (1978: pl. 28a,B), Braden a (1 fragment) 448867 s01?04 cup hamilton (1952: pls. 99c, 108, 110?117); Phillips and Brown (1984: pls. 241a, 257.1a, B-6ad) 448867 s01 cup Phillips and Brown (1984: pl. 243a,B) 448867 s02 cup one fragment is 448866 and 1 is 448880; 2 are 448867a; Phillips and Brown (1984: pl. 182B,c,g) and apical fragment not shown in Phillips and Brown (1984) 448867 s03 cup one large fragment is 448867, 4 others are 448880; Phillips and Brown (1984: pls. B6aa?ac, B6ad) 448868 cup Phillips and Brown (1978: pl. 91a,b,c); Braden, also 1 fragment 448880; Phillips and Brown (1984: pl. 186B) 448869 s01?03 cup Phillips and Brown (1984: pls. 183b, 186h, 208ha, 208.1ad, 209.1) 448869 s01 cup Phillips and Brown (1984: pls. 171cc,cb, 186h, 186B) (448868); Phillips and Brown (1984: pl. 209.1), woodpecker head with dotted beak; Phillips and Brown (1984: pl. 183b); several may not be pictured in Phillips and Brown 448869 s02 cup Barred oval, feathers, beaded forelock, divided arcs; similar to Phillips and Brown (1984: pl. 204.1) but does not match 448869 s03 cup Phillips and Brown (1984: pl. 208ha,b) 448870 cup Phillips and Brown (1978: pl. 86b) 448871 s01 cup Phillips and Brown (1984: pl.176), of a shot bison; pl. 176b is 448871 448871 s02 cup Maces and broken arrows; Phillips and Brown (1978: pl. 66a) 448871 s03 cup two fragments, neither of which is in Phillips and Brown (1978, 1984); show arrow fletching, arrowheads 448872 cup hamilton (1952: pl. 130a); Phillips and Brown (1984: pl. 270b,e,f,g); 2 large fragments 448872; 2 small fragments 448871 448873 cup Phillips and Brown (1984: pl. 201a) 448874 s01 cup hamilton (1952: pls. 134c,D, 135); Phillips and Brown (1978: pls. 59c [see 448850s01], 183, 174c [s03], 248a,b [s01]) 448874 s02 cup Very rare to see double outlining; see Phillips and Brown (1984: pl. a-5M) 448874 s03 cup consists of large fragments 423286, distal fragment 448874, small fragment 448880; single animal face with horns and snout on background of concentric scallop design; Phillips and Brown (1984: pl. 174) 448875 s01 cup Phillips and Brown (1984: pl. 170F) 448875 s02 cup Phillips and Brown (1984: pl. 172) 448875 s03 cup Phillips and Brown (1978: pl. 90) 448875 s04 cup Phillips and Brown (1984: pl. 333E) 448876 cup 448876?448876a is subsumed in s02; Phillips and Brown (1978: pls. 36a,b; 38b) 448876 s01 cup Phillips and Brown (1978: pl. 18) 448876 s02 cup one placed to Phillips and Brown (1978: pl. 37a), but the others are not in Phillips and Brown (1978, 1984) 448876 s03 cup Phillips and Brown (1984: pl. 179D) 448876 s04 cup not in Phillips and Brown (1978, 1984); spirals, dotted half circles, and sawtooth border 448877 cup Phillips and Brown (1978: pls. 34b, 122.1a, 122.3b; 1984: pl. 225); 448877a, Phillips and Brown (1984: pl. 333F) 448877 s01 cup Phillips and Brown (1984: pl. 338h) 448877 s02 cup Phillips and Brown (1984: pl. a-2 a,P) 448877 s03 cup Phillips and Brown (1984: pl. B-2c) 448877 s04 cup Phillips and Brown (1978: pl. 34B) 448877 s05 cup Phillips and Brown (1984: pl. 222Da) N U M B E R 4 9 ? 21 9 (continued) Catalog Subnumber number (if any) Object References and notes 448877 s06 cup not in Phillips and Brown (1978, 1984); Davis rectangle and concentric rayed, barred ovals 448877 s07 cup Phillips and Brown (1984: pl. 271); designated as 448877a by Phillips and Brown (1984) 448877 s08 cup Phillips and Brown (1984: pl. 272B) 448878 cup Phillips and Brown (1984: pls. 199K, B-3Rb, B-4Ba); 448878B, Phillips and Brown (1984: pl. 185f) 448878 s01 cup Phillips and Brown (1984: pl. 184) 448878 s02 cup Phillips and Brown (1984: pl. 185) 448878 s03 cup Phillips and Brown (1978: pl. 97a,b) (448880) 448879 cup see Duffield (1964: pl. XXXV2) for similar design; Phillips and Brown (1984: pl. 183.1a,c) 448879 s01 cup Phillips and Brown (1984: pl. 261) 448879 s02 cup Phillips and Brown (1984: pl. 260) 448880 s00 cup Phillips and Brown (1984: pl. 280.1) 448880 Various cup see Duffield (1964: pls. XXiX3, X1 [on pendant]); Phillips and Brown (1978: pls. 4c, 9b [with 448858], 13B, 14b, 15a, 20c, 21 [2 fragments, see 448865], 22E, 24 [with 448791], 26d [with 448866s02], 28B [with 448866s12]) 448880 Various cup Phillips and Brown (1978: pls. 30a, 31a, 33B, 43b, 53bc, 59F, 91a [with 448868], 92, 94a, 96a, 97B [with 448878s03], 99B [with 448803], 111a, 112a; 1984: pls. 161a,d [with 448862], 162f, 163Ba,c, 166.1, 168b,c, 168.1b [with 448858s12]) 448880 Various cup Phillips and Brown (1984: pls. 170a,D, 171a, 173aa,B, 174d [with 448874s03], 175d,e, 176a,e,f [with 448871s01], 177b, 179Ba,c, 179Da, 181F, 182d, 183c,d,e, 183.1d, 183.1b,e,f [with 448879s02]; 184d, 186D; 204ab,B) 448880 Various cup Phillips and Brown (1984: pls. 204.1c, 204.1a,b, 207a,Bb [with 448836], 208cb, 208.1aa, 208.1ab, 209.1E,F,g, 210ca [with 448858s10], 210Bb,E, 220a [with 448861s01], 222h, 224a,f [with 448866s05], 225e [with 448821], 227b,c [with 448848], 229b, 239h, 244a) 448880 Various cup Phillips and Brown (1984: pls. 262D, 266b [with 448815], 271.1e, 273g, 280.1b, 283.1 [mislabeled as 448823], 302.1, 310.1 [with 448832], 333a [with 448875], 334, 338ha,b [with 448877], a-4g,P, a-5i,K,R, B-1c,J,n, B-1h, B-2cb,c [with 448877], B-2h [with 448816]) 448880 Various cup Phillips and Brown (1984: pls. B-2i [with 448828], B-3E,g,n,Pa, B-3B,h,J,K,o, B-4Bb,E,g,L, B-6aa-c [with 448867], B-7a,D,h, B-7J,K, B-8F,h, B-8i, B-12ga, B-12Q) 448880 s01 cup Phillips and Brown (1978: pl. 43) 448880 s02 cup Phillips and Brown (1978: pl. 13B) 448880 s03 cup Phillips and Brown (1978: pls. 4c, 22E) 448880 s04 cup Phillips and Brown (1984: pl. 244a) 448880 s05 cup Phillips and Brown (1984: pls. 171a, 173a,B, 177B, 179Ba,c, 182D) 448880 s06 cup Phillips and Brown (1978: pl. 14B) 448880 s07 cup Phillips and Brown (1984: pl. 170a,D) 448880 s08 cup Phillips and Brown (1978: pls. 30a, 31a); Phillips and Brown (1984: pl. B-3K,o) 448880 s09 cup Phillips and Brown (1978: pl. 15) and Phillips and Brown (1984: pl. 197), fragment pictured twice 448880 s10 cup Phillips and Brown (1978: pl. 33B) 448880 s11 cup Phillips and Brown (1984: pl. a-4g) 448880 s12 cup Phillips and Brown (1978: pl. 111a) 448880 s13 cup not in Phillips and Brown (1978, 1984); bitriangular arrows 448880 s14 cup not in Phillips and Brown (1978, 1984); barred oval, snake, feathers 448880 s15 cup Phillips and Brown (1978: pl. 53B,c) 448880 s16 cup not in Phillips and Brown (1978, 1984); Bird Man 448880 s17 cup not in Phillips and Brown (1978, 1984); bird head on belt, pear-shaped object, pole with protuberance 448880 s18 cup Phillips and Brown (1984: pl. 302.1) 448880 s19 cup Phillips and Brown (1984: pl. 334) 448880 s20 cup Phillips and Brown (1984: pl. 183) 448880 s21 cup Phillips and Brown (1984: pls. 163Bb, 186D) 448880 s22 cup Phillips and Brown (1984: pl. 209.1E,F) 448880 s23 cup Phillips and Brown (1984: pls. 204.1c, 208.1ab, 209.1g); Phillips and Brown?s (1984) pls. 204ab,B and 204.1a,B may be included with these fragments, but there is no photographic evidence 2 2 0 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y