UNIVERSITY OF CALGARY

Traces of the Past: Microscopic Residue Analysis on the Canadian Plateau, British

Columbia

by

Shannon Croft

A THESIS

SUBMITTED TO THE FACULTY OF GRADUATE STUDIES

IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE

DEGREE OF MASTER OF ARTS

DEPARTMENT OF ARCHAEOLOGY

CALGARY, ALBERTA

DECEMBER, 2011

© Shannon Croft 2011 Library and Archives Biblioth&que et Canada Archives Canada Published Heritage Direction du HI Branch Patrimoine de l'§dition

395 Wellington Street 395, rue Wellington Ottawa ON K1A 0N4 Ottawa ON K1A 0N4 Canada Canada Your file Votre reference ISBN: 978-0-494-83417-6

Our file Notre reference ISBN: 978-0-494-83417-6

NOTICE: AVIS: The author has granted a non­ L'auteur a accorde une licence non exclusive exclusive license allowing Library and permettant £ la Biblioth&que et Archives Archives Canada to reproduce, Canada de reproduire, publier, archiver, publish, archive, preserve, conserve, sauvegarder, conserver, transmettre au public communicate to the public by par telecommunication ou par I'lnternet, pr§ter, telecommunication or on the Internet, distribuer et vendre des theses partout dans le loan, distrbute and sell theses monde, & des fins commerciales ou autres, sur worldwide, for commercial or non­ support microforme, papier, 6lectronique et/ou commercial purposes, in microform, autres formats. paper, electronic and/or any other formats.

The author retains copyright L'auteur conserve la propri6t§ du droit d'auteur ownership and moral rights in this et des droits moraux qui protege cette these. Ni thesis. Neither the thesis nor la these ni des extraits substantiels de celle-ci substantial extracts from it may be ne doivent etre imprimis ou autrement printed or otherwise reproduced reproduits sans son autorisation. without the author's permission.

In compliance with the Canadian Conformement a la loi canadienne sur la Privacy Act some supporting forms protection de la vie priv6e, quelques may have been removed from this formulaires secondares ont ete enleves de thesis. cette these.

While these forms may be included Bien que ces formulaires aient inclus dans in the document page count, their la pagination, il n'y aura aucun contenu removal does not represent any loss manquant. of content from the thesis. Canada Abstract

This study examines microscopic plant and animal residues from 106 stone tools dating to the Late Period (4500 to 200 BP) obtained from the Canadian Plateau site White Rock Springs (EeRj-226), in the Hat Creek valley, interior British Columbia, Canada. Microscopic residues gleaned from artifacts are used as direct archaeological evidence to assess the diet and technologies of prehistoric peoples. Root food use was a particular focus of the study and was targeted by characterizing starch grains found on the stone tools. In addition to starch, microscopic trace residues of coniferous wood, herbaceous and/or woody tissues, phytoliths, feathers, pollen, fungal hyphae and lichen were extracted from the stone tools.

ii Acknowledgements I express my gratitude to my supervisor Brian Kooyman (University of Calgary) for his patience and expert guidance in creating methodology appropriate for starch residue analysis and his anticipation of problems. Dr. Kooyman draws from his very large and varied archaeological and biological experience, and his insightful comments have been invaluable to my thesis. Brian Kooyman and Sandra Peacock (University of British Columbia Okanagan) allowed me to join their excavation team at White Rock Springs in Hat Creek, British Columbia. Dr. Peacock kindly shared her EeRj-226 research applications and reports with me and gave me the opportunity to learn from her extensive understanding of earth ovens. I am grateful for the use of Sonia Zarrillo's (Ph.D. candidate, University of Calgary) starch residue extraction procedure and her space in Dr. Scott Raymond's Latin American Studies laboratory at the University of Calgary. I was provided with the lithic database of White Rock Springs tools, information compiled by Dr. David Pokotylo (University of British Columbia) and his students. Drs. Len Hills, Heather Addy (University of Calgary), and Rolf Mathewes (Simon Fraser University) provided pundit advice on pollen and fungal identifications. TABLE OF CONTENTS

Abstract ii

Acknowledgements iii

Table of Contents iv

List of Figures vii

CHAPTER 1: INTRODUCTION 1

CHAPTER 2: THE SITE AND THE CANADIAN PLATEAU 3 Overview of the White Rock Springs Site (EeRj-226) 3 Chronology of the Canadian Plateau 7 Late Period 4500-200 BP (Plateau Pithouse Tradition) 7 Botanical Work on the Canadian Plateau 11 Direct and Indirect Plant Food Evidence 12 Ethnohistory 13 Ethnobotany 27 Archaeological Tools and Features 31 Plant Food Remains 44

CHAPTER 3: RESIDUE ANALYSIS: NEW INFORMATION FROM OLD TOOLS 51 Historical Overview of Lithic Residue Analysis with Emphasis on Starch 51 Artifact Residues 52 Starch as a Biological Subject 52 Starch as an Archaeological Residue 53 Residue Analysis as a Way Forward 55 Theoretical Considerations in Residue Analysis 59 Human Ecological Perspective 59 Women and Visibility. 60 Theory in Residue Analysis 60

iv CHAPTER 4: RESIDUES, ARCHAEOLOGY AND TAPHONOMY 66 Starch Biology 66 Starch Features Useful to Archaeology 69 Starch Look-alikes 73 Microscopic Indicators of Human Activity 74 Plant Micro-remains 75 Vascular Tissues 75 Phytoliths 78 Pollen 79 Fungal Micro-remains 80 Hyphae and Spores 80 Animal Micro-remains 81 Cells and Tissues 81 Feathers 81 Residue Taphonomy 82 Residue Preservation 83 Stone Tools 83 Soils 85 Special Contexts 87 Modified Starches 88 Contamination of Archaeological Material 91

CHAPTER 5: THE TOOLS AND PROCEDURES 94 Description of the Lithic Sample and Handling History 94 Procedures 96 Residue Extraction 96 Slide Preparation 99 Starch Grain Identification 99 Non Starch Residues 105

v CHAPTER 6: RESULTS 106 Plant Micro-remains 106 Starch 106 Overview and General Remarks 107 Starch Data Collected 107 Density of Starch by Tool 107 Morphotypes 113 Vascular Tissues 144 Softwoods 144 Hardwoods/Herbaceous Plants 147 Phytoliths 149 Pollen 154 Fungal Micro-remains 155 Hyphae and Spores 155 Animal Micro-remains 160 Feather Barbule 160 Variety and Density of Residues on Tools 162 Tool Form and Tasks 163 Absent Residues 167 Results of Controls 167 Summary of Results 170

CHAPTER 7: CONCLUSIONS 172 Goals Revisited 172 Implications 176 Recommendations 177 Pounding Tools and Future Research Potential 178 Contamination 178 Residue Extraction Procedure 181 Reference Collection 182 Limitations 183

REFERENCES 186

APPENDIX A: Residue Extraction and Slide Mounting Procedure 236

APPENDIX B: All Residues from Lithic Sample 242

APPENDIX C: Starch Morphotypes 256

vi List of Figures

Fig. 2-1. Diverse floral zones on the Canadian Plateau, near the town of Lillooet, British Columbia. Fig. 2-2. Les Miller holding a Chocolate Tips (Lomatium dissectum) plant with edible roots and shoots, found near excavations at the White Rock Springs site. Fig. 2-3. Example of a maul from British Columbia, held at the University of Calgary Archaeological Museum. Fig. 2-4. Trench through cultural feature at the White Rock Springs site showing high amounts of fire cracked rock and charcoal characteristic of earth ovens. Fig. 4-1. Light micrograph of storage starch grains from White Potato {Solatium tuberosum) under cross polarized light, 250x magnification. Note extinction crosses. Fig. 4-2. Different shapes of storage starch grains from White Potato, 250x magnification. The large grains display eccentrically placed hila and lamellae lines. Fig. 4-3. This large ruptured starch grain from water soaked macaroni pasta has been damaged, probably modified by physical milling, water and temperature changes, pH changes, enzymatic attack, or other methods of commercial processing. Tearing of the grain edge is apparent in normal light, 400x magnification. Fig. 4-4. Note the complete loss of birefringence and the extinction cross of the of the large ruptured grain when viewed in cross polarized light, 400x, image digitally lightened. Fig. 5-1. In this view and plane of focus, the hilum of Starch 15 (large grain in lower center) appears a solid dot. From Tool 1560, 400x. Fig. 5-2. When rotated, the hilum of Starch 15 appears an open ring form. From Tool 1560, 400x. Fig. 6-1. Morphotype A starch grain appears as a flat disc in polar view. Max length of grain is 28.5 |jm. Starch 1 from tool 1919, 400x. Fig. 6-2. Morphotype A starch grain from an oblique side view. When the grain is probed, the equatorial groove is visible. Starch 1 from tool 1919, 400x. Fig. 6-3. The extinction cross of Morphotype A grains usually become visible when probed into the oblique side view in cross polarized light. Starch 1 from tool 1919, 400x. Fig. 6-4. Morphotype A in polar view with notably roughened grain edges, max length 29.1 pm. Starch 2 from tool 3694, 400x. Fig. 6-5. Morphotype A in oblique side view with equatorial groove visible. Starch 2 from tool 3694, 400x. Fig. 6-6. Same grain in cross polarized light partial oblique view. Fig. 6-7. Morphotype A starch grain 3 from tool 3694 with the 'donut' or ringed area around the hilum, max length 28.5 pm, polar view, 400x.

vii Fig. 6-8. Morphotype A starch grain 3 from tool 3694 showing the 'donut' or ringed area around the hilum, slightly rotated, 400x. Fig. 6-9. Same grain showing the hilum area darkened in cross polarized light, 400x. Fig. 6-10. Starch grains in polar view of Bluestem Wheatgrass (Pascopyrum smithii) from the BC/AB reference collection, 400x. Image used with kind permission of B. Kooyman. Fig. 6-11. Same grains, note large grain rotated to show equatorial groove. Image by B. Kooyman. Fig. 6-12. Large Bluestem Wheatgrass starch grain rotated with equatorial groove, 400x. Image by B. Kooyman. Fig. 6-11. This Morphotype B grain is a 6.8 pm sphere with an open pink hilum. Starch 34 from tool 1790, 400x. Fig. 6-12. Same grain. Extinction cross is weak in cross polarized light, 400x. Fig. 6-13. Morphotype B grain is a slightly irregular sphere with a light red core in the open hilum and a dark red rim and is 5.7 pm. From tool 9991, 400x. Fig. 6-14. Same grain showing a weak extinction cross in cross polarized light, 400x. Fig. 6-15. Morphotype C oval grain with large pink depressed hilum area, 8.6 pm. Starch 1 from tool 3674, 400x. Fig. 6-16. (center, top) Morphotype C oval grain with depressed pink hilum area apparent, 8.8 pm. Starch 1 from tool 13575, 400x. Fig. 6-17. Morphotype C with large pink hilum area, 6 pm. Starch 11 from tool 9995, 400x. Fig. 6-18. Morphotype D starch 1 from tool 11356 is a bowl shaped bell with open hilum and rough surface texture, 14.3 pm, 400x. Fig. 6-19. Very large Morphotype D starch grain 12 from tool 9995, with a very rough, perhaps even cratered surface, 22.8 pm, 400x. Fig. 6-20. Huge Morphotype E reniform with lamellae, 25.9 pm. Starch 12 from tool 1560, 400x. Fig. 6-21. Huge Morphotype E reniform with lamellae, 25.97.1 pm. Starch 1 from tool 3824, 400x. Fig. 6-22. (center, top) Sphere shaped Morphotype F grain 6 with open pink hilum and rough edges, 7.1 pm. From tool 9995, 400x. Fig. 6-23. Morphotype F sphere shaped grain with rugged edges and open pink hilum. Starch 10 from tool 1560, 8.6 pm, 400x. Fig. 6-24. Same grain in cross polarized light, 400x. Fig. 6-25. Morphotype G polygon, open hilum with Y fissure, 20 pm. Starch 5 from tool 1790, 400x. Fig. 6-26. Morphotype G starch 5 from 1790, rotated with Y fissure still apparent, 400x. Fig. 6-27. Same grain in cross polarized light, 400x. Fig. 6-28. Morphotype G polygon with open hilum and rough edges, 20 pm. Starch 8 from tool 5364, 400x.

viii Fig. 6-29. Morphotype G from Starch 8 from tool 5364, rotated with edges now appearing smooth, 400x. Fig. 6-30. Same grain in cross polarized light, 400x. Fig. 6-31. Morphotype G Starch 1 from tool 10522, 17.1 pm, 400x. Fig. 6-32. Same grain in cross polarized light, 400x. Fig. 6-33. Morphotype G polygon,14.325.9 pm. Starch 8 from tool 1790, 400x. Fig. 6-34. Morphotype G polygon, rotated with Y fissure evident. Starch 8 from tool 1790, 400x. Fig. 6-35. Same grain in cross polarized light, 400x. Fig. 6-36. Commercial corn starch, 400x. Note polygons with open hila and grain with Y fissure in center. Fig. 6-37. Commercial corn flour, 400x. Note polygons with irregular bumpy edges and grain with deep Y fissure at top center. Fig. 6-38. Conifer tracheid wood fragment. Note large bordered pits. From tool 3694, 400x. Fig. 6-39. Tracheid of a softwood with large bordered pits. From tool 15182, 400x. Fig. 6-41. Tracheid of a conifer. Arrow points to a large bordered pit. From tool 15182, 400x. Fig. 6-42. Douglas Fir tracheid fragment displaying large bordered pits characteristic of conifers in cross polarized light. From tool 3651, 250x. Fig. 6-43. Close up of Douglas Fir tracheid in cross polarized light. Note diagnostic spiral thickening that is nearly perpendicular to the tracheid axis, 400x. Fig. 6-44. Conifer ray parenchyma cells on top overlapping two broken tracheids. Cross field pitting like this occurs in conifers, note bordered pits on top left of fragment. From microblade tool 13572, 400x. Fig. 6-45. The other end of the softwood fragment, also displaying large bordered pits, 400x. Fig. 6-46. Softwood fragment likely from Rocky Mountain Juniper (Juniperus scopulorum). Note cupressoid pit pairs in the cross fields of ray parenchyma cells and overlapping tracheids. From cobble tool 3674, 400x. Fig. 6-47. Alternate intervessel pitting on a woody or herbaceous eudicot vessel element. From flake tool 9995, 400x. Fig. 6-48. Same vessel element, with alternate intervessel pitting distinct in cross polarized light, 400x. Fig. 6-49. Vessel element with alternate intervessel pitting from tool 3830, 400x. Fig. 6-50. Vessel element with intervessel pitting from tool 3830,100x. Fig. 6-51. Same in cross cross polarized light. Note 'wavy lines' appearance, 100x. Fig. 6-52. Elongate trapeziform Pooid grass phytolith with 'corn cob' look. From tool 6368, 400x. Fig. 6-53. Roughened Panicoid bilobe phytolith. From tool 3677, 400x. Fig. 6-54. Triangular edge spine core phytolith from a trichome. From tool 17661, 400x. Fig. 6-55. Epidermal quadrilateral phytoliths from a leaf of a eudicot. From hand maul tool 3651, 400x.

ix Fig. 6-56. Epidermal tissue fragment with hexagonal shaped phytoliths. From hand maul tool 3651, 400x. Fig. 6-57. Interdigitating pitted epidermal phytoliths from steep-angled tool 16099, 250x. Fig. 6-58. Spruce (P/'cea sp.) bisaccate pollen grain Tool 3651, 400x. Fig. 6-59. Pine (Pinus sp.) bisaccate pollen grain Tool 3674, 400x. Fig. 6-60. Fungal hyphae end (on left) appears sippled. From tool 9984, 400x. Fig. 6-61. Brown fungal hypha with clamp connection, a feature of the Basidiomycota, and cross walls from tool 3651, 400x. Fig. 6-62. Fungal attack of plant material from tool 16099,100x. Fig. 6-63. Reproductive soredium of a lichen. Note green algal cells and surrounding hyphae that are hyaline, bubbly and smooth. From hand maul 3651, 400x. Fig. 6-64. Another lichen soredium fragment from the same tool, 400x. Fig. 6-65. Spore from acute-angled tool 3824, 400x. Fig. 6-66. Fungal spore from tool 9987, 400x. Fig. 6-67. Germinating fungal spore. Tool 3824, 400x. Fig. 6-68. Barbule from a feather of an Anseriform bird. Note triangular shaped nodes and lack of pigmentation. From acute-angled tool 13574, 250x. Fig. 6-69. Prong apparent on tapering distal end of the barbule, 400x. Fig. 6-70. Hard endosperm corn starch from back counter Control 2, 400x. Note polygon shape, open hilum and X fissure. Fig. 6-71. Hard endosperm corn starch from back counter Control 2, rotated, 400x. Fig. 6-72. Same in cross polarized light, 400x. Fig. 6-73. Blue thread from residue extraction Control 9989-9983, 400x. Fig. 6-74. Blue thread from residue extraction Control 11356-2417, 400x. Fig. 6-75. Hair from Control 2294-1951, 400x. Note cuticle and scales. Fig. 6-76. Hair from Control 2294-1951, 400x. Note scales. Fig. 6-77 Handmauls with residues and use wear evident on their working surfaces like this one from BC at the University of Calgary Archaeology Museum have good potential for residue analysis.

x 1

CHAPTER 1: INTRODUCTION

This broad spectrum residue analysis study is the first of its kind to be undertaken in the Canadian Plateau area of British Columbia and had a number of exploratory goals. These research goals were: 1) Learn what types of organic residues can be removed from ancient Canadian Plateau stone tools and viewed with light microscopy. 2) Describe, illustrate and identify the residues found. 3) Find direct evidence of plant foods. Investigate plant starch as a residue that might reveal information about which root foods were being processed at White Rock Springs. 4) Discuss where residues are likely to have originated. 5) Suggest how specific residues can be related to food and/or technological human use in the past.

To undertake these goals, the author selected a sample of 106 tools that represented all tool types in the lithic assemblage from the White Rock Springs site (EeRj-226) and extracted the residues. The site is regarded as a place where people gathered to collect root foods in the vicinity and prepare them by pit cooking in rock lined earth ovens. Direct evidence of botanical food processing at the White Rock Springs site is scarce and the use of plant foods in ancient diets is underrepresented archaeologically. This absence of visible plant food remains at what is assumed to be a plant processing site justifies the exploration of plants by microscopic means. Once residues were removed from the White Rock Springs tools, they were mounted on slides and viewed, photographed, and described. The microscopic residues discovered were identified wherever possible by consultation of literature, reference collection material, and advice from experienced researchers. Special attention was paid to plant starch grains, based on the assumption that earth oven sites were primarily used to cook root foods, and that some of these root foods may have left starch residues behind on processing tools. Starch was targeted since direct evidence of root remains is essentially absent in 2 the Canadian Plateau archaeological record. Fourteen characteristics were recorded for each of 151 starch grains encountered during microscopic examination, including size, shape, and surface features, among others (Appendix C). These characteristics were then used to group starch grains that shared similar morphology into Morphotype groups. Other microscopic remains such as phytoliths, fungal hyphae, and woody and herbaceous tissue fragments, were also photographed and recorded (Appendix B). All residues were assessed for their possible food, technological, and taphonomic importance. Laboratory and microscopic analysis was carried out in the Department of Archaeology at the University of Calgary, Alberta in 2010. 3

CHAPTER 2: THE SITE AND THE CANADIAN PLATEAU

Overview of the White Rock Springs Site (EeRj-226)

White Rock Springs was identified in 2004 as an earth oven site during ethnobotanical fieldwork being carried out by Sandra Peacock. The site is situated in Upper Hat Creek valley, an upland meadow between the Fraser and Thompson Plateaus in interior British Columbia. Survey and excavation of White Rock Springs has proceeded every year since 2005. The nearest modern communities to White Rock Springs are the town of Lillooet to the east and the villages of Ashcroft and Cache Creek to the west. The White Rock Springs site is named for its location at the base of a large towering white rock outcrop with a spring flowing from its base.

White Rock Springs is situated in the Upper Hat Creek valley, a place characterized by altitudinally diverse floral zones. Different habitats were visited by past peoples living in the area to obtain a variety of resources (Fig. 2-1). Hat Creek drains into the Bonaparte River, a tributary of the Thompson River. There are intermittent, seasonal and permanent stream tributaries that drain into Hat Creek and these probably reflect similar locations of water sources prehistorically. The Upper Hat Creek valley lies between the Fraser Plateau with the Clear Range to the west, and the Thompson Plateau, with the Trachyte and Cornwall Hills making up the eastern boundary (Pokotylo 1978:35-44). Climate in the Upper Hat Creek valley can be described as semi-arid due to the rain shadow effect of the Coast Mountains. The area has cold winters (3 to -11°C mean daily temperatures) with snowfall, and typically warm (12 to 15°C mean daily temperatures) but sometimes unpredictable summer weather. 4

Fig. 2-1. Diverse floral zones on the Canadian Plateau, near the town of Lillooet, British Columbia.

There are six biogeoclimatic zones in the Hat Creek watershed: Interior Douglas fir, Ponderosa Pine, Bunchgrass, Montane Spruce, Subalpine Englemann Spruce- Subalpine Fir, and Alpine Tundra (Pojar and Meidinger 1991:52; Peacock and Chambers 2003:2). The elevation of the site is about 4,000 feet (1,219 m). According to Alexander, the site falls in the Montane Parkland unit (1992a:76). The Montane Parkland unit is characterized as transitional between alpine meadows and subalpine forests; it is a mix of parkland meadows, clusters of stunted windblown trees, and very open stands of subalpine tree species. One can expect to find Whitebark Pine, Engelmann Spruce, Subalpine fir and Lodgepole Pine trees in this unit generally. Today at White Rock Springs, Pondarosa Pines, and Douglas fir trees are abundant, however there are no Lodgepole Pines. Mammals that can be found in the Upper Hat Creek valley include: shrews, bats, voles, mice, hares, squirrels, pikas, porcupines, beavers, gophers, marmots, deer, sheep, black and grizzly bears, wolves, coyotes, foxes, martens, fishers, weasels, minks, wolverines, badgers, skunks, cougars, bobcats and lynxes. Today the small lakes and streams 5 outside of Hat Creek proper do not support any fish, but trout, Dolly Varden, whitefish and minnows are found in Hat Creek (Pokotylo 1978:64-66). Some of the birds that can be found in the White Rock Spring site area include grouse, raptors, woodpeckers, and many perching birds, with the small lakes and ponds in the Hat Creek valley providing the best waterfowl habitat in the Lillooet area (Mathewes 1978:85). Each Hat Creek valley locality has its own mix and abundance of plant food species. Edible geophytes in the Upper Hat Creek valley were collected beginning in the spring and included: Balsam root (Balsamorhiza sagittata), Wild Onion (Allium spp.), Indian or Mountain Potato (Claytonia lanceolata), Chocolate Tips or Wild Celery (Lomatium dissectum, see Fig. 2-2), Desert Parsley or Wild Carrot (Lomatium macrocarpum and other Lomatium spp.), Silverweed (Potentilla anserina), Tiger Lily (Lilium columbianum), Mariposa Lily (Calochortus macrocarpum), and Yellow Avalanche Lily (Erythronium grandiflorum) (Hunn et al. 1998:526; Matthew 1986a:2; Peacock and Chambers 2003:4; Wyatt 1998:193). Other botanical foods in the area are Whitebark Pine nuts, Saskatoon berries, Soapberries and other berries (Kuhnlein and Turner 1991; Turner 1981, 1988).

Fig. 2-2. Les Miller holding a Chocolate Tips (Lomatium dissectum) plant with edible roots and shoots, found near excavations at the White Rock Springs site. 6

Sandra Peacock (Assistant Professor, Dept. of Anthropology, UBC Okanagan) and Brian Kooyman (Professor, Dept. of Archaeology, University of Calgary) are principal investigators of the White Rock Springs site. Other members of the research team include David Pokotylo (Professor, Dept. of Anthropology and Sociology, UBC Vancouver) and Richard Hebda (Curator, Botany and Earth History, Royal British Columbia Museum, Victoria). The research program is focused on the archaeobotany of hunter-gatherers during the Late Prehistoric Period (4500-200 BP) in the Upper Hat Creek valley, specifically examining the relationship between climate change and root processing and earth ovens. The main goals of this project are to "a) define the processes of root resource intensification in the mid-Fraser River region and b) explore the relationship between these processes and the changing climates of the Late Prehistoric Period (4500-200)" (Peacock 2007a: 11). Project analyses being carried out include: macrobotanical material (wood charcoal, needles, and seeds by Sandra Peacock, Monica Nicolaides, Sheena McKenzie, and Crystal Weins), microbotanical remains (phytoliths by Brian Kooyman and starch by Kooyman and Kooyman, in preparation), lithic remains (Pokotylo), faunal remains (McGill et al. 2010, Eddy 2011, and ongoing B. Kooyman), paleoecology (palynology by Brintnell and Hebda) and radiocarbon dating. To date, 28 earth ovens and two activity areas (tentatively classed as 'campsites') have been identified at White Rock Springs. Several of the ovens have been dated showing occupation between 2000 and 230 years ago. In addition to investigations at White Rock Springs, the project is excavating in another root processing locale, known as Houth Meadows, consisting of five sites (EeRj-55, EeRj-56, EeRj-57, EeRj-58, EeRj-159) and collectively representing 19 earth ovens. There are currently over 300 archaeological sites in the Hat Creek valley watershed with evidence for the use of earth ovens, commencing 3300 years ago during the Shuswap Horizon and continuing up to the historic period (Peacock 2007a:10; Peacock and Chambers 2003:5). 7

Chronology of the Canadian Plateau

The prehistory of the Canadian Plateau is organized into three major cultural historical units, based on evidence of changing lithic technologies, settlement types, major food resources, and climate. These archaeological time units include: Early 12,000/11,000 BP to 7000 BP, Middle 7000 to 3500 BP, and Late 3500 to 200 BP Periods (Fladmark 1982; Rousseau 1993:140; Stryd and Rousseau 1996:177-226). The Middle and Late Periods as used recently appear to be now set at 7000 to about 4500 BP and 4500 to 200 BP, respectively (Nicholas and Westfall, in press; Peacock 2007b:3; Peacock and Chambers 2003:1). The Late Period has the most complete archaeological information. The dates from White Rock Springs fall within the Late Period and this is the time frame of most interest, as the stone tools examined in this study represent about the last 2,000 years.

Late Period 4500-200 BP (Plateau Pithouse Tradition) It has been argued that climate change was the driving force that made Canadian Plateau people more sedentary during the transition from the Middle to the Late Period (Kuijt 1989). Around 4500 BP, modern grasslands became established and cooling to modern climatic conditions occurred (Hebda 1995). This cooling trend made winters longer and snows deeper on the Canadian Plateau. This lessened peoples ability to hunt during the winter and made taking shelter in semi- subterranean habitations more attractive (Rousseau 2004:11). This reduction of mobility and use of house pits has been noted across the Plateau around the 5000 BP mark (Chatters and Pokotylo 1998:75). Stryd and Rousseau (1996) refined the Plateau Pithouse Tradition chronology set out originally by Richards and Rousseau (1987) to contain four cultural divisions: the Lochnore Phase ca. 5000-3500 BP, the Shuswap Horizon ca. 3500-2400 BP, the Plateau Horizon ca. 2400-1200 BP, and lastly the Kamloops Horizon ca. 1200-200 BP. The Plateau Pithouse Tradition sequences all fall within the Late Period, as do the dates from White Rock Springs. By the Shuswap Horizon (3500 BP), Late Period peoples were less mobile than previously, and a degree of semi-sedentism in house pit villages occurred. People 8 also took part in planned seasonal harvests at specific locations, with salmon contributing increasingly to the diet (Rousseau 2004:13). Good evidence for the intensive utilization of plant foods exists for the Late Period, with the number of earth ovens increasing beginning 2400 BP (Plateau Horizon) (Lepofsky and Peacock 2004:135). The widespread appearance of house pits during the Late Period is often associated with storage facilities, demonstrating that food surpluses were gathered, processed and stored for future consumption (Kuijt 1989:102).