Catalog Subnumber number (if any) Object References and notes 448880 s24 cup Phillips and Brown (1984: pl. 210Bb; 222h) 448880 s25 cup Phillips and Brown (1984: pl. 175D,E) 448880 s26 cup Phillips and Brown (1984: pl. 168B,c,D) 448880 s27 cup Phillips and Brown (1978: pl. 112a) 448880 s28 cup Phillips and Brown (1978: pl. 94a) 448880 s29 cup Phillips and Brown (1978: pl. 96a); Phillips and Brown (1984: pl. B-7Jb) 448880 s30 cup Phillips and Brown (1984: pl. 166.1a,B,c,D) 448880 s31 cup Phillips and Brown (1978: pl. 20c) 448880 s32 cup Phillips and Brown (1984: pl. B-1ha,hb) 448880 s33 cup Phillips and Brown (1978: pl. 92a) 448880 s34 cup Phillips and Brown (1984: pls. 208cc, B-3h,J, B-4Bb); 35 other pieces that are unintelligible 448880 s35 cup Phillips and Brown (1984: pls. 273g, a-5i, B-3B, B-4Ba,L); other pieces not in Phillips and Brown (1978, 1984) 448880 s36 cup Phillips and Brown (1984: pls. 210E, a-5R); other pieces not in Phillips and Brown (1978, 1984) 448880 s37 cup Phillips and Brown (1984: pls. 262D, 271.1E) 448880 s38 cup Either too little of design present to matter or unengraved 448880 s39 cup Phillips and Brown (1984: pl. 162F) 448880 s40 cup Phillips and Brown (1984: pls. a-5K, B-4E,g); one other fragment not found in Phillips and Brown 448880 s41 cup Phillips and Brown (1984: pls. 229B, 239h, B-12ga,Q) 448880 s42 cup Phillips and Brown (1978: pl. 59F); Phillips and Brown (1984: pls. a-4P, B-7K, B-8i); other pieces not in Phillips and Brown (1978, 1984) 448880 s43 cup all badly deteriorated; not in Phillips and Brown (1978, 1984) 448880 s44 cup worked shell; not in Phillips and Brown (1978, 1984) 448880 s45 cup Phillips and Brown (1978: pl. 62) 448881 cup not in Phillips and Brown (1978, 1984); unclear designs due to a heavy mat impression 448882 cup not in Phillips and Brown (1978, 1984) 448885 s04 worked shell not in Phillips and Brown (1978, 1984) 448890 Figure hamilton (1952: pl. 26) (shown with head) 448891 Effigy hamilton (1952: pl. 25) 448892 Effigy hamilton (1952: pl. 26); Brose (1985: pls. 96, 138); identified by Brose as seated male figure, caddoan culture spiro phase, Mississippian period ad 1200?1350 448893 Effigy bowl hamilton (1952: pl. 29B) 448894 Effigy bowl hamilton (1952: pl. 29a); see griffin (1952:92) for relation to Middle Mississippi area effigy bowls 448895 Effigy hamilton (1952: pl. 30a); griffin (1952:92); these ?dog heads? resemble shell specimens from the huasteca, Mexico 448896 Effigy hamilton (1952: pl. 26) 448897 worked wood hamilton (1952: pl. 24B) (wooden blade) 448899 worked wood hamilton (1952: pl. 24c); Burnett (1945, pl. LXXXV) 448915 cloth fragment Rogers et al. (2002) 448916 cord Rogers et al. (2002) 448917 cordage, cloth Rogers et al. (2002) 448919 cloth Rogers et al. (2002) 448934 Basketry hamilton (1952: pl. 67) References anderson, D. g., and K. E. sassaman, eds. 1996. The Paleoindian and Early Archaic Southeast. tuscaloosa: University of alabama Press. Baerreis, D. a. 1947. spiro Focus Basketry. The Museum of the University of Oklahoma, Information Series, circular 2:1?12. ???. 1957. the southern cult and the spiro ceremonial complex. Bulletin of the Oklahoma Anthropological Society, 5:23?28. Banks, L. D. 1984. ?Lithic Resources and Quarries.? in Prehistory of Oklahoma, ed. R. E. Bell, pp. 65?95. new York: academic Press. ???. 1990. From Mountain Peaks to Alligator Stomachs: A Review of Lithic Sources in the Trans- Mississippi South, the Southern Plains, and Adjacent Southwest. okla- homa anthropological society Memoir, no. 4. oklahoma city: oklahoma anthro- pological society. Barker, a. w., and t. R. Pauketat, eds. 1992. Lords of the Southeast: Social Inequality and the Native Elites of Southeastern North America. archeological Papers of the american anthropological association, no. 3. washington, D.c.: american an- thropological association. Barker, a. w., c. E. skinner, M. s. shackley, M. D. glascock, and J. D. Rogers. 2002. Mesoamerican origin for an obsidian scraper from the Precolumbian southeastern United states. American Antiquity, 67(1):103?108. Bauxar, J. J. 1953. Evidence of a substructure chamber under the Brown Mound at spiro. American Antiquity, 19(2):169?170. Beck, R. a. 2003. consolidation and hierarchy: chiefdom Variability in the Mississip- pian southeast. American Antiquity, 68(4):641?661. Bell, R. E. 1947. trade Materials at spiro Mound as indicated by artifacts. American Antiquity, 12(3):181?184. ???. 1953. Pottery Vessels from the spiro Mound, cr- 1, Le Flore county, oklahoma. Bulletin of the Oklahoma Anthropological Society, 1:25?38. ???. 1958. Guide to the Identification of Certain American Indian Projectile Points. special Bulletin of the oklahoma anthropological society, no. 1. oklahoma city: oklahoma anthropological society. ???. 1960. Guide to the Identification of Certain American Indian Projectile Points. special Bulletin of the oklahoma anthropological society, no. 2. oklahoma city: oklahoma anthropological society. 2 2 2 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y ???. 1972. The Harlan Site, Ck- 6, a Prehistoric Mound Center in Cherokee County, Eastern Oklahoma. oklahoma anthropological society Memoir, no. 2. oklahoma city: oklahoma anthropological society. ???. 1980. ?Fourche Maline: an archaeological Manifesta- tion in Eastern oklahoma.? in Caddoan and Poverty Point Archaeology: Essays in Honor of Clarence Hungerford Webb, ed. J. L. gibson. Louisiana Archaeology, 6:83?125. ???. 1984. ?arkansas Valley caddoan: the harlan Phase.? in Prehistory of Oklahoma, ed. R. E. Bell, pp. 221?240. new York: academic Press. Bennett, c., and J. weymouth. 1982. ?a Magnetic survey in the Plaza area of the spiro Mounds site.? in Spiro Archaeol- ogy: 1980 Research, by J. D. Rogers, pp. 215?226. nor- man, okla.: oklahoma archaeological survey. Bense, J. a. 1994. Archaeology of the Southeastern United States: Paleoindian to World War i. san Diego, calif.: aca- demic Press. Berthoud, g., and F. savelli. 1979. our obsolete Production Mentality: the heresy of the communal Formation. Cur- rent Anthropology, 20:745?760. Braecklein, J. g. 1936. ?Letter to smithsonian institution.? ac- cession File 137937. national Museum of natural history, smithsonian institution, washington, D.c. Brain, J. P., and P. Phillips. 1996. Shell Gorgets: Styles of the Late Prehistoric and Protohistoric Southeast. cambridge, Mass.: Peabody Museum Press. Brose, D. s., J. a. Brown, and D. w. Penney. 1985. Ancient Art of the American Woodland Indians. new York: harry abrams, in association with the Detroit institute of arts. Brown, J. a. 1966a. Description of the Mound Group. First Part of the second annual Report of caddoan archaeology? spiro Focus Research. norman, okla.: University of okla- homa Research institute. ???. 1966b. The Graves and Their Contents. second Part of the second annual Report of caddoan archaeology?spiro Focus Research. norman, okla.: University of oklahoma Research institute. ???. 