The Lochnore Phase began around 5000 years ago and ended at 3500 BP. Early Lochnore sites are ephemeral campsites with lithics, faunal remains and hearth features (Prentiss and Kuijt 2004:50). The Lochnore side-notched point is diagnostic of this phase (see photographs in Stryd and Rousseau 1996:194, and Rousseau 2004:9) and bone and antler technology was well-developed. The lithic material most commonly used for tools was non-vitreous basalt (Stryd and Rousseau 1996:193). People in the Lochnore Phase had a generalized broad spectrum diet, however the emphasis was on terrestrial game. They hunted deer, elk, beaver, porcupine, and birds. They fished for salmon and other fish, collected fresh water mussels, and some kept domestic dogs. Greater coastal cultural influences became apparent on the Canadian Plateau at this time (Rousseau 2004:11), signaling a closer relationship between plateau and coastal peoples. For example, shell beads from sea snails (Olivella sp.) and limpets from the coast are present in Lochnore Phase sites (Stryd and Rousseau 1996:193).

The Shuswap Horizon (3500-2400 BP) commenced with moister and cooler climatic conditions, reflected in a documented decrease in the frequency of forest fires in south west British Columbia generally (Hallett et al. 2003:303). House pits of various sizes appear during the Shuswap. One of the earliest examples of house pits from the Canadian Plateau comes from the Tezil site (FgSd-1). The Tezil site has 46 depressions and three styles of dwellings: semi-subterranean structures with a floor surrounded by a raised bench, saucer shaped depressions, and raised earthen rings (Donahue 1977:119-121). A date of 3850 ± 140 BP was assayed on a 9 charcoal sample from a raised bench area above a Tezil house pit floor (Donahue 1977:154). Shuswap projectile points are characterized as being variable in form but usually stemmed. The Shuswap lithic technology is described as 'crude' when compared to later horizons, and local low to medium quality lithic raw materials were generally used for all tool types (Richards and Rousseau 1987:25-27). Leaf- shaped and lanceolate bifaces, key-shaped unifaces and bifaces, thumbnail scrapers, hide scrapers, and split cobble tools are all common. Bone and antler working is present during this period with antler wedges, bone beads, points, harpoon heads and awls being included in the tool types. Most of the very large and medium-sized housepits at the Keatley Creek site (EeRI-7) appear to have been initially constructed during the Shuswap Horizon (3500-2400 BP) and to have been abandoned around 1100 BP (Hayden 1997:246). According to Peacock (2007a: 10), and Peacock and Chambers (2003:5), there is evidence for the use of earth ovens in the Hat Creek valley watershed from 3300 years ago which persists up to the historic period, although no radiocarbon dates are cited. Plant resources and roots specifically were eaten during this period. There is evidence for trade between the Canadian Plateau and southern Nothwest Coast peoples in the form of high quality nephrite stone being exported to the coast and Dentalium shell to inland groups. Foods of primary importance during the Shuswap Horizon include terrestrial mammals of all sizes, birds, fresh water mussels, salmon, trout, and other fresh water fish species (Richards and Rousseau 1987:29).

Larger villages of more than 10 house pits appear in the Plateau Horizon (2400- 1200 BP) and are usually near optimal fishing locations (Prentiss et al. 2005a:71). At the Bridge River site (EeRI-4), the number of simultaneously occupied house pits jumped from 7 to 29 between ca.1800 and 1250 cal. BP (Prentiss et al. 2008:59), and Prentiss and her colleagues have argued for large villages as a late development occurring after 2000 BP. In contrast, Hayden asserts that large villages and house pits are at least as old as 2600 BP or earlier (2005b:169). Richards and Rousseau (1987:32) state that house pits become smaller during the this horizon, but this was also challenged by Hayden (2000a:23), who argued for 10

the continued occupation and even expansion of large house pits at the Keatley Creek site during the Plateau Horizon. Controversy still exists as to the timing of climate change and the nature of the development of large villages, wealth-based inequalities, demographic changes, and village abandonment (see Hayden 1997, 2000b, 2000c, 2005a, 2005b; Hayden and Mathewes 2009; Kuijt 2001; Kuijt and Prentiss 2004; Morin et al. 2008; Prentiss 2009, 2011; Prentiss et al. 2003, 2005b, 2007, 2008, 2011). These issues are certainly not yet resolved and new research on the Canadian Plateau will aim to refine chronological accuracy and reassess established models of cultural change. For example, ongoing house pit settlement research in the mid-Fraser region by Sheppard and Muir (2010) will likely lead to further revision of our understanding of the relationship between changes in the sizes and densities of house pits, and time scales.

Earth ovens and storage pits are evident in the Plateau Horizon. The older dates of earth ovens from the Upper Hat Creek (2245 BP) (Beirne and Pokotylo 1979:Part1, 1:5), White Rock Springs (2000 BP), and Keatley Creek (1580 BP) (Hayden and Cousins 2004) fall within this time frame. The Plateau Horizon lithic industry is characterized by an increase in chipped stone endscrapers, with unifacial and bifacial flake tools still dominating assemblages. The calibre of stone knapping is higher than during the Shuswap Horizon, which may be partly explained by predominant use of high quality raw lithic materials during the Plateau Horizon (Richards and Rousseau 1987:34). More bone and antler artifacts appear in the Plateau Horizon, including tooth pendants, animal incisor tools, bone beads and/or gaming pieces, and composite harpoon heads.

The Kamloops Horizon (1200-200 BP) is the best represented horizon of the Late Period archaeologically. In the Kamloops Horizon, subsistence remained focused on seasonally available resources and winter house pits continued to be used. Excavated Kamloops Horizon house pits usually have a thin roof insulation layer, but some have heavy wood and earth covered roof deposits. Storage pits and central hearths are commonly encountered in house pits (Richards and Rousseau 11

1987:43). The small Kamloops side-notched point appears and is common and diagnostic of the horizon. The dominance of these small points in Kamloops assemblages is interpreted as evidence of heavy use of bow and arrow technology (Hayden 2000a:25). The variety and frequency of ground stone artifacts and bone, antler and tooth artifacts are higher when compared with the preceding Plateau Horizon. Utilitarian items became more decorated (Rousseau 2004:19) and skilled artists carved zoomorphic and anthromorphic designs from steatite (Richards and Rousseau 1987:45). Dogs were used for transporting food surpluses and were sometimes eaten. Interregional trade networks between the Pacific Coast and the Canadian Plateau continue to be significant. For example, abalone armbands, Olivella and Dentalium shell bead necklaces were found in a triple burial at the Talus Burial Site (EqQw-1), estimated to be 150 years old (Hills 1971:27). According to stable carbon isotope analysis by Chisholm, marine protein accounted for 40 to 60 (±10%) of the dietary protein based on sampled bone collagen (1986:122). There is a decline in the number (but not the size) of earth ovens at upland locations between 1500 and 800 BP, and presumably upland root resource use also declined (Lepofsky and Peacock 2004:136). Although further sampling is required, it appears that the number of earth ovens increased after 800 BP, but became smaller in size (Lepofsky and Peacock 2004:136).

Botanical Work on the Canadian Plateau

In the archaeology of the Canadian Plateau, the plants that formed a major component of past diets have not been as well studied as the faunal contributions. There has been a biased picture of Canadian Plateau subsistence patterns, with an overemphasis on the salmon fishery to the exclusion of plant and other animal resources (eg. Kuijt 1989). This is especially true in the Late Period (ca. 4500-200 BP) characterized by large housepit villages, also known as the Plateau Pithouse Tradition (Ames and Marshall 1980; Lepofsky 2004:368; Peacock 2002:44; Pokotylo and Froese 1983:127,152-153). Research that deals with the contribution that plant resources made to Plateau diet is scarce which has, in effect, diminished 12 the apparent importance of women's roles in traditional economies (Hunn 1981). This meat and male dietary bias is not surprising given that paleobotanical studies on the Canadian Plateau were rarely incorporated into excavation programs (Lepofsky 2004:367,439; Lepofsky and Peacock 2004:130). Furthermore, this bias is a common problem of hunter-gatherer studies generally (Hather and Mason 2002:1). Recently, a reversal of this imbalance has been underway due to the work of ethnobotanical and paleobotanical specialists such as Nancy Turner, Sandra Peacock and Dana Lepofsky. It is now clear that plant resources figured prominently in the diets of peoples on the Canadian Plateau, with root food carbohydrates (including fleshy bulbs, corms, tubers, rhizomes, and taproots) making a critical contribution. However, direct evidence of root food remains has been exceedingly rare archaeologically, which is why this study turns to microscopic residue techniques in order to identify the use of plant foods. The major carbohydrate of most tubers and roots is starch, comprising 16-24% of their total weight (Hoover 2001:253), making starch residues a sensible target of research when trying to 'see' geophyte use in the past. However, not all geophytes store carbohydrates as starch. For example the Balsamroot (Balsamorhiza sagittata) stores sugars as the complex carbohydrate inulin. Thus, root foods that store sugars as inulin may not be easily detected by residue analysis.

Direct and Indirect Plant Food Evidence In the following sections I will discuss the information available on botanical food resources of the Canadian Plateau. To accomplish this, I have followed Lepofsky's (2004:368) and Lepofsky and Peacock's (2004:127) definitions of direct and indirect evidence, with the addition of ethnographic accounts of food plant use as indirect evidence. Direct evidence of food plants is defined here as the edible or inedible remains of food plants found in secure archaeological contexts. Examples of direct evidence include pollens, seeds, cones, pericarp fragments, charred parenchymous tissue, dried fruits, plant food residues found on cooking vessels and tools, micro or macrobotanical remains from sediments, stomach contents, coprolites and dental calculus. Indirect evidence is any material exclusive of the 13 plant remains themselves that can inform us about plants in ancient diets. Indirect evidence is often found to affirm and give context to direct archaeological evidence of plant food remains. Oral histories, early and modern ethnographies, biogeographic information, theories on demographic shifts and resource availability, palynological and other climate/vegetation records are pieces of indirect evidence that can support the direct evidence of plant food remains. Plant food processing tools and features such as digging sticks, sap scrapers, mortars and pestles, earth ovens, roasting pits or mounds, and cache pits are also important sources of indirect evidence of plant foods in the diet.

Ethnohistory For archaeologists, ethnographies are sources that enrich and inform the interpretations we build about what life was like thousands of years ago. Of course it is necessary to accept that no matter how accurate or reliable the ethnographic literature is, it will remain an analogy and its applicability to the time period in question is unknown until supported by the archaeological record (Wylie 1985:83). Indeed, even at the time of historical documentation, "Every living community is in the process of continuous change with respect to the material which it utilizes," and will not remain the same (Ascher 1961:324). Thus, uniformitarian cultural assumptions must be avoided, as the fit between actual earlier lifeways and historical descriptions cannot be expected to be perfect. Few would deny that the strongest foundation for interpretations is found where different lines of evidence converge. Indirect evidence of the processing and consumption of food plants from early ethnographic accounts is the first area that will be explored here.

The oldest in-depth specific information available about Canadian Plateau plant foods is found in the ethnographies of the late nineteenth and early twentieth centuries, carried out by American anthropologist James A. Teit. He married a Thompson (Nlaka'pamux) woman and lived with the people for forty years, thus acquiring special access and close knowledge of their customs and lifeways (Wickwire 1993:539). In his ethnographies of the Thompson (Nlaka'pamux), 14

Lillooet (Sta'atl'imx), Shuswap (Secwepemc) and Okanagan, Teit documented which plant food species were consumed and the way in which many were collected, prepared and cooked (Teit 1900:231-239, 1906:222-224, 1909:513-518, 1930:237-240; Steedman 1930:477-492;). He also included drawings of some of the tools used in plant collecting and processing such as root digging sticks, bark peelers, sap scrapers, birch bark baskets, and a berry drying frame. Good ethnographic information about Canadian Plateau people was also collected by Charles Hill-Tout on the Thompson, Okanagan, and Lillooet in the late 1800s and early 1900s (Hill-Tout 1978a, 1978b), George M. Dawson on the Shuswap (1891) and Verne F. Ray (various groups 1939, 1942). The similarities between how the Thompson, Lillooet, Shuswap and Okanagan utilized plant foods are so striking they are easily compared.

Canadian Plateau First Nations ethnographies recorded ways in which plant foods were collected and prepared. Underground geophytes (bulbs, stems, tubers, rhizomes and corms) were dug up using digging sticks made of antler or wood (commonly from Saskatoon Berry) or a composite of these. Digging sticks for uprooting geophytes had handles made mostly of birch wood, goat and sheep horns, and the antlers of elk and deer (Teit 1900:231-232, 1906:223, 1909:514, 1930:240; Ray 1942:145; see also Hunn et al. 1998:527 for a photograph of a digging stick collected by Teit before 1903 whose shaft is made of Saskatoon Berry wood and handle of birch wood). The women placed the unearthed roots in birch bark and cedar root baskets carried on the body and head, which were periodically unloaded into larger baskets (Teit 1900:231). Some types of roots, such as Spring Beauty (Claytonia lanceolata) and Mariposa Lily (Calochortus macrocarpus), were simply prepared by dessication. The roots were threaded on a bark or grass string and hung up to dry (Steedman 1930:477; Teit 1900:235) or spread out on mats to dry (Teit 1909:516). The ethnohistoric record details the use of earth ovens (also called roasting pits) to cook a variety of geophytes (Dawson 1891:21; Ray 1942:137-139; Steedman 1930:477; Teit 1906:223, 1909:516, 1930:240), with large flat heated stones acting as heating elements (Teit 1900:236). Using earth 15 ovens with large rock heating elements is an efficient way to cook large quantities of roots and/or meats, as well as prepare food items such as cetain geophytes that require prolonged heating times to convert inulin to digestible simple sugars (Peacock 1998:315; Wandsnider 1997:29). This method of cooking saves labour because a large volume of food can be prepared in one firing event in an earth oven. The oven can be re-used, which means no additional effort is required to dig a new pit and find appropriate heating stones for each cooking event. Since it is documented ethnographically that earth ovens were used for roasting meats such as bear, deer and 'ground hog' as well as plants (Ray 1942:137), earth ovens can only be inferred as indirect evidence of root roasting activity. George Dawson saw the Shuswap transform raw root foods sealed inside bark baskets into a type of flour after earth oven baking (1891:21). Some roots were threaded and dried on strings or cooked inside earth ovens with layers of vegetation matting (1900:235,236, 1906:223, 1909:516). Some roots were boiled with hot rocks in cedar root or birch bark containers (Steedman 1930:477). Birch bark containers and cedar root woven baskets were used to collect and cook roots as well as berries (Teit 1900:235, 1906:216, 1909:517). Some cooked roots were pounded in mortars, made into cakes and dried (Teit 1930:240). Prepared roots were stored in sacks and baskets, in or near habitations or inside cache pits dug into the ground (Steedman 1930:477).