1971a. ?the Dimensions of status in the Burials at spiro.? in Approaches to the Social Dimensions of Mor- tuary Practices, ed. J. a. Brown, pp. 92?112. washington, D.c.: society for american archaeology. ???. 1971b. Pottery Vessels. First Part of the third an- nual Report of caddoan archaeology?spiro Focus Re- search. norman, okla.: University of oklahoma Research institute. ???. 1976. The Artifacts. second Part of the third annual Report of caddoan archaeology?spiro Focus Research. norman, okla.: University of oklahoma Research institute. ???. 1983. ?spiro Exchange connections Revealed by sources of imported Raw Materials.? in Southeastern Natives and Their Pasts: A Collection of Papers Honoring Dr. Robert E. Bell, ed. D. g. wyckoff and J. L. hofman, pp. 129?162. norman, okla.: oklahoma archeological survey. ???. 1984a. Prehistoric Southern Ozark Marginality: A Myth Exposed. Missouri archaeological society special Publica- tion, no. 6. columbia, Missouri: Missouri archaeological society. ???. 1984b. ?arkansas Valley caddoan: the spiro Phase.? in Prehistory of Oklahoma, ed. R. E. Bell, pp. 241?263. new York: academic Press. ???. 1996. The Spiro Ceremonial Center: The Archaeology of Arkansas Valley Caddoan Culture in Eastern Oklahoma. Volumes 1?2. Memoirs of the Museum of anthropology, no. 29. ann arbor: University of Michigan. ???. 2010. ?cosmological Layouts of secondary Burials as Political instruments.? in Mississippian Mortuary Practices: Beyond Hierarchy and the Representationist Perspective, ed. L. P. sullivan and R. c. Mainfort, Jr., pp. 30?53. gaines- ville: University of Florida Press. Brown, J. a., and R. E. Bell. 1964. The First Annual Report of Caddoan Archaeology?Spiro Focus Research. norman, okla.: University of oklahoma Research institute. Brown, J. a., R. E. Bell, and D. g. wyckoff. 1978. ?caddoan settlement Patterns in the arkansas River Drainage.? in Mississippi Settlement Patterns, ed. B. D. smith, pp. 169? 200. new York: academic Press. Brown, J. a., and J. Kelly. 2000. ?cahokia and the south- eastern ceremonial complex.? in Mounds, Modoc and Mesoamerica, ed. s. R. ahler, pp. 469?510. illinois state Museum scientific Papers, Vol. 28. springfield: illinois state Museum. Brown, J. a., and J. D. Rogers. 1989. Linking spiro?s artistic styles: the copper connection. Southeastern Archaeology, 8:1?8. ???. 1999. aMs Dates on artifacts of the southeastern cer- emonial complex from spiro. Southeastern Archaeology, 18:11?18. Burnett, E. K. 1945. The Spiro Mound Collection in the Mu- seum. new York: Museum of the american indian, heye Foundation. clements, F. E. 1945. Historical Sketch of the Spiro Mound. contributions to the Museum of the american indian, heye Foundation, no. 14:48?68. new York: Museum of the american indian, heye Foundation. clements, F. E., and associates. 1935?1938. Preliminary Gen- eral Report on Archaeological Work in Oklahoma Spon- sored by the University of Oklahoma as a Works Progress Administration Project. wPa Progress Report 4844. wash- ington, D.c.: national anthropological archives, smithso- nian institution. Dickson, D. 1989. Lithic Raw Materials in the southern ozark highlands. Unpublished manuscript. Department of an- thropology, national Museum of natural history, smithso- nian institution, washington, D.c. Duffield, L. F. 1964. Engraved Shells from the Craig Mound at Spiro, Le Flore County, Oklahoma. Memoir 1. oklahoma city: oklahoma anthropological society. N U M B E R 4 9 ? 22 3 ???. 1973. the oklahoma craig Mound: another Look at an old Problem. Bulletin of the Oklahoma Anthropological Society, 22:1?10. Fischbeck, h. J., J. D. Rogers, s. R. Ryan, and F. E. swenson. 1989. sourcing ceramics in the spiro Region: a Preliminary study Using Proton- induced X- ray Emission (PiXE) analy- sis. Midcontinental Journal of Archaeology, 14(1):3?18. Fowler, M. L. 1989. The Cahokia Atlas: A Historical Atlas of Cahokia Archaeology. studies in illinois archaeology, no. 6. springfield: illinois historic Preservation agency. Fundaburk, E. L., and M. D. Foreman, eds. 1957. Sun Circles and Human Hands: The Southeastern Indians Art and In- dustries. Luverne, ala.: authors, self- published. galloway, P. 1989. The Southeastern Ceremonial Complex: Ar- tifacts and Analysis. Lincoln: University of nebraska Press. galm, J. R. 1978. Archaeological Investigations at Wister Lake, Le Flore County, Oklahoma. archaeological Research and Management center, Research series, no. 1. norman, okla.: archaeological Research and Management center. ???. 1981. Prehistoric cultural adaptations in the wister Valley, Eastern oklahoma. Ph.D. diss., washington state University, Pullman. ann arbor, Mich.: University Micro- films. ???. 1984. ?arkansas Valley caddoan Formative: the wister and Fourche Maline Phases.? in Prehistory of Oklahoma, ed. R. E. Bell, pp. 199?219. new York: academic Press. galm, J. R., and P. Flynn. 1978. The Cultural Sequences at the Scott (34Lf-11) and Wann (34Lf-27) Sites and Prehistory of the Wister Valley. archaeological Research and Manage- ment center, Research series, no. 3. norman, okla.: ar- chaeological Research and Management center. goodman, D., Y. nishimura, and J. D. Rogers. 1995. gPR (ground Penetrating Radar) time slices in archaeological Prospection. Archaeological Prospection, 2:85?89. griffin, J. B. 1952. an interpretation of the Place of spiro in southeastern archaeology. The Missouri Archaeologist, 14:89?106. hall, R. L. 1997. An Archaeology of the Soul: North Ameri- can Indian Belief and Ritual. Urbana: University of illinois Press. hamilton, h. w. 1952. the spiro Mound. The Missouri Archae- ologist, 14:17?88. hamilton, h. w., J. t. hamilton, and E. F. chapman. 1974. Spiro Mound Copper. Memoir, Missouri archaeological society, no. 11. columbia, Mo.: Missouri archaeological society. henry, D. o. 1978. Big hawk shelter in northeastern okla- homa: Environmental, Economic, and cultural changes. Journal of Field Archaeology, 5:269?287. henry, D. o., B. h. Butler, and s. a. hall. 1979. the Late Prehis- toric human Ecology of Birch creek Valley, northeastern oklahoma. Plains Anthropologist, 24(85):207?238. hilgeman, s. L. 1985. Lower ohio Valley negative Painted ce- ramics. Midcontinental Journal of Archaeology, 10(2):195? 214. hitchcock, R. K. 1982. ?Patterns of sedentism among the Bo- sarwa of Eastern Botswana.? in Politics and History in Band Societies, ed. E. Leacock and R. B. Lee, pp. 223?268. cambridge: cambridge University Press. hoffman, M. J. 1978. conservation systems for the spiro tex- tiles. Master?s thesis, University of arkansas, Fayetteville. horton, E., and g. sabo iii. 2011. spiro Perishable Project Makes Exciting new Discovery. Field Notes, 358:9?13. Johnson, g. a. 1982. ?organizational structure and scalar stress.? in Theory and Explanation in Archaeology: The Southampton Conference, ed. c. Renfrew, M. J. Rowlands, and B. a. segraves, pp. 389?421. new York: academic Press. Justice, n. D. 1987. Stone Age Spear and Arrow Points of the Midcontinental and Eastern United States. Bloomington: indiana University Press. Kay, M., g. sabo iii, and R. J. Merletti. 1989. ?Late Prehis- toric settlement Patterning: a View from three caddoan civic- ceremonial centers in northwest arkansas.? in Con- tributions to Spiro Archaeology: Mound Excavations and Regional Perspectives, ed. J. D. Rogers, D. g. wyckoff, and D. a. Peterson, pp. 129?157. studies in oklahoma?s Past, no. 16. norman, okla.: oklahoma archaeological survey. King, a. 2007. Southeastern Ceremonial Complex: Chronology, Content, Context. tuscaloosa: University of alabama Press. King, a., and M. s. Meyers, eds. 2002. Frontiers, Backwaters, and Peripheries: Exploring the Edges of the Mississippian world. Southeastern Archaeology, 21(2):113?116. King, M. E., and J. gardner 1981. ?the analysis of textiles from spiro Mound, oklahoma.? in The Research Potential of Anthropological Museum Collections, ed. a. E. cantwell, J. B. griffin, and n. a. Rothschild. Annals of the New York Academy of Science, 376:123?139. Kozuch, L. 2002. Olivella beads from spiro and the Plains. American Antiquity, 67(4):697?709. Kuttruff, J. 1988. textile attributes and Production complexity as indicators of caddoan status Differentiation in the arkan- sas Valley and southern ozark Regions. Ph.D. diss., Depart- ment of anthropology, ohio state University, columbus. La Vere, D. 2007. Looting Spiro Mounds: An American King Tut?s Tomb. norman: University of oklahoma Press. Leonhardy, F. 1989. ?the Physical stratigraphy of copple Mound.? in Contributions to Spiro Archaeology: Mound Excavations and Regional Perspectives, ed. J. D. Rogers, D. g. wyckoff, and D. a. Peterson, pp. 15?26. norman, okla.: oklahoma archaeological survey. Lintz, c. 1978. An Archaeological Survey of the Kerr?McGee Choctaw Coal Mine Facility, Haskell County, Oklahoma. Project Report series no. 2. norman, okla: archaeological Research and Management center, University of oklahoma. MacDonald, a. B. 1935. ?a ?King tut? tomb in the arkansas Valley.? Kansas City Star, 14 December 1935. Meillassoux, c. 1960. Essai d?interpr?tation du ph?nomene ?conomique daus les soci?ti?s traditionnelles d?anto- subsistence. Cahiers d?Etudes Africains, 4:36?67. 2 2 4 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y Merriam, L. g., and c. J. Merriam. 2004. The Spiro Mound: A Photo Essay. oklahoma city, okla.: Merriam station Books. Milner, g. R. 1998. The Cahokia Chiefdom: Archaeology of a Mississippian Society. washington, D.c.: smithsonian insti- tution Press. Mohrman, D. 1939. the Detection of Fraudulent copper needles. Transactions of the Illinois Academy of Sciences, 32:60?61. Moorehead, w. g. 1897. the hopewell group. The Antiquarian 1(10):254?264. ???. 2000. The Cahokia Mounds. tuscaloosa: University of alabama Press. Munsell color co. 1973. Munsell soil color charts. Baltimore, Md.: Munsell color co. Muto, g. 1978. The Habiukut of Eastern Oklahoma: Parris Mound, Part 1, Phase 1, an Archaeological Report. series in anthropology, no. 3. oklahoma city: oklahoma histori- cal society. national Museum of natural history. 1936. ?J. g. Braecklein loan.? accession File 137937. smithsonian institution, washington, D.c. ???. 1937. ?inventory list of J. g. Braecklein gift.? accession File 145540. smithsonian institution, washington, D.c. ???. 1960. ?harry M. trowbridge purchase.? accession File 228126. smithsonian institution, washington, D.c. orr, K. g. 1946. the archaeological situation at spiro, oklahoma: a Preliminary Report. American Antiquity, 11(4):228?256. Pauketat, t. R. 1993. Temples for Cahokia Lords. Memoirs of the Museum of anthropology, no. 26. ann arbor: Univer- sity of Michigan. ???. 1994. The Ascent of Chiefs: Cahokia and Mississippian Politics in Native North America. tuscaloosa: University of alabama Press. ???. 2007. Chiefdoms and Other Archaeological Delusions. Lanham, Md.: altamira Press. ???. 2009. Cahokia: Ancient America?s Great City on the Mississippi. new York: Viking. Pauketat, t. R., and t. E. Emerson, eds. 1997. Cahokia: Domi- nation and Ideology in the Mississippian World. Lincoln: University of nebraska Press. Payne, c., and J. F. scarry. 1998. ?town structure at the Edge of the Mississippian world.? in Mississippian Towns and Sacred Spaces: Searching for an Architectural Grammar, ed. R. B. Lewis and c. stout, pp. 22?48. tuscaloosa: University of alabama Press. Pendergast, J. F. 1989. ?the significance of some shell Excavated on iroquoian archaeological sites in ontario.? in Proceed- ings of the 1986 Shell Bead Conference, Selected Papers, ed. L. ceci, pp. 97?112. Research Records, no. 20. Rochester, new York: Rochester Museum and science center. Perino, g. 1968. A Guide to Certain Projectile Points. special Bulletin of the oklahoma anthropological society, no. 3. oklahoma city: oklahoma anthropological society. ???. 1970. A Guide to Certain Projectile Points. special Bulletin of the oklahoma anthropological society, no. 4. oklahoma city: oklahoma anthropological society. Perttula, t. K. 1989. ?the Late Prehistoric Period in the western ozarks of southwest Missouri.? in Contributions to Spiro Archaeology: Mound Excavations and Regional Perspec- tives, ed. J. D. Rogers, D. g. wyckoff, and D. a. Peterson, pp. 117?128, studies in oklahoma?s Past, no. 16. norman, okla.: oklahoma archaeological survey. Peterson, D. 1989. ?copple Mound Excavations.? in Contri- butions to Spiro Archaeology: Regional Perspectives and Mound Excavations, ed. J. D. Rogers, D. g. wyckoff, and D. a. Peterson, pp. 27?41. studies in oklahoma?s Past, no. 16. norman, okla.: oklahoma archaeological survey. Phillips, P., and J. a. Brown. 1978. Pre- Columbian Shell Engrav- ings from the Craig Mound at Spiro, Oklahoma, Part 1. cambridge, Mass.: Peabody Museum of archaeology and Ethnology. ???. 1984. Pre- Columbian Shell Engravings from Craig Mound at Spiro, Oklahoma, Part 2. cambridge, Mass.: Peabody Museum of archaeology and Ethnology. Phillips, P., J. a. Ford, and J. B. griffin. 1951. Archaeological Sur- vey on the Lower Mississippi Alluvial Valley, 1940?1947. Pa- pers of the Peabody Museum of archaeology and Ethnology, no. 25. cambridge, Mass.: Peabody Museum Press. Plog, F. t. 1974. The Study of Prehistoric Change. new York: academic Press. Potter county historical survey committee. 1964. Alibates Flint Quarries: A National Monument, 2 Volumes. amarillo, texas: Panhandle geological society. Powell, s. 1983. Mobility and Adaptation: The Anasazi of Black Mesa, Arizona. southern carbondale: illinois University Press. Rafferty, J. E. 1985. the archaeological Record on sedentari- ness: Recognition, Development, and implications. Ad- vances in Archaeological Method and Theory, no. 8, ed. M. B. schiffer, pp. 113?156. new York: academic Press. Rogers, J. D. 1980. Spiro Archaeology: 1979 Excavations. stud- ies in oklahoma?s Past, no. 6. norman, okla.: oklahoma archeological survey. ???. 1982. Spiro Archaeology: 1980 Research. studies in oklahoma?s Past, no. 9. norman, okla.: oklahoma ar- chaeological survey. ???. 