According to Teit (Teit 1906:256; Steedman 1930:478), root digging grounds were considered common property among the Lillooet, Thompson, and other tribes, although there is some indication that some patches were privately owned but with an expectation that the 'crops' would be shared. One rich root gathering area called "Many Roots," or "Wealthy-in-Roots" was shared by the Lillooet, Chilcotin (Tsilhqot'in) and Shuswap (Teit 1906:256). Root digging, berry picking, and fishing places were inhabited temporarily by groups in certain seasons (Hill-Tout 1978a: 131). These impermanent camps were also set up to exploit and prepare geophytes. Earth oven construction involved digging a pit, as well as collecting appropriate heating rocks and insulating vegetation and wood. Roots were 16 harvested by parties of women, with children helping alongside. Women may have spent days collecting suitable green vegetation with which to enclose and protect the roots to be cooked (Dawson 1891:20). Men were also probably involved to some extent in preparing root foods at temporary camps, perhaps cutting down pine trees to fuel the earth oven fires (Dawson 1891:20). However, men were excluded from the actual cooking of the geophytes (Dawson 1982:19; Steedman 1930:509; citations within Alexander 1992b: 127). Locales where root resources were concentrated were seasonally-visited by task groups. For example, the Botanie valley (near present day Lytton), was a place where the Thompson, Shuswap, Lillooet and other nations met every summer; the women dug roots and collected berries, and men hunted (Davidson 1915:48; Ommer and Turner 2004:134; Turner and Loewen 1998:52). For example, ethnographically known bitter root (Lewisia rediviva) gathering grounds at Three Sisters valley between Spences Bridge and Ashcroft (in the vicinity of the White Rock Springs site) were visited by groups of Thompson women (Davidson 1916:135). Bitter root processing was described thus by Davidson (1916:136):

From an examination of several Indian camping-places it was seen that the women evidently collect a supply of roots and return to camp to strip them. One can picture half a dozen Indian women, squatted before the camp-fire at the close of a day's digging, busy peeling the roots preparatory to packing them. The small heaps of skins [cortical tissue] left at most of the camps indicated that each party returned with many hundreds-perhaps thousands-of roots.

Similarly, when speaking of the collection of camas (Camassia quamash) in Northwestern North America, Brown (1868:379) noted:

The gathering is nearly wholly done by women and children, who use a sharp- pointed stick for the purpose; and it is surprising to see the aptitude with which the root is dug out. A botanist, who has attempted the same feat with his spade, will appreciate their skill. About this period the Indians come from their permanent village, and encamp under the shade of trees in little brush camps. 17

Roots and root digging are also culturally embedded in myth or stories, taboos and girls' puberty training. For instance, Teit recorded a Thompson story in which a woman who lived in Lytton was taken away by a chief (perhaps the Sun), but left roots in Botani for her people to dig and eat (Teit 1917:21). Another Thompson tradition describes how a man named Tsuntia, or Child-of-Hog-Fennel, became the moon, replacing Coyote who was 'being' the moon. The story involves a maiden who, while digging hog-fennel roots (Desert Parsley, Lomatium macrocarpum), found and cohabited with one of the roots and gave birth to a son, Child-of-Hog- Fennel. The child eventually becomes the moon (Steedman 1930:510; Teit 1912:224-226). Traditional stories may reflect socially enforced plant food restrictions or taboos that existed in past societies. For example, the Thompson had specific protocols concerning the gathering and processing of Balsamroot (Balamorhiza sagittata). Teit reports women must abstain from sexual intercourse while digging or cooking the roots and men are not allowed near the earth oven where the Balsamroot is being baked (Steedman 1930:509). Young people were expected to address a prayer to the Balsamroot whenever eating berries or roots for the first time that season, and failing this the person would not be able to get up in the morning: "I inform thee that I intend to eat thee. Mayest thou always help me to ascend, so that I may always be able to reach the tops of mountains, and may I never be clumsy! I ask this from thee, balsam root. Thou art the greatest of all in mystery" (Steedman 1930:509). It it reasonable to speculate that these controls on peoples actions associated with Balsamroot may have been reinforced by stories as a mode of cultural transmission. In another tradition, the roots of Fool's Onion (Brodiaea douglasii) saved a family from starvation. In yet another, Coyote was in the sky country and dug up Spring Beauty (Claytonia lanceolata); he discovered these roots were stars (Steedman 1930:511). During the isolation puberty rituals, young Lillooet women prayed to be lucky in finding roots, and to be skilled at digging them quickly and with endurance (Teit 1906:264-265). The Thompson and Shuswap practiced similar rituals, with girls obliged to dig trenches with the hope that this 'practice' would improve stamina for root digging and other challenging work (Teit 1900:312-313; 1909:587). Wright believes that these intense and 18 monotonous tasks that girls carried out were designed to gain spiritual power (2003:7).

Task groups of women and children collected berries by hand into woven and bark baskets, as well as birch bark trays (Brown 1868:384; Teit 1909:515; 1930:222,240). Small blueberries were removed from the bush by passing the teeth of wooden combs through the branches (Teit 1909:515). A variety of fruits, including Saskatoon Berry, Soapberry, wild cherries (species unspecified, probably Prunus pensylvanica, and/or P. emarginata), huckleberries, raspberries, brambleberries and rose hips were dried in the sun on mats with no further preparation for storage (Teit 1900:235; 1930:240). Berries were often preserved in dried cakes (Hill-Tout 1978a:56; Teit 1900:235; 1906:222), similar in concept to dried fruit leather or bars today. One way of making the cakes was by first baking the fresh berries, then boiling them with hot rocks in bark baskets. Once cooked, the berries were mashed with a stick, wooden or stone pestle, or by hand. Then they were laid out in the sun to dehydrate on one of the following: rush, tule, grass or willow twig mats; layers of leaves, dry grass, dry or fresh pine needles, sticks or a wooden frame/scaffold (Ray 1942:139,143; Teit 1900:235; 1930:218,240). After being dried out for a period of time, the processed berries were formed into cakes (Teit 1900:235). Berries were also mashed up with meat using ground stone pestles on large flat stones. Dried berries or cakes were stored in birch bark baskets. Although there is only one species of Amelanchier alnifolia (Saskatoon Berry) recognized by Western scientific nomenclature, the Thompson ascribe different names to six types of the berry (Steedman 1930:488). Like root digging areas, for the Lillooet berrying patches were common property and anyone could pick berries, as long as they were ripe and the berry picking season had been ceremonially opened by the clan chief responsible for that particular patch(es) (Hill -Tout 1978a: 133-134; Teit 1906:256). One berry that the Thompson identified as the little-coyote berry or northwestern serviceberry, was commonly talked of in traditions as coyote food (Steedman 1930:510). 19

To collect tree cambium, a bark-peeler made of antler, wood and horn was used to separate the bark from the tree. Then a sap-scraper made of antler, bear scapula, deer ulna, or animal rib was used to scrape the cambium layer (Ray 1942:131-132; Teit 1900:233; 1909:515-516). The cambium layer from Lodgepole Pine (Pinus contorta), Whitebark Pine (Pinus albicaulis), Ponderosa Pine (Pinus ponderosa), spruce (Picea spp.), Grand Fir (Aibes grandis), poplar (Populus spp.), Trembling Aspen (Populus tremuloides), Douglas Fir (Pseudotsuga menziesii), Choke Cherry (Prunus virginiana), and alder (Alnus spp.) were used. Sap from wild cherry (species unspecified, likely Prunus pensylvanica, and/or P. emarginata) was also eaten (Ray 1942:131; Teit 1900:233; 1906:222; 1909:515; 1930:238-239). The cambium was collected with the sap scraper using an upward motion on the tree and put into baskets, according to informants from the Lillooet, Shuswap and Thompson (Ray 1942:132). Cambium could be eaten fresh or was dried and stored for future use by the Shuswap and Thompson (Ray 1942:132), with sap from the Ponderosa pine noted specifically to be prepared for winter use (Teit 1900:233).

Canadian Plateau people harvested hazelnuts wherever available, as well as cones containing nuts (Teit 1906:222; 1909:515; 1930:237). They also sought out and took nuts from squirrel and mouse caches (Teit 1900:231). Whitebark Pine nuts were a favourite among the upper Thompson (Teit 1900:233), but were also collected by the Shuswap (Teit 1909:515). They roasted the nuts in ovens or ashes and then stored them away in sacks for winter (Teit 1900:233). Ponderosa Pine seeds were also used by some groups of the Shuswap (1909:515). Balsamroot seeds, sometimes referred to simply as 'sunflower' in old ethnographies, were eaten (Ray 1942:132; Teit 1900:233; 1930:237). Balsamroot seeds were boiled using hot stones in baskets with or without deer fat (Teit 1900:237), then pounded in mortar bags with stone pestles to a mealy flour (Teit 1930:204). The Okanagan sometimes crushed seeds in mortar bags made of rawhide (Teit 1930:221), and the Lillooet used flat rawhide plaques to pound pine nuts (Ray 1942:143). 20

Green vegetables eaten included the young stalks of Cow Parsnip, Fireweed, Balsamroot, salmon berry, thimble berry, and wild celery (Teit 1900:233; 1906:222; 1909:515; 1930:238-239). Mushrooms were peeled and eaten fresh or roasted (Teit 1900:233; 1909:515). Cactus and black tree lichen was also baked or steamed in earth ovens (Teit 1900:233,237; 1909:515).

The following is a list of plant species that were recorded in Teit's ethnographies of the Thompson, Lillooet, Shuswap and Okanagan (1900; 1906; 1909; 1930, respectively). I have organized the categories by food type, not botanical taxonomy, to group foods that were used or prepared in similar ways. Species are listed by modern binomial name, followed by common names, then species name as documented by Teit, if different from the modern name. Plant species in each food type group are numbered consecutively.

Roots: bulbs, tubers, norms, rhizomes Apiaceae (Carrot Family) 6. Lomatium dissectum (Nuttall) Mathias & Constance (Chocolate Tips) = Leptotaenia dissecta Nuttall 7. Osmorhiza berteroi de Candolle (Sweet Cicely) = Osmorhiza nuda Torrey 8. Lomatium macrocarpum (Nuttall) Coulter & Rose (Desert Parsley) = Peucedanum macrocarpum Nuttall 9. Sium suave Walter (Water Parsnip) = Sium laeve Walter, Sium lineare Asteraceae (Sunflower Family) ~\0.Balsamorrhiza sagittata Nuttall (Balsamroot) 11 .Cirsium edule Nuttall (Edible Thistle) '\2.Cirsium hookerianum Nuttall (Hooker's Thistle) 13.Cirsium undulatum (Nuttall) Sprengel (Wavy-leaved Thistle) = Cnicus undulatus Gray Boraginaceae (Borage Family) \4.Lithospermum incisum Lehm. (Long-flowered Stoneseed, Narrowleaved Stoneseed) = Lithospermum angustifolium Michaux 21

Cyperaceae (Sedge Family) 15. Scirpus sp. (propbably Scirpus lacustris ssp. glaucus, aka Scirpus acutus. The thick fleshy rootstocks of one Scirpus sp. was reported as being baked and eaten Steedman 1930:481) (Bulrush, Tule) Hydrophyilaceae (Waterleaf Family) 16. Hydrophyllum capitatum Douglas ex Bentham (Ball-head Waterleaf) = H. capitatum was probably mistakenly referred to as Hydrophyllum occidentale Gray by Teit, since H. occidentale does not occur in B.C.; the plant is native to the western United States from California to Idaho) Lamiaceae (Mint Family) 17. Lycopus uniflorus Michaux (Bugleweed) = Cycopus uniflorus Liliaceae (Lily Family) 18. Allium spp. (Onion) 19. Allium acuminatum Hooker (Hooker's Onion) 20. Allium cernuum Roth (Nodding Onion) 21. No modern name, (Autumn onion, Prairie onion) Allium stellatum Ker. (probably a wild onion species that was eaten, A. stellatum not present in guide books for interior B.C. or E-Flora BC) 22. Triteleia grandiflora Lindley (Fool's Onion, Large-flowered Triteleia) = Brodiaea douglasii Wats. 23. Triteleia hyacinthina (Lindley) Greene, also Brodiaea hyacinthina (Lindley) Baker (White Brodiaea, White Triteleia, Fool's Onion) = Brodiaea grandiflora Smith 24. Calochortus macrocarpus Douglas (Mariposa Lily) = called Calechortus macrocarpus (Lavender Lily) by Teit 25. Camassia quamash (Pursh) Greene (Blue Camas) = Quamasia quamash (Pursh) Coville, also Camassia esculenta (noted by Teit as confined to the southern parts of the country, Columbia River valley) 22

26. Erythronium grandiflorum Pursh (Yellow Avalanche Lily, Glacier Lily, Dog-tooth Violet) = Erythronium grandiflorum Pursh var. minor, Erythronium grandiflorum parviflorum Wats. 27. Fritillaria lanceolata Pursh (Chocolate Lily) 28. Fritillaria pudica (Pursh) Sprengel (Yellowbell) 29. Lilium columbianum Hanson ex Baker (Tiger Lily, Columbia Lily) = Lilium parviflorum (Hooker) Holzinger (Panther Lily) Portulacaceae (Pursulane Family) 30.Claytonia sp., 31. Lewisia columbiana (Howell ex Gray) Robinson (Columbian Bitterroot) 32. Claytonia lanceolata Pursh (Spring Beauty) = Claytonia sessilifolia, Claytonia lanceolata Pursh 33. Lewisia pygmaea (A. Gray) Robinson (Pygmy Bitterroot) 34. Lewisia rediviva Pursh (Bitterroot) Rosaceae (Rose Family) 35. Potentilla sp. 36. Potentilla anserina Linnaeus (Common Silverweed) Pteridophyta (Fern Division) 37. Pteridium aquilinum (Linnaeus) Kuhn (Common Bracken, Bracken Fern) = Pteris aquilina Linnaeus var. lanuginosa (Bory) Hooker Typhaceae (Cat-tail Family) 38. Typha latifolia Linnaeus (Cat-tail)

N.B.: Two species not included in ethnographic documentation by Teit, but noted as regularly pit cooked plants by Peacock (2002:47), are Sagittaria latifolia Wildenow (Wapato) and Perideridia gairdneri (Hooker & Arnott) Mathias (Gairdner's Yampah).

Fruits: berries, currants, cherries Adoxaceae (formerly part of the Caprifoliaceae Honeysuckle Family) 1. Sambucus spp., (Elderberry) 23

2. Sambucus cerulea Rafinesque (Blue Elderberry) or S. racemosa Linnaeus (Red Elderberry), misidentified as Sambucus canadensis Linnaeus (American Elderberry) by Teit, since S. canadensis is not found in B.C.) Berberidaceae (Barberry Family) 3. Mahonia spp. Nuttall (Oregon Grape) = Berberis sp. Caprifoliaceae (Honeysuckle Family) 4. Lonicera involucrata (Richardson) Banks ex Sprengel (Twinflower Honeysuckle, Black Twinberry) = Lonicera involucrata Banks 5. Viburnum sp. (Cranberry) 6. Viburnum edule (Michaux) Rafinesque (High-bush Cranberry, Mooseberry, Squashberry) = Viburnum pauciflorum Pylaie Cornaceae (Dogwood Family) 7. Cornus sp. (Bitter berry) 8. Cornus sericea Linnaeus (Red Osier Dogwood) = Cornus pubenscens Nuttall Elaeagnaceae (Oleaster Family) 9. Shepherdia canadensis Nuttall (Soapberry) Ericaceae (Heath Family) 10. Arctostaphylos uva-ursi Sprengel (Bearberry) 11. Gaultheria shallon Pursh (Salal berry, Salal) 12. Vaccinium sp. (huckleberries) 13. Vaccinium myrtillus Linnaeus (Dwarf Bilberry, Low Bilberry, Whortleberry) = called Vaccinium myrtillus Linnaeus, var. microphyllum Hooker (Blueberry) by Teit 14. Vaccinium membranaceum Douglas ex Torrence (Mountain Bilberry, Black Blueberry, Black Huckleberry, Thinleaf Huckleberry) Grossulariaceae (Currant Family, Gooseberry Family) 15. Ribes sp. (Gooseberry) 16. Ribes lacustre (Persoon) Poiret (Swamp Gooseberry, Black Gooseberry, Prickly Currant) 17. Ribes cereum Douglas (Waxy Currant, Wax Currant) 24

18. Ribes hudsonianum Richardson (Northern Black Currant) Rosaceae (Rose Family) 19. Amelanchier alnifolia Nuttall (Saskatoon Berry) 20. Probably Crataegus columbiana Howell (Red Hawthorn) or C. douglasii Lindley (Black Hawthorn) = Teit names Crataegus rivalaris Nuttall (hawberry), although this species is not present in B.C. 21. Probably Fragaria vesca Linnaeus (Tall Strawberry) or F. virginiana Duchesne (Blue-leaf Strawberry)= Teit identified a strawberry as Fragaria californica Chamisso and Schlecht, but F. californica is not found in B.C.) 22. Prunus virginiana Linnaeus (Choke Cherry) = Prunus demissa Walpers 23. Rubus sp. (Raspberry, Thimbleberry, Blackberry) 24. Rubus lecodermis Douglas ex Torrey & Gray (Blackcap) 25. Rubus parviflorus Nuttall (Salmonberry) = Rubus nutkanus Mocino 26. Rosa gymnocarpa Nuttall (Dwarf Wild Rose, Baldhip Rose) 27. Probably Sorbus sitchensis Roemer (Sitka Mountain-ash, Western Mountain Ash) = Sorbus sambucifoiia (Chamisso and Schlechtendal) Roemer

Stems and stalks Apiaceae or Umbelliferae (Carrot or Celery Family) 1. Heracleum lanatum Michaux (Cow Parsnip, Wild Rhubarb) 2. Peucedanum spp. = Lomatium spp. 3. Lomatium nudicaule (Pursh) Coult. & Rose (Indian Celery, Barestem Lomatium, Barestem Biscuitroot, Barestem Desert-parsley) = Peucedanum leiocarpum Nuttall (Wild Celery, Indian Consumption Plant) Asteraceae (Sunflower Family) 4. Balsamorrhiza sagittata (Pursh) Nuttall (Balsamroot) Cactaceae (Cactus Family) 5. Opuntia sp. (Prickly-pear Cactus) 6. Probably Opuntia fragilis (Nuttall) Haworth (Brittle Prickly-Pear), listed by Teit as Opuntia polyacantha Haworth (Plains Prickly-pear) 25

Onagraceae (Evening Primrose family) 7. Epilobium angustifolium Linnaeus (Fireweed) Rosaceae (Rose Family) 8. Rubus sp. 9. Rubus leucodermis Douglas ex Torrey & Gray (Blackcap, Blackcap Raspberry, Black Raspberry, Dwarf Bramble, Whitebark Raspberry) = Rubus occidentalis Linnaeus probably is R. leucodermis since R. occidentalis only occurs in Eastern North America 10. Rubus spectabilis Pursh (Salmonberry)

Flowers 1. Calochortus macrocarpus Douglas (Mariposa Lily)

Cambium Betulaceae (Birch Family) 1. Alnus rubra Bongard (Red Alder) Pinaceae (Pine Family) 2. Abies grandis (Douglas ex Don) Lindley (Grand Fir) 3. Picea sp. (Spruce) 4. Pinus albicaulis Engelmann (Whitebark Pine) 5. Pinus contorta Douglas ex Louden (Lodgepole Pine, Shore Pine)= Pinus murrayana Balfour 6. Pinus ponderosa Douglas (Ponderosa Pine) 7. Pseusotsuga menziesii (Mirbel) Franco (Douglas fir) = Pseudotsuga mucronata (Rafinesque) Sudworth, Pseudotsuga douglasii Carr Salicaceae (Family) 8. Populus tremuloides Michaux (Tembling Apsen)

Douglas-fir sugar Pinaceae (Pine Family) 1. Pseusotsuga menziesii (Mirbel) Franco (Douglas fir) 26

Funai and lichen Agaricaceae 1. Agaricus sp. Parmeliaceae 2. Bryoria fremontii (Tuckerman) Brodo & Hawksworth (Black Tree Lichen, 'Black Moss', Edible Horsehair) = Alectoria jubata Linnaeus, Alectoria jubata Acharius

Seeds and nuts Asteraceae (Sunflower Family) 1. Balsamorrhiza sagittata Nuttall Betulaceae (Birch Family) 2. Corylus cornuta Marshall (Beaked Hazelnut) Pinaceae (Pine Family) 3. Pinus albicaulis Engelmann (Whitebark Pine) 4. Pinus ponderosa Douglas (Ponderosa Pine) Santalaceae (Sandalwood Family) 5. Comandra umbellata (Linnaeus) Nuttall (Bastard Toadflax) = Comandra pallida

Favouring Asteraceae (Sunflower Family) 1. Balsamorrhiza sagittata Nuttall seeds Rosaceae (Rose Family) 2. Probably Fragaria vesca Linnaeus (Tall or Wild Strawberry) or F. virginiana Duchesne (Blue-leaf Strawberry)= Teit identified Fragaria californica Chamisso and Schlecht, but F. californica is not found in B.C.) Scrophulariaceae (Figwort Family) 3. Probably Pentstemon fruticosus (Shrubby Penstemon), which was used in to flavour roots during pit-cooking = Pentstemon menziesii Hooker flowers 27

Ethnobotany Canadian Plateau ethnobotanical studies are critical to understanding what plant foods were processed at White Rock Springs. A familiarity with past harvesting, preparation and cooking techniques, as well as plant management practices, from the statements of elders in modern times is vital to bring the past to life and relevancy. The botanical knowledge of elders can inform archaeological expectations. This ethnobotanical information is helpful to the archaeologist because it enumerates plant species that can be particularly targeted for identification and analysis, and provides geographical information about where plant gathering and processing sites are likely to be located. For example, in a history of the Upper Lillooet, Trefor Smith documented accounts from elders such as Edith O'Donaghey of the Shalalth (Tsal'alh) of the Lillooet (St'at'imc) (1998:10). O'Donaghey recounted how the mountains were visited to gather berries, geophytes, lichen, and nuts, and to hunt game:

[In mid August] a whole bunch [of families] would go up ... on top of the mountain, way up high, close to the snow line ... by horseback every year... for a few days, three or four days or a week, mainly to pick huckleberries ... We got huckleberries and dug roots while we were up there, wild potatoes, wild onions, whatever we could find up there on the mountain. They [the men] would hunt too ... I haven't been up there since I was young, but that used to be a tradition in the olden days ... They'd gather a whole bunch of tiger-lily bulbs,... that black moss ... we'd bring down six or eight baskets with pack horses ..., baskets and sacks of mountain potatoes ... Well [in late August and September], there were the pine cones up there too. My dad used to bring home sacks full from his hunting trips up in the mountains ... Granny used to keep a special pan just for roasting them [pine cone seeds] in the oven ... My dad used to go hunting with about six or eight other men and boys ... Sometimes they were gone for a week, or ten days or more.

Although the description above is just one account of who, when, how, and what foods were gathered at upland sites, it provides a good baseline scenario that helps us imagine how White Rock Springs could have been utilized.

Nancy Turner is undoubtedly the leading expert on the ethnobotany of First Nations in British Columbia. She has provided the most comprehensive current work of 28 plants used as food, medicine and technology, past and present, by indigenous peoples of the Canadian Plateau. Through interviews with elders, Turner has collected and documented the old stories and modern ways in which plant species were, and are, used and managed (Szczawinski and Turner 1980; Turner 1992, 1997, 1998, 1999; Turner et al. 1980, 1990, 2000). Extensive accounts of the ethnobotany of the Lillooet (Turner 1992), Okanagan-Colville (Turner et al. 1980), Shuswap (Palmer 1975) and Thompson (Turner et al. 1990) have been published. I will not review the plant species eaten ethnographically in the Canadian Plateau region, as this has been adequately documented in the sources listed above. The main focus of this study is the identification of direct archaeological evidence of microscopic botanical remains and to document past plant use.

Ethnobotany has drawn attention to the dramatic loss of First Nations plant knowledge in British Columbia and in Canada on a large scale. Ethnobotanical studies have illustrated that First Nations posess a holistic knowledge and understanding of the natural environment which is integrated with practices and beliefs that have been passed down through generations. This collection of environmental knowledge has been variously termed as Traditional Ecological Knowledge (TEK) (Berkes 1999), Indigenous Ecological Knowledge (Berkes 1993), Traditional Ecological Knowledge and Wisdom (Turner et al. 2000), and Traditional Phenological Knowledge (Lantz and Turner 2003). TEK has been highlighted as having the potential to make meaningful contributions to environmental conservation efforts and to promotion of biodiversity (Berkes and Turner 2006; Huntington 2000; Peacock and Turner 2000; Usher 2000). A related objective of some TEK work has been to increase awareness of the plants and animals used for food, technology, and medicine, and revitalize such knowledge so that it may contribute to the physical and cultural health of First Nations communities (eg. Turner and Turner 2007). Turner has examined the importance of preserving and promoting women's plant knowledge specifically, and notes the value this knowledge will provide in future resource management and land restoration plans (Turner 2003; Turner and Turner 2008). 29

The nutritional value of traditional foods is a topic being explored in ethnobotany and paleoethnobotany today. Nutritional value has been posited both as a way to promote reintroduction of traditional foods into modern diets and a way to assess the food choices people would have made in the past. It is generally assumed that foods that were high in energy and calories would have been valued and ranked higher than those which did not contribute significantly to energy needs. However, this logic does not hold in times of food stress and famine, when many 'foods' that are potentially toxic and harmful to the body were consumed in desperation in order to survive (Turner and Davis 1993). Ignored in such a narrow view is the fact that dietary needs are not just in terms of calories, but also in essential vitamins, minerals, and amino acids that are critical nutritional components. Another problem with studies that assess nutritional value of traditional foods is that values of a food plant that are collected from different locations or micro-habitats may not be comparable.

Mullin et al. compared the energy of Balsamroot (Balsamorhiza sagittata) taproots and Yellow Glacier Lily (Erythronium graniflorum) bulbs in raw and earth oven cooked states (1997). They found that these two food plants provided about the same amount of energy in both samples that were fresh and earth oven cooked for 20 hours at temperatures between 65 and 95°C (364 to 395 kcal per 100 g dry basis) (1997:772, 774). Balsamroot specifically was 369 (peeled) and 370 (unpeeled) kcal per 100g dry basis for fresh samples, and 364 kcal per 100g dry basis for earth oven cooked peeled and unpeeled samples (1997:774). Peacock collected very interesting data from experimental cooking of Balsamroot in earth ovens under high (~58-99°C) and low (~46-78°C) temperature conditions for 20 and 21 hours, respectively (2008). The earth oven that had a higher temperature yielded Balsamroot samples with the highest caloric energy; 224.18 (peeled sample) and 233.86 (unpeeled) kcal per 100 g dry basis. The earth oven with the lower temperature yielded roots with values of 169.96 (peeled) and 163.66 (unpeeled) kcal per 100 g dry basis, and fresh raw Balsamroot samples had energy values of 168.75 (peeled) and 112.45 (unpeeled) kcal per 100 g dry basis 30

(2008:125). The differences between Balsamroot energy values reported by Mullin et al. (1997) and Peacock (2008) (differences of 139.82 kcal between earth oven cooked, peeled samples and 130.14 kcal between earth oven cooked, unpeeled samples) were surprising given that samples treated in the same way did not produce similar results. Both studies tested earth oven baking for 20 hours in the 'high' temperature range (within 58-95°C), both froze, then freeze-dried the samples, both tested peeled and unpeeled Balsamroot, and both carbohydrate analyses were conducted by the same person, John Mullin. I suggest that these differences in Balsamroot caloric value may be indicative of differences in plant soil chemistry, plants dug at different stages of development or plants dug at different times of year.