1983. ?social Ranking and change in the harlan and spiro Phases of Eastern oklahoma.? in Southeastern Na- tives and Their Pasts: Papers in Honoring Robert E. Bell, ed. D. F. wyckoff and J. hofman, pp. 17?128. studies in oklahoma?s Past, no. 11. norman, okla: oklahoma ar- cheological survey. ???. 1989. ?settlement contexts for shifting authority in the arkansas Basin.? in Contributions to Spiro Archaeology: Mound Excavations and Regional Perspectives, ed. J. D. Rogers, D. g. wyckoff, and D. a. Peterson, pp. 159?176. studies in oklahoma?s Past, no. 16. norman, okla.: okla- homa archeological survey. N U M B E R 4 9 ? 22 5 ???. 1991a. ?Patterns of change on the western Margin of the southeast, a.D. 600?900.? in Stability, Transformation, and Variation: The Late Woodland Southeast, ed. M. s. nas- saney and c. R. cobb, pp. 221?248. new York: Plenum. ???. 1991b. Regional Prehistory and the spiro site. South- eastern Archaeology, 10:63?69. ???. 1995. ?Dispersed communities and integrated house- holds: a Perspective from spiro and the arkansas Basin.? in Mississippian Communities and Households, ed. J. D. Rog- ers and B. D. smith, pp. 81?98. tuscaloosa: University of alabama Press. ???. 1996. Markers of social integration: the Development of centralized authority in the spiro Region. in Political Structure and Change in the Prehistoric Southeastern United States, ed. J. F. scarry, pp. 53?68. gainesville, Florida: Uni- versity Press of Florida. ???. 2006. chronology and the Demise of chiefdoms: East- ern oklahoma in the sixteenth and seventeenth centuries. Southeastern Archaeology, 25:20?28. ???. 2009a. ?the spiro site.? in Archaeology in America: An Encyclopedia, ed. F. P. McManamon, pp. 324?329. west- port, connecticut: greenwood Press. ???. 2009b. ?the harlan site.? in Archaeology in America: An Encyclopedia, ed. F. P. McManamon, pp. 322?323. westport, connecticut: greenwood Press. ???. 2011. stable isotope analysis and Diet in Eastern okla- homa. Southeastern Archaeology, 30. Rogers, J. D., L. E. albert, and F. winchell. 2000. chronometrics at the norman site. Caddoan Archaeology, 11:11?18. Rogers, J. D., c. J. Dove, M. heacker, and g. R. graves. 2002. identification of Feathers in textiles from the craig Mound at spiro, oklahoma. Southeastern Archaeology, 21:245? 251. Rogers, J. D., M. Moore, and R. greaves. 1982. Spiro Archaeol- ogy?The Plaza. studies in oklahoma?s Past, no. 10. nor- man, okla.: oklahoma archaeological survey. Rogers, J. D., and g. sabo iii. 2004. ?the caddos.? in Hand- book of North American Indians, ed. w. c. sturtevant. Volume 14: Southeast, ed. R. D. Fogelson, pp. 616?631. washington, D.c.: smithsonian institution Press. Rogers, J. D., and B. D. smith, eds. 1995. Mississippian Commu- nities and Households. tuscaloosa: University of alabama Press. Rogers, J. D., D. g. wyckoff, and D. Peterson, eds. 1989. Con- tributions to Spiro Archaeology: Regional Perspectives and Mound Excavations. studies in oklahoma?s Past, no. 16. norman, okla.: oklahoma archaeological survey. Rohrbaugh, c. L. 1982. spiro and Fort coffee Phases: chang- ing cultural complexes of the caddoan area. Ph.D. diss., University of wisconsin, Madison. ann arbor, Mich.: Uni- versity Microfilms. ???. 1985a. Lfgei, the geren site, 34LF36, of the spiro Phase. Bulletin of the Oklahoma Anthropological Society, 34:9?81. ???. 1985b. the subdivision of spiro Phase. Midcontinental Journal of Archaeology, 10(2):155?170. Rolingson, M. a. 1987. ?an assessment of the significance of clay- tempered ceramics and Platform Mounds at the toltec Mounds site.? in The Emergent Mississippian: Pro- ceedings of the Sixth Mid- South Archaeological Conference, ed. R. a. Marshall, pp. 107?116. occasional Papers 87- 01. Mississippi state: cobb institute of archaeology, Missis- sippi state University. scarry, J. F., ed. 1996. Political Structure and Change in the Pre- historic Southeastern United States. gainesville: University Press of Florida. schambach, F. F. 1982. ?an outline of Fourche Maline culture in southwest arkansas.? in Arkansas Archeology in Review, ed. n. L. trubowitz and M. D. Jeter, 132?197. Research se- ries, no. 15. Fayetteville: arkansas archeological survey. ???. 1993. ?some new interpretations of spiroan culture history.? in Archaeology of Eastern North America: Papers in Honor of Stephen Williams, ed. J. B. stoltman, pp. 187? 230. archaeological Report, no. 25. Jackson: Mississippi Department of archives and history. ???. 1998. Pre- Caddoan Cultures in the Trans- Mississippi South: A Beginning Sequence. Research series, no. 53. Fay- etteville: arkansas archeological survey. schroeder, s. 2004. Power and Place: agency, Ecology, and his- tory in the american Bottom, illinois. Antiquity, 78:812? 827. sievert, a. K. 1992. Maya Ceremonial Specialization: Lithic Tools from the Sacred Cenote at Chichen Itza, Yucatan. Madison, wis.: Prehistory Press. ???. 2003. ?spiro Painted Maces and shell cups: the scien- tific Use of artifacts without context.? in Theory, Method, and Practice in Modern Archaeology, ed. R. J. Jeske and D. K. charles, pp. 182?194. westport, conn.: Praeger. smith, B. D. 1984. Mississippian Expansion: tracing the histor- ical Development of an Explanatory Model. Southeastern Archaeology, 3:13?32. ???. 1986. the archaeology of the southeastern United states: From Dalton to de soto, 10,500?500 B.P. Advances in World Archaeology, 5:1?95. ???, ed. 1990. The Mississippian Emergence. washington, D.c.: smithsonian institution Press. story, D. a., ed. 1981. Archaeological Investigations at the George C. Davis Site, Cherokee County, Texas: Summers of 1979 and 1980. occasional Papers, no. 1. austin: texas archaeological Research Laboratory. suhm, D. a., a. D. Krieger, and E. B. Jelks. 1954. An Introduc- tory Handbook of Texas Archaeology. Bulletin of the texas archeological society, no. 25. austin: texas archeological society. thoburn, J. B. 1930. the northern caddoan Peoples of Prehis- toric times and the human origin of the natural Mounds, so called, of oklahoma and neighboring states. thoburn collection, oklahoma historical society, oklahoma city. 2 2 6 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y ???. 1931. ?the Prehistoric cultures of oklahoma.? in Ar- chaeology of the Arkansas River Valley, ed. w. K. Moore- head, pp. 53?82. andover, Mass.: Phillips academy. thorne, R. M., and B. J. Broyles, eds. 1968. handbook of Pot- tery types Found in Mississippi. SEAC Bulletin, 7. thurmond, J. P. 1991. archeology of the Dempsey Divide: a Late archaic/woodland hotspot on the southern Plains. Bulletin of the Oklahoma Anthropological Society, 39:103?157. trowbridge, h. M. 1938. analysis of spiro Mound textiles. American Antiquity, 4(1):51?53. ???. 1944. the trowbridge collection of spiro Mound arti- facts. Journal of the Illinois Archaeological Society, July:21?23. ???. 1958. ?catalog of artifacts from spiro Mound.? na- tional anthropological archives, w. wedel collection, cor- respondence file: ?trowbridge.? smithsonian institution, washington, D.c. turgeon, D. D., h. E. Bogan, E. V. coan, w. K. Emerson, w. g. Lyons, w. L. Pratt, c. F. E. Roper, a. scheltema, F. g. thompson, and J. D. williams. 