The Cottonwood Mushroom (Tricholoma populinum) was eaten by the Thompson, Lillooet, Okanagan and Shuswap. Turner et al. (1987:926) reported a low energy value of raw peeled Cottonwood Mushrooms, only 21-28 kcal per 100 g fresh weight. However, the authors considered the Cottonwood Mushroom nutritionally valuable because all the amino acids are present in mushrooms (except perhaps sulphur-containing amino acids), and small but important amounts of iron, copper and zinc are present. Fresh and frozen samples of Bitterroot (Lewisia redivia) yielded 90 and 94 kcal per 100 g dry basis, respectively (Benson et al. 1973:145).

Dering (1999) has argued that despite the high archaeological visibility of earth ovens and associated remains of lechuguilla (Agave lechuguilla) and sotol (Dasyliriort texanum) from archaic period sites in the Lower Pecos River region of Texas/Mexico, these plants have a relatively low food value and could not have contributed as significantly to the diet as has been previously modeled. He experimentally earth oven cooked lechugilla and sotol for a minimum of 36 hours, then each plant was pounded into separate cakes and dried. The average caloric value on a dry basis for lechugilla was 319 kcal per 100 g and for sotol was 343 kcal per 100 g (Dering 1999:664). Along similar lines, Smith et al. (2001) found Sego Lily bulbs (Calochortus nuttalli) experimentally cooked in earth ovens on the 31

Plains in the Big Horn Basin, Wyoming, returned low energy values of 96.5 kcal per 100 g. These authors see these values as especially low considering the digging time and effort required to build earth ovens. Smith et al. suggest that the Sego Lily was a very low ranked resource and would not have been added to the diets of prehistoric foragers unless absolutely necessary. I find Smith et al.'s conclusion that earth oven cooked Sego Lily has a low caloric value objectionable given that this value is similar to the common white potato Solanum tuberosum, which has 99 kcal per 100 g when baked in skin, flesh only (Woolfe 1987:34). White potatoes were and are staple foods in many places, and would not be a preferred food unless they provided a contribution to caloric needs. In general, the caloric value of the baked potato is lower than to that of earth oven cooked wild geophytes such as Balsamroot and Yellow Glacier Lily available in the interior of British Columbia by 66 to 300 kcal. High carbohydrate foods such as rice, corn, wheat flour, and beans were not available to prehistoric First Nations of the Plateau region, and thus people choose geophytes that were the most palatable and provided the highest nutritional value. Teit noted that potatoes, flour, rice, beans, and others were carbohydrate staples in the diet of many Shuswap bands by 1905 (1909:517).

Archaeological Tools and Features Material culture indicative of plant harvesting, processing and storage has been excavated at archaeological sites on the Canadian Plateau. I divide the indirect evidence of the consumption and storage of plant foods into two categories: tools and features. Tools include digging sticks, baskets, boiling stones, pestles and hammerstones, plant grinding stones, sap scrapers and drying racks. Features include earth ovens and caches pits. I have classed Harlan Ingersoll Smith's excavations published in Archaeology of Lytton, British Columbia 1899, and his Archaeology of the Thompson River Region, British Columbia 1900 works as archaeological evidence, albeit not to modern standards.

Digging sticks, according to Rousseau (2004), appear at the start of the Plateau Horizon (2400-1200 BP). Hayden and Schulting (1997:60,64) provide a list of 32 antler digging stick handles from midden, burial and housepit contexts that have been found at nine archaeological sites on the Canadian Plateau. On the Canadian Plateau, digging stick handles have been found at: Chase burial site (EeQw-1) (Sanger 1968a); Texas Creek burial site (EdRk-1) (Sanger 1968b:10); Mitchell Site (EeRI-22) (Stryd 1972:27; 1973:69); sites in the Lytton area (Smith 1899:137-138); site in the Thompson River region (Smith 1900:411); in a burial context with charred bones at the Government site (Smith 1900:436); and in a grave at the head of Nicola Lake (Smith 1900:439). Wooden digging sticks and handles would have likely decomposed long before they could be recovered in archaeological excavations.

Baskets were used to harvest, cook and store berries, roots and nuts. Baskets were probably used for multiple purposes, with other functions such as water collection and burial inclusions. Therefore, a basket can only be associated with food plants with some certainty when food plant remains are found adhering to or inside the vessel. Discoveries of baskets, particularly those made of birch bark, have been made at many Canadian Plateau archaeological sites in a variety of contexts. Birch bark containers have been reported archaeologically from a number of sites (see examples in: Beirne and Pokotylo 1979 Section 3:74-75, 93; Carlson 1980:94; Pokotylo et al. 1987:4,5; Prentiss et al. 2005b; Sanger 1968a: 124,179, 1970:19,101; Smith 1900:437; Stryd 1970:7, 1985:78; Stryd and Hills 1972:199, 205-206; Richards and Rousseau 1987:36,45;). Birch bark containers appear to have been commonly used in the past but what contents the above containers might have held remains a mystery, as no reports of floral remains from the sediments of these baskets were made. Croft and Mathewes (n.d.) have shown that sediments from excavated baskets should be examined for evidence of plant and animal foods. They found charred and uncharred seeds of Saskatoon berry {Amelanchier alnifolia), Goosefoot (Chenopodium spp.), Blackcap Raspberry (Rubus cf. leucodermis) and two unknown seeds, as well as fragments of fish bone, charcoal and grass, in the sediments of a birch bark container from the Ollie site (EeRk-9) (Blake 1974:35). Ketcheson reported that a birch bark basket that 33 contained salmon bones was excavated from an earth oven at EeRj-55A (Beirne and Pokotylo 1979 Section3:74-75; Ketcheson 1979:23). She suggested that the paucity of roots found in earth ovens may be due to past peoples placing roots in baskets or woven mat vessels that were removed from the ovens (1979:41), leaving little chance for spillage and loss. A fragment of a coiled basket was found at the Keatley Creek site (Wittke et al. 2004), however no associated botanical contents were reported.

Pottery and wooden boxes are unknown from the Plateau of pre-contact British Columbia (Smith 1913:38; Conn and Schlick 1998:600), and the main cooking vessels relied on were probably birch bark and Western redcedar root vessels used with heated boiling stones. Boiling stones are difficult to distinguish from earth oven stones, both of which are fire altered from heating. Piles of fire cracked rock have been found in the American Southwest that authors suggest indicate animal and plant processing and other activities (Sullivan et al. 2001), and Thorns (2008a: 445) suggests that at Plateau sites where above ground containers were used with boiling stones, the boiling stones might be found in scatters and concentrations. Perhaps small heaps of boiling stones showing heat alteration could be a contextual indication of use of stones for cooking in vessels on the Canadian Plateau. Oven stone residues, including amino acids (constituents of proteins) and fatty acids (constituents of lipids), have been explored by liquid and gas chromatography (eg. Fankhauser 1993; Quigg et al. 2001). Oven stones thus are an artifact class that clearly would profit from analysis of residues. It is reasonable to assume that trace amounts of the foods that were cooked with boiling stones would adhere to the rock surfaces. Not only does residue analysis of boiling stones have the potential to discover the type of plants that were eaten, but also may be a good way to find modified starch that can indicate cooking techniques.

Ground stone pestles, also known as handmauls, pounders, or pestle shaped hammers, recovered from archaeological sites could be indicators of plant food processing and/or woodworking (Fig. 2-3). Preliminary work by Croft found that 34 starch does preserve on handmaul artifacts from British Columbia that have been held in museum collections (2009). There is a good body of ethnographic evidence that hand mauls, pestles and mortars of stone (but also sometimes wood) were used to crush and grind root foods, berries, seeds, nuts and perhaps also the tree lichen Bryoria fremontii in British Columbia (Brennan 1975:122; Keddie 1988; Smith 1899:138, 1900:415; Stewart 1973:52; Teit 1930:240).

In addition to pulverizing plant foods, excavated pestles could have been used for other tasks such as wood working, grinding up plant or mineral dyes, and macerating meats, making their function(s) impossible to interpret without microscopic examination of use wear and/or residue analysis. Harlan I. Smith noted that in the Thompson River region of interior B.C., "Stone pestles served for crushing dried meat, berries, etc., as well as for driving wedges, splitting wood, and in like industries. Many of these pestles are mere cylinders of tough rock, often but slightly changed from the natural pebble by a little pecking or rubbing," (1900:412). Pestles were found to be commonplace in Shuswap territory along the Fraser and Thompson Rivers (Dawson 1891:19). Pestles have been found at the Lehman site (EdRk-8) (Sanger 1970:75), the Chase burial site (EeQw-1) (Sanger 1968a), Kamloops Reserve site (EeRb-3) (Wilson 1973:6), sites in the Lytton area (Smith 1899:138-139), and in the Nicola valley (Smith 1900:412-413). Hammerstones or hand mauls were found at the Fountain site (EeRI-19) (Stryd 1972:19), Texas Creek burial site (EdRk-1) (Sanger 1968b:6), Lehman site (EdRk-8) (Sanger 1970:75), Cache Creek site (EfRh-3), Kamloops Reserve site (EeRb-3 and EeRb-10) (Wilson 1973:2,6,11; 1980:65), Guichon Slough site (EbRc-6) (Wyatt 1971:61), Honest Paul site (EcRh-1) (Wyatt 1971:63), Curr site (EdRa-22) (Carlson 1980:111), and Van Male site (EeRb-10) (Wilson 1980:65). A small stone mortar was found in the burial of a male at Brocklehurst Burial site (EeRc-8) (Wilson 1973:9), and stone mortars have been found in sites near Kamloops (Smith 1900:413-414). 35

Fig. 2-3. Example of a maul from British Columbia, held at the University of Calgary Archaeological Museum.

Mortars, ground stone bowls, anvils, flat stone slabs, and boulders with depressions are also possible indicators of plant food processing, but they could have been used for many functions in the past. However, there is mention of their use specifically for processing nuts and berries in ethnohistoric sources. With reference to the Thompson, cones of Whitebark Pine (Pinus albicaulis) were collected and the "...seeds were parched and pounded into a fine flour in a mortar" (Steedman 1930:491). Often, large flat grinding stones were used to mix berries and animal fat together with a pestle (Steedman 1930:486). Sandstone grinding slabs likely for processing seeds and berries were excavated at the Fountain site (Stryd 1972:39; 1973:69). Stone slabs showing use wear, as well as mortars or anvils, were found on the surface of village sites in Lytton (Smith 1899:139).

Although there is no direct archaeological evidence of cambium being consumed, ethnographies state that bone, horn and wood tools specifically for harvesting edible inner bark and sap from trees were used (Ignace 1998:208; Teit 1900:233; 36

1906:223; 1909:516,780). Tools used to scrape up tree cambium certainly could leave resinous residues that trap starch grains and other microscopic plant particles, so this may be a good tool type to explore plant foods via residue analysis in the future. Sap scrapers have been recovered from the Chase burial site (EeQw-1) (Sanger 1968a), sites in the Lillooet area (Stryd 1973:69), and three examples made of bone were recovered from a grave at a large burial at Kamloops (Smith 1900:411,412).

Drying mats and racks were used to prepare berries for winter storage (Teit 1900:235; 1909:516). At the bottom of a cache pit at EeRb-140, Nicholas excavated "numerous long and thin wood fragments, arranged in a mat-like fashion, ... this could possibly represent a portion of a berry cake-drying frame" (Nicholas and Westfall, in press).

From the surface, earth ovens appear as cultural depressions on the landscape and have been found in areas upland of deeply incised river valley floors, among other areas. Circular pits in the ground have been found to be numerous and highly visible during archaeological survey, but their function(s) are difficult to determine without excavation due to surface dimensions that fall within the range of three possible types of cultural features: earth ovens, cache pits and house pits (Beirne and Pokotylo 1979; Fladmark 1986; Peacock 1998, 2002; Pokotylo and Froese 1983). Thus, earth ovens can only be securely identified by shovel testing or excavation and the presence of fire cracked rock and sediments blackened by charcoal that would not be expected in either cache pits or house pits (see Fig. 2-4). Earth ovens that fit ethnographic descriptions and were presumably used to bake, steam or roast roots have been found at numerous sites, with over 300 earth ovens sites in the Hat Creek watershed alone. It is also probable that some meats were cooked in the same fashion (Hayden and Cousins 2004). Fig. 2-4. Trench through cultural feature at the White Rock Springs site showing high amounts of fire cracked rock and charcoal characteristic of earth ovens.

The earliest archaeological investigation to excavate earth ovens was conducted by Beirne and Pokotylo for the Hat Creek Project (1979). Pokotylo and Froese later published an article on the cultural features and associated surface lithic scatters in the upland sites of Upper Hat Creek valley (1983). The study involved survey, measuring and excavation of eleven earth ovens and highlighted the need for Canadian Plateau archaeologists to research 'small' non-housepit sites that documented non-salmon subsistence activities. This was seen as necessary to capture a wider and more representative picture of resource exploitation activities. Radiocarbon dates obtained from eleven earth ovens in this study ranged between 2245 ± 50 and 600 ± 40 BP. They showed that some earth ovens contained multiple rock lined basins dated hundreds of years apart, providing evidence that they were reused. Another issue pointed out by Beirne and Pokotylo (1979: Part 1, 3:69) and Pokotyo and Froese (1983:151) was that the diameter measurements 38 from excavated cooking pits overlapped with other cultural depressions, namely housepits and cache pits. This meant that earth ovens, housepits and cache pits could not be differentiated from each other based on surficial dimensions alone; excavation or at least shovel testing was required. This point was also illuminated by Hayden and Cousins, who found that only eight out of fifteen suspected geophyte roasting pits or earth ovens at the Keatley Creek site were confirmed by excavations of the pits (2004:147). Pokotylo conducted a study of 44 lithic debitage assemblages from sites in the Upper Hat Creek valley. He found that those assemblages associated with earth ovens were different from the other lithic scatter sites in that they represented a wide range of tool manufacturing activities, supporting the idea of intensive use, and likely prolonged stays, at earth oven sites (1981:387, 391).