1988. Common and Scientific Names of Aquatic Invertebrates from the United States and Canada: Mollusks. special Publication no. 16. Bethesda, Md.: american Fisheries society. watson, V. D. 1950. The Wulfing Plates: Products of Prehistoric Americans. st. Louis, Mo.: washington University Press. white, D. R., g. P. Murdock, and R. scaglion. 1971. natchez class and Rank Reconsidered. Ethnology, 10:369?388. white, s. 1940. human Effigy Pipes from spiro Mound, Le Flore county, oklahoma. The Oklahoma Prehistorian, 3(1): 10?11. whitford, a. c. 1941. Textile Fibers Used in Eastern Aboriginal North America. anthropological Papers of the american Museum of natural history, no. 38. new York: american Museum of natural history. willoughby, c. c. 1952. textile Fabrics from the spiro Mound. The Missouri Archaeologist, 14:107?118. wilson, g. D. 2010. community, identity, and social Memory at Moundville. American Antiquity, 75:3?18. wyckoff, D. g. 1970. A Preliminary Report on 1970 Field Re- search at the Spiro Site, Le Flore County, Oklahoma. nor- man, okla.: oklahoma archaeological survey. ???. 1980. caddoan adaptive strategies in the arkansas Basin, Eastern oklahoma. Ph.D. diss., washington state University. ann arbor, Mich.: University Microfilms. wylie, a. a. 1995. ?archaeology and the antiquities Market: the Use of ?Looted? Data.? in Ethics in American Archaeol- ogy: Challenges for the 1990s, ed. M. J. Lynott and a. a. wylie, pp. 17?21. washington, D.c.: society for american archaeology. Zirkan, c. 1956. ?Record Unit 305: Letter to clifford Evans.? accession 210052. national Museum of natural history, smithsonian institution, washington, D.c. index Page numbers for tables are in italics and for figures are in bold. agriculture, 9 and sedentism, 9 alibates agatized dolomite, 76, 97 animal effigies, 166 animal hair, 57, 62, 65, 71, 72, 72 arkansas novaculite, 75?76, 85 Balloun, Joseph w., 14, 19, 22, 94 basketry, 25, 67?68, 185 coiled, 68, 69, 69 impressions on other artifacts, 40, 41, 70?71 nMnh collection of, 68 open- work woven, 70 square shallow, 68, 68?69 twill woven, 68, 68, 69?70, 70, 71 beads, 25, 113?23, 186 composite, 121 copper (see copper: beads) shell (see shell beads) stone, 121?22, 122, 122 stringing of, 57, 58 and textiles, 66 boatstones, 100, 101 bone artifacts, 167 awls, 134?35, 135, 167 gorgets, 128, 132?33, 133, 167 pendants, 125, 127 pins, 134, 134?35, 135 plaques, 135, 136 ritual objects, 25, 166, 167, 167, 168 Boone chert, 75, 88, 90, 91, 93 Braden style engraving, 43?47, 44, 127 2 2 8 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y Braden style engraving (continued) a, 43?45, 44, 44, 45, 51?52 B, 43, 44, 45?46, 45, 46, 51?52 c, 43, 44, 46?47, 46, 47, 51?52 Braecklein, J. g., 19?20, 20, 22, 96, 102, 123, 131, 134, 156, 179, 181 Brown, aunt Rachel, 14 Brown Mound, 3, 4, 6, 7, 11, 14 Busycon, 27, 39?40, 40, 41, 115, 116, 123, 127, 131, 132, 161, 165 caddoan, 14 northern region, 1, 9, 10 cahokia, 1, 36, 87, 181 cannel coal, 98, 99, 100, 126, 126 celts, 25, 91, 93, 93, 93, 97, 98, 99, 100, 100, 102 elongate, 98, 98, 99 spatulate, 98, 99, 100, 100, 100 ceramics, 25 characterizing features of, 27?29, 28, 183?84 clement Redware, 29, 37, 38 coles creek, 2, 31, 32, 36 crockett curvilinear incised (see croc- kett curvilinear incised ceramics) effigy, 29, 35?36, 36 French Fork- like agee incised, 2 Friendship Engraved, 29, 33?34 glazed, 29, 39 grit- tempered, 31, 33, 34, 36, 38 grog- tempered, 29, 30?35 haley Engraved, 29, 34, 35 hatchel Engraved, 29, 34, 35 Le Flore Plain (see Le Flore Plain ceramics) manufacture of, 30 Mississippian Plain, 35 old town Red, 29, 30, 32, 36?37 Pennington Punctate (see Pennington Punctate ceramics) Plum Bayou, 2 polychrome, 29, 39 Poteau Plain, 29, 36 raw material used for, 29?30 sanders Engraved, 29, 37, 37?38 sanders Plain, 29, 30, 30, 31, 32, 36?37 sanders Plain/clement Redware, 29, 37, 38 sanders Plain/old town Red, 29, 30, 32, 36?37 shell- tempered, 29, 35?36 sherd counts, 29 slipped, 29, 36?38 smithport Plain, 29, 30, 31, 31, 32 southwestern, 29, 38, 38?39 undifferentiated grog- tempered incised, 29, 35 undifferentiated grog- tempered plain, 29, 35 undifferentiated shell tempered, 29, 36 use of, 30 white Engraved, 29, 30, 31, 33, 34, 34 williams Plain (see williams Plain ceramics) woodward Plain, 29, 35, 36 chert Boone, 75, 79, 82, 86, 88, 90, 91, 93 Burlington, 75, 82, 83, 85, 86, 91 Devonian woodford, 76, 88 Dover, 76, 93, 95 Florence (Kay county), 75, 83, 91, 92, 92, 93, 95 Jefferson city?cotter, 75, 82, 85, 90, 90 Keokuk, 75, 85, 90 Mill creek, 76, 93, 95 Mississippian compton, 75 Mississippian Pierson, 76, 85 Mississippian Reed springs, 75, 79, 82, 83, 84, 85, 86, 88, 90, 95 ordovician cotter, 75, 85 Pitkin, 75, 82, 83, 86, 87 Red River jasper, 76, 79, 83 smoky hills (niobrara) jasper, 76, 91, 95 chiefdoms, 1, 9, 11, 12 chipped lithic weaponry, 73, 74 damage to, 75, 76, 77, 89 haftable bifaces, 91 large bifaces (see large bifaces) large projectile points (see large pro- jectile points) raw materials used for, 75?76, 81 small projectile points (see small pro- jectile points) clements, Forest, 15, 23 cloth, 25, 57, 57 decorations, 66 dyed, 60, 63, 64, 66, 67 simple weave, 57, 65, 67 spaced weft twining, 57, 61?62, 61, 62, 63, 64, 64, 65 twined tapestry, 57, 65, 65, 66, 66 weaving styles, 60?67 wrapped warp, 57, 64 See also textiles cones, 135?36, 136 copper, 160 beads, 25, 58, 122, 122?23, 123, 123 decoration on earspools, 107?8, 108, 109, 111, 113 on headdresses, 135 (see also copper pins) pins (see copper pins) preservation of organic materials by, 162, 170 ritual objects, 25 sheet, 24, 69, 135, 160?62, 161, 161 copper pins authentic, 133, 133?34 fraudulent, 25, 133, 134, 134 copple Mound, 3, 4, 6 coprolites, 172 corner- notched points, 77, 81, 89 agee, 77, 78, 79, 80, 81, 82?83, 82, 86, 89 ashley, 77, 80, 81, 83?84 ashley chocoville, 81, 84, 90 collier, 81, 84 coryell, 81, 84, 86 Martindale, 78, 81, 84 Massard, 81, 84, 84?85, 86 Rockwall, 79, 81, 83 scallorn, 81, 83, 85, 85 sequoyah, 79, 81, 83, 87 craig, william, 14 craig Mound, 3, 4, 5, 6, 7, 11, 16 closing of, 7 construction of, 7, 17 destruction of, 15, 16 pre- excavation, 14 premound burials, 2 craig style engraving, 43, 44, 127, 128 a, 43, 44, 47?48, 48, 51, 128, 129 B, 43, 44, 48?49, 49, 50, 51, 128, 130, 132 c, 43, 44, 49?51, 50, 51, 51, 52 crockett curvilinear incised ceramics, 29, 30, 31, 31?33, 33, 171 Pennington Punctate ceramics com- pared to, 33 crown maces. See maces Davis site, 9 discoidals, 159?60, 160, 160 Dollop Mound, 3, 4, 5?6 earspools, 25, 105?13, 106, 108 analysis of, 185?86 central boss, 106, 108, 111 composite, 106, 108, 111, 111 n u m b e r 4 9 ? 