In 1985, Alexander and Matson undertook an archaeological survey and surficial artifact collection of sites in the Potato Mountain Range and adjoining Eagle Lake area, located on the western edge of the interior Plateau and representing part of Chilcotin traditional territory (1987). They identified three summer root gathering mountain sites (Middle Mountain site (EjSb-52), Mountain Fan site (EjSb-39) and Mountain Pond site (EjSb-54)). Earth ovens that were test excavated were found to contain moderate to high quantities of fire cracked rock and charcoal, with no artifacts. Six Late Period earth ovens from P2-3 (EjSb-26), P2-9 (EjSb-33), Echo Ridge site (Ejsb-12), and Mountain Fan site, provided radiocarbon dates ranging from 100 ± 60 to 1910 ± 50 (Alexander and Matson 1987:95). Intense use of the alpine/subalpine ecotones to collect and process Claytonia lanceolata (Spring Beauty, Indian Potato or Mountain Potato) was suggested, although no macrobotanical remains were found (Alexander and Matson 1987:96).

Rousseau and Howe surveyed (but did not shovel test or excavate) and tentatively identified food roasting ovens at 26 sites on the Reserve at Scheidam Flats, near Kamloops (1987:8). They noted that most of the sites with cultural depressions they found were near creeks and suggested these were favourable locations to set 39 up seasonal root gathering camps. Earth ovens may have been constructed near fresh water for reasons of cleaning the roots or game before and/or after cooking, providing water for steam inside the oven, and to generally wash up after the very dirty task of constructing and opening the earth ovens. Rousseau and Howe propose that Balsamroot was likely the main target of harvest and processing, since it is prolific in the Scheidam Flats area today.

Sandra Peacock's 1998 dissertation brought much needed attention to the important role of edible geophytes in ancient Canadian Plateau diets. Peacock argued that earth ovens that contained remains of wood charcoal were the indirect evidence of baking and steaming of fleshy root foods. She noted that these root foods tend to not leave identifiable resistant parenchymous tissues behind, at least not macroscopically. The ethnographic descriptions of earth oven construction do indeed appear to match traits of the excavated archaeological cultural features, with their heating elements of fire cracked rock, dark, carbon rich deposits and charcoal fragments (Peacock 2002:51). Earth ovens logically represent the cooking stage of food procurement, but Peacock considers them in a wider context. She connects the baking of geophytes in earth ovens with a suite of related activities: collecting, cultivating, processing and storing. Peacock also views the use of earth ovens as a technological strategy past hunter-gatherers used to intensify plant food production. For Peacock the driving force behind the changes in intensity of use of earth ovens through time is climate change. She hypothesizes that earth ovens were more intensively used during cold climatic periods when storage of food over long winters was necessary for survival. She further developed these ideas in her 2002 chapter entitled "Perusing the pits: the evidence for prehistoric root resource processing on the Canadian Plateau." In this work, she makes clear her perspective that earth ovens are evidence of the transition from foraging to wild plant food production (Peacock 2002:59).

Stahl (1989) argued that various methods of plant processing such as milling, soaking, fermentation and heat treatment can increase nutritive value of plant 40 foods, and Peacock supports this perspective. Peacock emphasized that processing geophytes by pit cooking puts underground storage organs (some of which are otherwise inedible) on the menu as food resources. Not only are the roots made palatable by earth oven baking or steaming, but cooking also increases digestibility and bioavailability of nutrients for absorption in the gastrointestinal tract. Peacock has demonstrated that cooking geophytes in earth ovens increases their palatability and the nutritional value humans can obtain from these foods (2008). Native or 'raw' starch present in geophytes is poorly digested in the human alimentary tract. Cooking disrupts starch granule integrity, allowing amylolytic enzymes of the digestive system to more easily attack (Galliard 1987:3-4). Inulin, the storage carbohydrate found in some geophytes, cannot be digested by stomach acids, nor by enzymes in the small intestine, and can only be degraded (selectively fermented) by indigenous bacteria in the large intestine (Belitz et al. 2004:334; Gibson and Delzenne 2008:55). Thus without heat cooking treatment, inulin is not a digestible source of energy for humans.

Eight tested earth ovens, ranging in age from 30 ± 50 to 1,580 ± 60, from Keatley Creek yielded 1 Lomatium sp. skin, 1 Allium sp. bulb, 2 unidentified bulbs, 4 unidentified root skin fragments, 5 Ericaceae seeds, and 6 unidentified seeds (Hayden and Cousins 2004:147). Based on wood charcoal analysis, species found in earth ovens at White Rock Springs include Douglas Fir, Cottonwood, Ponderosa Pine, Engelmann Spruce, Choke Cherry, Birch, Alder and Willow (McKenzie et al. 2010; Peacock 2007b:2). Some of these species were probably used to heat earth oven stone elements, and some as matting to protect and provide moisture to foods being cooked.

Macrobotanical remains from flotation samples of the basin and toss zones of three earth ovens (CF 14, 15 and 18) from White Rock Springs were identified by Nicolaides (2010:94). Wood that was used as fuel included: Picea glauca (x engelmannii) (Spruce), Pinus sp. (Pine), Pseudotsuga menziesii (Douglas Fir), cf. Alnus sp. (Alder), cf. Betula sp. (Birch), Populus sp. (Trembling Aspen/ 41

Cottonwood), and cf. Salix sp. (Willow) (Nicolaides 2010:114). It is possible that different fuel types may have been selected for combustion properties suitable for the species of roots being cooked in the oven. For example, since certain types of wood burn to a higher temperature, different fuel species would have had an impact on how much and for how long heat was retained by the earth oven rock elements. Based on ethnographic accounts, Nicolaides suggests that dry dead Douglas Fir wood might have been preferred to newly felled green wood for use as earth oven fuel (2010:113). Materials that represent matting in the three White Rock Springs earth ovens include the needles of: Picea glauca (x engelmannii) (Spruce), Pinus ponderosa (Ponderosa Pine), Pinus contorta (Lodgepole Pine), and Pseudotsuga menziesii (Douglas Fir); twigs and branches of Picea glauca (x engelmannii) (Spruce), Pinus sp. (Pine), Pseudotsuga menziesii (Douglas fir), Populus sp. (Trembling Aspen/Cottonwood), and cf. Shepherdia canadensis (Soapberry) (Nicolaides 2010:94). A number of seeds from shrubs, wildflowers, grasses and sedges were also found by Nicolaides from the three earth ovens, including: Arctostaphylos uva-ursi (Kinnikinnick), cf. Berberis aquifolium (Oregon Grape), cf. Rosaceae (Rose), Rubus sp. (Raspberry/Thimbleberry), cf. Shepherdia canadensis (Soapberry), Asteraceae cf. Achillea sp. (Yarrow), Chenopodium cf. capitatum (Chenopod), cf. Fabaceae (Pea), cf. Polygonaceae (Buckwheat), cf. Ranunculaceae (Buttercup), Cyperaceae (Sedges), and Poaceae (Grasses) (2010:94). It is interesting that so many seeds were found in the earth ovens, but the interpretation that is most logical is that the seeds were probably unintentional inclusions from the branches and stems of earth oven matting (Nicolaides 2010:117). Nicolaides also suggested that green vegetation used to protect roots was also present in the earth ovens (Nicolaides 2010:119), although the species that would have been used, such as Western Skunk Cabbage (Lysichiton americanus) do not carbonize or preserve macroscopically. Unfortunately, no root remains came to surface in Nicolaides macrobotanical analysis (2010:118-119). There is potential for earth oven stones and sediments to harbour starch residues on their surfaces, particularly resistant heat modified starches. Horrocks has successfully collected sweet potato (Ipomoea batatas) and taro (Colocasia 42 esculenta) starch from a sediment core taken from a Maori earth oven site called Rangihoua Bay in New Zealand (Horrocks et al. 2004b).

It appears that meats were cooked in earth ovens in addition to plant foods, as animal remains have been excavated from earth ovens. Awareness of the evidence for animals being cooked in earth ovens aids in figuring out the dominant use of earth ovens. Cooked faunal remains were reported by Ham (1979) from test excavations from 15 cultural depression features in the Upper Hat Creek valley. Remains included Mule Deer, Elk, a species of grouse, and unidentified fish. Cultural pit feature 2 at White Rock Springs was found to contain bones from Elk (Cervus elaphus), Snowshoe Hare (cf. Lepus americanus), Fisher (Martes pennanti), Bushy Tailed Wood Rat (cf. Neotoma cinerea), Mule Deer (Odocoileus hemionus), and White Tail Deer (cf. Odocoileus virginiana) (McGill et al, 2010). Butchering marks were apparent on 47 of 241 bones identified to element and taxon, and approximately 75% were burned. The authors suggest the earth oven was being used as a dump for the bones, not a cooking facility. EeRj-71 is a site in Hat Creek valley that is composed of a lithic scatter and a large (5.6m diameter) earth oven with multiple rock lined basins (Pokotylo and Froese 1983:142; Turnbull 1977:155). The oven appears to have been used multiple times since there are many fire cracked rock lined basins with charcoal, with charcoal from the primary and secondary basins dated to 2250 ± 50 BP and 2120 ± 65 BP, respectively. Four 1m x 1m units excavated in the oven revealed faunal remains, including Mule Deer, and unidentified plant remains. Ketcheson found floral and flotation samples from cultural pit features at EeRj-1 and EeRj-55D to contain unburnt bone fragments, but these small remains likely have nothing to do with cooking (1979:12, 29).

Thorns argues that an increase in the number and sizes of earth ovens means there must have been population pressure on food resources (1989; 2008a; 2009), and Peacock has argued that this might be caused by environmental pressures (Peacock 1998). Thorns also suggests that the increase in earth ovens over time represents a 'carbohydrate revolution' (Thorns 2008b). I believe that root foods 43

were probably always carbohydrate staples in peoples' diets. Possible explanations for the appearance of more and bigger earth ovens could include: 1) the desire to trade/export prepared commodities, 2) good 'bumper crop' root season, so lots were harvested and processed, 3) competitive feasting and aggrandizing behaviour (Hayden 1996, 1998, 2009, 2011), and 4) social networks- more groups coming together to harvest and process roots. These are just initial thoughts and the questions related to why intensive use of earth ovens occurred prehistorically is by no means solved. Complicating the question of intensity of earth oven use over time is also the fact that older earth oven sites are less likely to be discovered and studied due to taphonomic factors.

Thorns work can be described as Binfordian processual, with optimal foraging and cost/benefit thinking and assumed rational behaviour of past peoples as a given. Thorns (2008a:457-458) argues that in the absence of food remains, ethnographic data and actualistic experiments are good ways to suggest functions of features such as earth ovens. I would agree that ethnographic analogy is a good first step, but used on its own to decipher archaeological features, it is inadequate. Thorns suggests that the food species eaten by past people can be inferred by feature type (eg. boiling pit, earth oven, etc.) and its characteristics (depth of pit, size of FCR), since different foods have different cooking requirements. He suggests that experiments can be designed to figure out what the feature types and their characteristics would look like archaeologically. It is true that there probably are some limits as to which plant and animal species were likely cooked in earth ovens, and this can be informed by ethnographic data. However, I find it problematic to assume which foods were cooked in a particular pit feature simply based on its size, depth, fuel types, etc., especially since there were probably a myriad of species cooked in pits in multiple baking events, each of which could potentially have its own 'feature type'. Furthermore, people likely did not always construct earth ovens for maximal cooking efficiency; at least some nutritive value is obtained from geophytes, regardless of cooking time. The most reasonable and profitable way to resolve feature function is to find the actual direct 44 archaeobotanical evidence of the foods that were cooked. Thorns et al. (2011) have shown that microfossils of baked foods (False Garlic, Nothoscordum bivalve, and Eastern Camas, Camassia scilloides) can be found on earth oven stones, but more support for these initial experiments is needed. Particular attention should be paid to differentiating between authentic residues and residues present naturally in the soil. Residues may be authentic when they are found on tools in higher concentrations than are present in the soil.

Cache pits are oval or round depressions in the ground .75 to 3.5m in diameter, with a depth of .2 to 1m, commonly lined with bark, grass, or stone and capped with a layer of dirt (Stryd and Hills 1972:195). Stryd and Hills called cache pits "... ancient 'root-cellars' excavated for the storage of perishable foodstuffs" but did not elaborate on the use of pits to store root foods (1972:195). Cache pits where great amounts of edible plant seeds, charred parenchymous tissues, starch in the soil or phytoliths are found could potentially be features in which plant foods were stored.

Plant Food Remains Over 100 plant food species have been identified from archaeological sites on the northern and southern Canadian Plateau. Lepofsky and Peacock (2004) identified five categories of plant foods used on the Canadian Plateau, of which some types are far better represented in the archaeological record than others. These categories include: root foods, fruits and berries, seeds and nuts, green vegetables and tree cambium. In the following sections, I will review the five categories of direct evidence of plant foods that Lepofsky and Peacock discuss.