2 2 9 decorations on, 107?8, 108, 109, 112?13 depicted on shell cups, 45, 47, 48, 110 divided flange, 106, 108, 110?11, 111 effigy, 106, 108, 112, 112, 136 flanged ring, 106, 108, 112, 112 nesting, 106, 108, 111 perforated pulley- shaped (see perfo- rated pulley- shaped earspools) raw materials used for, 108 unperforated pulley- shaped (see un- perforated pulley- shaped earspools) wooden disc, 106, 108, 112 effigy pipes, 153, 154, 154 human, 154, 156?59, 156, 157, 158, 159 zoomorphic, 154, 154?56, 155, 158?59, 159 Emergent Mississippian period, 8, 9 Evans, george, 14?15 Evans phase, 2, 4, 8, 10, 85 population during, 11 transition to, 8?9 feathers, 57, 65, 72, 72 figurines, 162, 162 human effigy, 25, 162, 162?65, 163, 164, 165 shell, 162, 164?65, 165 stone, 162, 164, 164 wooden, 162, 162?64, 163 See also maskettes Fort coffee phase, 2, 4, 7 population during, 11 Fourche Maline phase, 2, 8?9, 85, 89 transition to Evans phase from, 9 fraudulent artifacts, 25, 133, 134, 134 functional contexts, 24 galena, 164, 164, 170, 171?72, 172 glass slag, 172 glauconite, 33, 171 gorgets, 25, 127?33 bone, 128, 132?33, 133, 167 design distribution of, 128 shell (see shell gorgets) solid, 130 great Mortuary, 7, 17 ?great temple Mound?, 14 hamilton, henry, 16, 17, 20, 20, 23 harlan phase, 2, 4, 4?7, 8, 9, 10 population during, 11 transition to norman phase from, 11 harlan site, 4, 10 harris, R. King, 23 hasinai caddo, 9 headdresses bone pins, 134, 134?35, 135 bone plaques, 135, 136 copper foil materials on, 135 copper pins (see copper pins) plaques and plates, 135 hematite, 126 house Mound 1, 3, 4 house Mound 2, 3, 4 house Mound 3, 3, 4 house Mound 4, 2, 3, 4, 6 house Mound 5, 3, 4?5, 6 buildings at, 4?5, 5 house Mound 6, 3, 4, 6 house Mounds, 7. See also specific House Mound human effigies. See effigy pipes: human; figurines: human effigy human remains, 175?78 femora, 176, 177, 178 foot bones, 178 hand bones, 177, 178 humerus, 176, 178 innominates, 176, 178 mandibles, 175?76, 177, 178 ribs, 177 skulls, 175, 177, 178 teeth, 175, 176 tibia, 176, 178 ulna, 177?78 vertebrae, 176?77, 178 large bifaces, 25, 91?95, 93, 92 celts (see celts) Kay, 90, 92, 93 maces (see maces) ovoid, 91?92, 92, 93 preforms, 93, 93, 93 sword- form (see sword- form bifaces) large projectile points, 25, 89?91, 90 eccentric, 90, 91 expanding stem, 90, 91 gary, 89?90, 90 graham cave, 90, 90, 91 Johnson, 90, 91 Lange, 90, 90, 90?91 leather, 170 Le Flore Plain ceramics, 2, 29, 31 compared to williams Plain ceramics, 31 Lightner Museum of hobbies, 20, 20?21 maces, 25, 93, 95?97, 96, 97 depicted on shells, 46, 51, 97 maize, 1, 9 Marginella, 115, 119, 120 maskettes, 165?66, 166 material types, 24 mats, 57, 67 twilled, 69?70, 70, 71 twined, 70 medallions. See maskettes Meyer, Richard K., 22 mica, 111, 172 middle- range societies, 1, 9 mineral lumps, burned, 171, 171 Mississippian cultural ?style?, 1, 2 Mississippian period, Emergent, 8, 9 Mound a (see Plaza Mound a) Mound B (see Plaza Mound B) Mound c (see Plaza Mound c) natchez, 9 national Museum of natural history spiro collection accession history, 19?23, 20 harris collection, 23 Meyer collection, 22, 39, 180 Mississippian research and, 181 representativeness of, 179?80 significance of, 180?81 summary of, 25 trowbridge collection, 21?22, 39, 55, 56, 56, 180 see also specific artifact type nMnh. See national Museum of natural history spiro collection norman phase, 2, 4, 6, 7, 10, 11 population during, 11 transition to, 11 novaculite, 75?76, 79, 81, 82, 83, 84?85, 86, 87, 88, 91, 92, 92, 93, 93, 96, 102 ocher, 171 Olivella, 115, 119?20, 120 Parris Mound, 6 pendants, 25, 123?27 bone, 125, 127 ceramic, 127 shell (see shell pendants) stone, 126, 126?27, 127 Pennington Punctate ceramics, 29, 33, 34 crockett curvilinear incised ceramics compared to, 33 2 3 0 ? s M i t h s o n i a n c o n t R i B U t i o n s t o a n t h R o P o L o g Y perforated pulley- shaped earspools, 106, 106, 106, 107, 108, 109, 110 decorations on, 107?8, 108, 109 manufacture of, 106?7 raw material used for, 107, 108 pipes, 25, 149, 150, 154, 154, 154, 186 effigy (see effigy pipes) elbow, 150, 153, 153, 158 Red River, 150, 153 t- shaped (see t- shaped pipes) plant fibers, 57?58, 72 platform mounds, 4, 6 Plaza Mound a, 3, 4, 5?6 Plaza Mound B, 3, 4, 5?6 Plaza Mound c, 3, 4, 5?6 Plaza Mounds, 7. See also specific Plaza Mound Plum Bayou culture, 2 Pocola Mining company, 15, 19, 73, 179, 180 population levels and social change, 11 at spiro site, 11 pulley- shaped earspools. See earspools quartz, 172 rabbit hair, 65, 71, 72 ranked societies, 1, 8, 9, 12 relic hunters, 13?15, 19, 166, 179, 180 ritual materials, 25, 149, 170, 173 animal effigies, 166 bone objects, 25, 166, 167, 167, 168 discoidals, 159?60, 160, 160 figurines (see figurines) leather pieces, 170 maskettes, 162, 165?66, 166 minerals, 171?72 pipes (see pipes) unworked, 171?73 wooden objects, 25, 168, 168?69, 169, 170 scullyville Mounds, 15 sedentism, 9 and agriculture, 9 shell beads, 25, 115 bowed (asymmetrical convex), 114, 115, 115?16, 117, 118 bulging, 114, 115, 117, 118 concave disc, 114, 115, 116 concavo- cylindrical, 114, 115, 116, 117 convexo- cylindrical, 114, 114, 115 cylindrical, 114, 115, 116, 117 disc, 113?14, 114, 115, 116, 119 drop- shaped, 114, 115 elliptical, 114, 114, 115 flat circular, 114, 114, 115 flattened hexagonal, 114, 115, 117, 118 gastropod and bivalve, 119?20, 120, 120 manufacture of, 113 mask, 115, 119, 119 pearls, 115, 120?21, 121 sculpted, 115, 115 spherical, 114, 115, 116, 116 styles of, 113?19, 114, 115 tooth, 114, 115, 119 shell cups. See shell vessels shell discs, 112?13 shell fragments, engraved, 51?52 shell gorgets, 127, 128 engraved fragments, 51?52 fenestrated, 51, 128?30, 128, 128, 129, 130, 131 iconographic styles, 127?28, 128 manufacturing styles, 127 mask, 128, 131?32, 132 perforated, 128, 130 sidewall, 128, 131, 131 solid, 128, 130 shell pendants, 123?26, 125, 126 Columella, 123?25, 124, 124, 125 shell vessels, 39, 25 analysis of, 184?85 conservation of, 184?85 engraved fragments, 51?52 engraving styles (see Braden style engraving; craig style engraving) incising on, 42, 42, 43?51, 44, 45, 46, 47, 48, 49, 50, 51, 52, 138?48 (see also Braden style engraving; craig style engraving) manufacture of, 41?42 nMnh collection of, 44, 51, 51?52 raw material used for, 39?40, 40 residues on, 40?41, 41 unengraved, 51?52 use of, 42?43 side- notched points, 78, 81, 88 haskell, 78, 79, 80, 81, 87 Keota, 78, 80, 81, 85?86, 86, 86, 87, 89 Morris, 79, 81, 87, 89 Reed, 77, 78, 80, 81, 85, 86, 86, 87, 88, 89 Reed/Keota variation, 78, 81, 86?87, 87 washita, 81, 86, 86, 87?88 small projectile points, 25, 76?88, 77, 78, 79, 80, 81, 86 caches of, 89 corner- notched (see corner- notched points) damage to, 77, 89 Edwards, 79, 81 heat alteration of, 88?89, 89 side- notched (see side- notched points) stemmed (see stemmed points) unshouldered (see unshouldered points) spiro excavations history of, 13?17 publicity of, 16 spiro phase, 2, 4, 6, 7?8 population during, 11 spiro site, 3 ceramics from (see ceramics) copper from (see copper) cultural phases, 2?8, 4 decline of, 12 excavations at (see spiro excavations) fraudulent artifacts from, 25, 133, 134, 134 grave lots, 2, 4, 7, 17 ?high- status? artifacts at, 10 ?Plaza? area, 4 population at, 11 role in region, 10?11 shell from, 23?24 (see also shell beads; shell discs; shell gorgets; shell pendants; shell vessels) stone bowls from, 53?54, 54 structure density at, 6 textiles (see textiles) transition to Fort coffee phase, 7 utilitarian implements from (see utili- tarian implements) weaponry from (see weaponry) wooden vessels from, 52?53, 53 stemmed points, 78 alba, 78?79, 78, 80, 80, 81, 86, 89 Bonham, 78, 81, 82 Bonham tahlequah, 81, 82 hayes, 78, 79?80, 81, 86, 89 hayes short, 78, 80, 81 homan, 81, 82 Perdiz, 77, 78, 78, 81, 82, 89 sword- form bifaces, 25, 93 Duck River, 93, 93?95, 94 n u m b e r 4 9 ? 2 3 1 fusi- elliptical, 93 general, 93, 95, 95 tapestry. See cloth: twined tapestry textiles, 15, 20, 23, 24, 25, 180 analysis of, 185 basketry (see basketry) beads and, 66 braid, 57, 57, 59, 59, 60 cloth (see cloth) conservation of, 185 cordage, 25, 57, 57, 58?60, 59 raw materials used for, 57?58, 71?72, 72 styles of, 57, 57, 58?65 trowbridge collection of, 21?22, 55, 56, 56 uses of, 55, 57, 67 wrapped basket fiber, 57, 59 yarn, 57, 57, 58, 58 thoburn, Joseph B., 13?14 tools, 101 abrading, 101, 102?3, 103 drills, 42, 92, 101, 102 flakes, 101, 102 hammerstones, 101, 103 hoes and spades, 93, 101, 101, 102 incising (graver), 86, 92, 101, 102 manos, 101, 103, 103 polishing stones, 101, 103, 103 scrapers, 95, 102, 102, 166, 167, 168 trowbridge, harry, 19, 20, 20, 21?22, 24, 38, 55, 56 t- shaped pipes, 149?50, 150, 152?53 chandler, 150, 150?51, 151 contracting bowl, 150, 152, 153 cylindrical bowl, 150, 151, 151?52, 152, 153 unperforated pulley- shaped earspools, 106, 110 raw materials used for, 108, 110 unshouldered points, 80, 88 Fresno, 80, 81, 88 Maud, 77, 81 nodena, 81, 88 Young, 81, 88 utilitarian implements, 101 chipped stone, 101?2 ground stone, 102?3 Vandagriff excavations, 14 Vandergriff excavations. See Vandagriff excavations Vandigriff excavations. See Vandagriff excavations ward Mound 1, 2, 3, 4, 6?7, 13 ward Mound 2, 2, 3, 7 weaponry, 25, 185 atlatl components, 100, 101, 101 chipped lithic (see chipped lithic weaponry) williams Plain ceramics, 2, 29, 30, 30, 31 compared to Leflore Plain ceramics, 31 wister Valley, 2 wood earspools, 106, 108, 112 figurines, 162, 162?64, 163 ritual objects, 25, 168, 168?69, 169, 170 vessels, 52?53, 53 woodland period, 2, 89, 127 Early, 91, 100 Late, 8, 85 Middle, 173 works Progress administration. See wPa craig Mound excavation wPa craig Mound excavation, 15?16, 179?80 yarn, 57, 57, 58, 58 All MAnuSCrIPTS ArE rEVIEwEd FOr AdHEr- EnCE TO THE SISP MAnuSCrIPT PrEPArATIOn And STylE GuIdE FOr AuTHOrS (available on the ?Submis- sions? page at www.scholarlypress.si.edu). Manuscripts not in compliance will be returned to the author. Manuscripts in- tended for publication in the Contributions Series are evalu- ated by a content review board and undergo substantive peer review. Accepted manuscripts are submitted for funding ap- proval and scheduling to the Publications Oversight Board. MInIMuM MAnuSCrIPT lEnGTH is thirty manuscript pages. If a manuscript is longer than average, an appropriate length will be determined during peer review and evaluation by the Content review Board. Authors may be asked to edit manuscripts that are determined to be too long. TExT must be prepared in a recent version of Microsoft word; use a Times font in 12 point for regular text; be double spaced; and have 1" margins. Each chapter/section must be saved in a separate file. rEquIrEd ElEMEnTS are title page, abstract page, table of contents, main text, and reference section. See the SISP Manuscript Preparation and Style Guide for Authors for the order of all elements. HEAdInGS should be styled so different levels of headings are distinct from each other and so the organization of the manu- script is clear. Insert one line space above and one line space below all headings. FrOnT MATTEr should include title page, abstract page, and table of contents. All other sections are optional. Ab- stracts must not exceed 300 words. Table of contents should include A-, B-, and C-level headings. TABlES (numbered, with captions, stubs, rules) should be submitted in separate MS word files; should include foot- notes, if appropriate; should have rules only at top, bottom, and beneath column heads. Printouts of each table should accompany the manuscript to ensure correct layout of data. Tabulations within running text should not be numbered or formatted like formal tables, and should be included in the text of the manuscript. FIGurE CAPTIOnS should be provided in a separate MS word file. FIGurES (e.g., photographs, line art, maps) should be num- bered sequentially (1, 2, 3, etc.) in the order called out; be placed throughout text, not at end of manuscript; have all components of composites lettered with lowercase letters and described in the caption; include a scale bar or scale descrip- tion, if appropriate; include any legends in or on the figure rather than in a caption. ArT must not be embedded in the main text. Figures must be original and submitted as individual TIFF or EPS files. resolution for art files must be at least 300 dpi for grayscale and color images and at least 1200 dpi for line art. Electronic images should measure no more than 100% and no less than 75% of final size when published. JPG files will not be accepted. Color images significantly increase costs so should be included only if required. Funding for color art is subject to approval by SISP and the Publications Oversight Board. In-TExT rEFErEnCES should be used rather than biblio- graphic notes and should follow the author-date system in the following format: ?(author last name, year)? or ?. . . author (year)?; ?(author, year:page used within the text)? or ?. . . author (year:page).? A full citation should be included in a ?references? section. EndnOTES are to be used in lieu of footnotes and should be keyed manually into a separate MS word file, in a sec- tion titled ?notes.? notes should not contain bibliographic information. Manually type superscript numerals in text and use full-sized numerals at the beginning of each note in the ?notes? section. SISP will determine the best placement of the notes section, either at the end of each chapter or at the end of the main text. rEFErEnCES should be in alphabetical order, and in chron- ological order for same-author entries. Each reference should be cited at least once in main text. Complete bibliographic in- formation must be included in all citations (e.g., author/editor, title, subtitle, edition, volume, issue, pages, figures). For books, place of publication and publisher are required. For journals, use the parentheses system for volume(number):pagination [e.g., ?10(2):5?9?]. do not use ?et al.?; all authors/editors should be included in reference citations. In titles, capitalize first word, last word, first word after colon, and all other words except articles, conjunctions, and prepositions. Exam- ples of the most common types of citations are provided in the SISP Manuscript Preparation and Author Style Guide. For questions regarding the guidelines, please email SISP at schol.press@si.edu. requirementS for SmithSonian SerieS Publication