Direct evidence of fleshy root foods is difficult to recover archaeologically, especially when more durable parts of the plant such as the seeds and herbaceous stems were not used. Disappointingly, very few geophyte or root taxa have been identified from excavations (Lepofsky and Peacock 2004:132). 45

As with seeds found in archaeological contexts, only charred remains should be considered cultural material; unburnt fragments of soft root food tissue are almost certainly modern (Brian Kooyman, pers. comm.). Lyons (2000) conducted a limited paelobotanical study with the aim of identifying root foods from eight earth oven samples from Keatley Creek. She found unburnt remains that resembled the skin (epidermis) of Chocolate Tips (Lomatium dissectum), and other possible bulb or root tissues, all tentatively. The only charred root material that can be inferred as archaeological from Lyons analysis was a charred bulb that resembled Geyer's onion (Allium geyeri) (2000:5). Ketcheson conducted an archaeobotanical analysis based on 99 botanical samples (almost exclusively consisting of wood charcoal) from 30 flotation samples from Hat Creek valley sites EeRj-1, 46, 55, 71, 92 and 101. She found capsules belonging to the lily Family (Liliaceae) associated with a large amount of charcoal in an earth oven at EeRj-55A, tentatively identified as Allium sp. root tissue (1979:23). It was not stated whether the Allium sp. capsules were burnt along with the associated charcoal. Possible food remains Ketcheson found in floral and flotation samples collected from EeRj-101 included: a small burnt Allium sp. bulb, burnt bulb fragments from the Liliaceae, and a burnt Pteridophyte (fern) rhizome (1979:34-39).

Examples of Allium sp. have been identified (Ketcheson 1979; Lyons 2000; Wollstonecroft 2000, 2002) and are likely the remains of the very widely distributed and culturally significant Nodding Onion (Allium cernuum). Flotation of sediments from EeRb-140 yielded charred plant material that is probably Nodding Onion (Nicholas and Westfall, in press; Wollstonecroft 2002:63). Other vegetative tissues were found in the flotation samples from EeRb-140, but probably cannot be identified further without attempting SEM (as done by Hather 1991) or DIC metallurgical microscopy, and having access to good reference collection material. Endo also found what she suspects are charred root remains in flotation samples from the Bridge River site (EeRI-4) housepits (Endo 2010:145). 46

The above examples are exceptional finds, since fleshy roots are rarely charred and disintegrate rapidly. The low chance of root preservation is noted by Peacock (2002:50) as "...root foods, which were placed in containers between protective layers of foliage to avoid burning, would be even less likely to become carbonized." So to be found archaeologically, a root food needs to have been charred and not collected from the earth oven, representing a mistake by people and an item not utilized by soil fauna or microbes (a very unlikely scenario).

J. G. Hather conducted an experimental study in which he attempted to find distinctive characteristics of charred root tissues that would be useful for species identification of archaeological materials (1991). Based on his results, it seems that fleshy organs 1.) need to have been dried prior to pit cooking for the tissue to maintain enough morphological features to have a chance of being identified, and even less likely, 2.) the root food needs to have been charred and either not collected from the earth oven, discarded in a midden dump, or left unconsumed in a storage pit. Granting charred remains of organ tissues are recovered, there is a possibility they will be overlooked altogether. Hather showed that species taxonomic identification is indeed a difficult task even at high SEM magnifications, but did find five useful guidelines to identify charred parenchymous tissues at low magnifications from wood, seeds, feces or other charred fragments. Pieces of charred vegetative parenchymous organs are often rounded, cells are spherical and not elongate, cells appear arranged in an organized way (not randomly distributed), may contain reflective vascular bundles, and often will contain regular or irregular patterns of cavities (1991:673). Hather found that "In many charred tap roots separate areas are apparent; an outer region external to the cambium, and a region internal to the cambium. The roots of different species vary in structure and one of these regions may be larger than the other," (1991:665). He also found that the distribution and pattering of vesicles and tension fractures in charred tubers differed between species, based on the gross morphology of the tuber. For example in small tubers such as Bitter Vetch (Lathyrus linifolius) the vesicles and tension fractures are tangentially oriented in the outermost tissues with other 47 cavities radiating from the central point of the tuber, whereas in large and irregularly shaped tubers from plants such as Wild Arum (Arum maculatum), the placement of the vesicles and tension fractures was random. Random placement of of vesicles was also seen in rhizomes such as Wood Anemone (Anemone nemorosa) (1991:665).

Direct evidence of fruits and berries furnish the most abundant food plant information recovered from Canadian Plateau sites, particularly as seeds. At the Keatley Creek site, sediments from floor, rim, and roof contexts of three housepits (3, 7, and 12) were floated and analysed spatially by Lepofsky (2000a, 2000b). This work yielded in-depth direct food plant evidence, including the seeds of Saskatoon Berry (Amelanchier alnifolia), Kinnikinnick (Arctostaphylos uva-ursi), Red Osier Dogwood (Cornus stolonifera), Heath or Heather Family (Ericaceae), Cherries (Prunus sp.), Gooseberry (Ribes sp.), Woods' Rose (Rosa cf. woodsii), Blue Elderberry (Sambucus cerulea), and Star-flowered Solomon's Seal (Smilacina stellata). The seed assemblage from housepit 7 at Keatley Creek was dominated by food seeds, and clusters of Blue Elderberry and unburned pine nuts (Pinus ponderosa) were excavated from the housepit (Prentiss et al. 2007:312-313). In addition to the food seed species from Keatley housepit contexts identified by Lepofsky (2000a:120), Prentiss et al. also list hawthorn (Crataegus sp.), prickly pear (Opuntia sp.), Ponderosa Pine, raspberry (Rubus sp.), Mountain Ash (Sorbus cf. sitchensis), and blueberry/huckleberry (Vaccinium sp.) as botanical foods represented in housepit 7. Cousins compared flotation samples from four small housepits (9,12, 90, and 104) from Keatley Creek and found seeds from edible foods including: Saskatoon Berry, Ericaceae sp., Kinnikinnick, blueberry/ huckleberry, Pinaceae sp., and Ponderosa Pine (2000:95). Remarkably, a flattened 'cake' of Saskatoon berries was recovered from an archaeological site in the Lillooet and examined by Rolf Mathewes; this is the only example of its kind noted in the literature (Stryd 1980:7). In the past, Saskatoon berries were the most important fruit collected by Interior First Nations, and this is still true today (Kuhnlein and Turner 1991:234; Wilson 1916:17). 48

Unidentified berry seeds were recovered from a female burial at the Terry site (EeRI-167). A broken comb carved from maple wood was found at her feet (Stryd 1985:78); it could have been a berry collecting comb. Concentrations of berry seeds were found at the Fountain site in a Kamloops Phase burial. They may have been placed beside the corpse in a pouch, and were also found in other contexts (Stryd 1972:39; 1973:68). In another burial near Kamloops, Kinnikinnik seeds were found with a woman of about 20 years old at the Government Hill site (Smith 1900:437). I identified fruit seeds from eight sites excavated in the 1970s by Arnoud Stryd during the Lillooet Archaeological Project (Stryd 1974,1978). These remains were recovered from the following sites: Bell, Ollie, Mitchell, Squatter's, Terrace, West Fountain West, EeRI-45, and EeRI-80 (Croft 2008). The seeds identified include Amelanchier alnifolia (Service or Saskatoon Berry), Rubus idaeus (Wild Raspberry), Prunus virginiana (Choke Cherry), Elaeagnus commutata (Silverberry), as well as cone seeds from Pinus ponderosa (Ponderosa Pine), Pinus albicaulis (Whitebark pine), Pseudotsuga menziesii var. glauca (Rocky Mountain Douglas Fir).

At the Bridge River village site (EeRI-4), flotation samples recovered from various activity areas within six housepits were examined by Naoko Endo (2009; 2010). The following seeds from likely food remains were recovered: Amelanchier alnifolia, Arctostaphylos uva-ursi, Ericaceae, Pinus spp., Rosaceae, Sheperdia canadensis, Crataegus spp., Opuntia spp., Rubus spp., and Sambucus spp.

Field schools and excavations of the Secwepemc Cultural Education Society- Simon Fraser University (SCES-SFU) between 1991 and 2004 have aimed to recover information about mobile hunter gatherer sites away from the flood plains on the Canadian Plateau before 4,500 years ago. The SCES-SFU research program directed by Geor^Ntcholas has incorporated the collection of paleobotanical flotation samples. The project has investigated over 60 sites, sites EeRb-130,140,144 and 149, being the most intensively explored by testing and excavation. Only light fraction flotation samples for one site, EeRb-140, have been 49 analyzed thus far. Seeds of Amelanchier alnifolia, Cornus sericea, Pinus spp., Prunus virginiana, Ribes spp., Rubus spp., and Vaccinium spp., have been identified from flotation samples from EeRb-140 (Nicholas and Westfall, in press; Wollstonecroft 2002:63). Wollstonecroft also found fleshy fruit vegetative tissue of Amelanchier alnifolia, Ribes spp. and other berries that were unidentifiable.

As mentioned previously, seeds were found by Nicolaides (2010:94) in three White Rock Springs earth ovens: Arctostaphylos uva-ursi (Kinnikinnick), cf. Berberis aquifolium (Oregon Grape), cf. Rosaceae (Rose), Rubus sp. (Raspberry/ Thimbleberry), cf. Shepherdia canadensis (Soapberry), Asteraceae cf. Achillea sp. (Yarrow), Chenopodium cf. capitatum (Chenopod), cf. Fabaceae (Pea), cf. Polygonaceae (Buckwheat), cf. Ranunculaceae (Buttercup), Cyperaceae (Sedges), and Poaceae (Grasses) (2010:94). Nicolaides suggests some of these seeds may represent foods, flavourings, or may have been included incidentally in earth ovens.

Seeds and nuts are also sometimes encountered archaeologically. Charred Whitebark Pine (Pinus albicaulis) cones were excavated at several housepit sites in the Lillooet area, with cones found at the Fountain site in a 1500 year old small earth oven lined with fire cracked pebbles and charcoal (Stryd 1972:23; 1973:69; 1974:29). The sealed cones of Whitebark Pine contain edible seeds that are usually broken open and dispersed by birds, but can also be released by exposure to heat. Clark's Nutcracker (Nucifraga columbiana) transports and stores Whitebark Pine nuts, but the seeds have a hard coat that is not broken by the bird (Len Hills, pers. comm.). Charred axes of Whitebark Pine cones were also recovered from Squatters site (EeRI-36) and West Fountain West site (EeRI-63) (Croft 2008).

Macrobotanical remains of fungi are incredibly rare from archaeological contexts. So the discovery of eight mature puffballs from the Bridge River site (EeRI-4) and the Mitchell site (EeRI-22) was novel (Compton et al. 1995). These puffballs were 50 probably collected for ritual or healing reasons because puffballs are only palatable when young and the spores are undeveloped. Ketcheson found a fragment of unburnt Fomes sp. bracket fungus from an earth oven at EeRj-101 that was perhaps, but not likely, used as a food (1979:39).

There is currently no direct evidence of green vegetables such as shoots, leaves, or tree cambium (inner bark). Such foods are nearly archaeologically invisible with current analytical techniques. I feel it is possible that leaves and stalks of green vegetables may be visible via phytolith analysis and/or identification of vascular tissues, and tree cambium may be detectable by residue analysis. 51

CHAPTER 3: RESIDUE ANALYSIS: NEW INFORMATION FROM OLD TOOLS

Historical Overview of Lithic Residue Analysis with Emphasis on Starch

Residue analysis offers a new way to approach Canadian Plateau stone tool analysis and provide objective direct evidence of past activities. I will briefly review how analysis of ancient use residues began and its development as a discipline in archaeology. The development of starch as a research subject in biology, and later archaeology, will be examined in particular since starch was a focus of this study.

Artifact Residues George Odell (2004:156) succinctly explains what archaeological trace residues are and the main obstacle to be hurdled when they are recovered:

Residue is the non-soil stuff that clings to a stone tool [or any artifact]. It can be organic or inorganic and it may have gotten there through precipitation, meaningful contact with the origin of the substance, or because a cave bear sauntered through and sat on the tool for awhile. In other words, the association of a substance with the surface of a stone does not necessarily tell you how that substance got there.

Residues come in many forms. Organic residues are the carbon-based (in combination with H, N, O, P and S) remains from bacteria, fungi, plants, and/or animals. These can be isolated as biomolecules, as has been done with lipids, proteins, starches, DNAand plant lignin (Eerkens and Barnard 2007:1). Studies of archaeological organic residues have employed chemical techniques for identification, such as gas chromatography, mass spectrometry, infrared spectroscopy, energy dispersive spectroscopy, and stable isotope analysis (eg. Boyd et al. 2008; Deal 1990; Evershed 2008; Hill and Evans 1987; Jahren et al. 1997). However, many residues including starch, phytoliths, hair, feathers, and plant and animal tissues are identified by using microscopes to examine morphological traits. The residues of particular concern in this study are those organic traces which represent material that were of food or technological 52 importance to past peoples, and that can be identified using light microscopy (LM) at magnifications ranging from 100x to 400x.

Starch as a Biological Subject Starch residue analysis in archaeology has been primarily informed by biological science. The study of archaeological starch has benefited from progressive advancements in their recognition, morphological classification and chemical properties. Dutch microscopist and 'father of microbiology' Antoni van Leeuwenhoek likely published the first drawings of starch grains, along with other plant and animal cellular illustrations (1719). German botanist Jakob Matthias Schleiden, one of the co-founders of cell theory, was the first to construct a key to classify plants based on the recognition of the taxonomic potential of unique starch grains (1849). Schleiden used microscopic observations of starch grain morphological traits to devise his classification scheme. Simple, compound or amorphous were the three major divisions of Schleiden's starch key. Exhaustive investigations into the physical and chemical nature of starch grains were undertaken by Swiss botanist Karl Wilhelm von Nageli (1858), and 37 years later by German botanist Arthur Meyer (1895). Nageli devised a classification system of starch grains from wild and cultivated plants, improving upon the three major divisions originally set out by Schleiden. Meyer made his classification system of starch grains by taking a developmental approach, carefully studying the stages of starch grain development. He conducted many years of painstaking comparative experiments of the changes to the grain that vary by: species of plant organ tested, season, and laboratory tests (Ward 1895:640). Meyer also showed that the presence of fissures and pores on starch grains had an effect on the level of enzyme and solvent damage to the grains (Meyer 1895:240; Ward 1895:641). Meyer's system of starch taxonomy proved cumbersome to use, and botanists came to regard Nageli's classification as more practical (Ugent 2006:116).

American Edward Tyson Reichert and his assistants produced a detailed and extensive account of quantitative and qualitative properties of various plant 53 starches (1913). The goal of this work was to find an underlying concept to explain some of the major biological questions of the day such as the processes of heredity, influence of environment, mutations and sex differentiation. Reichert believed the key to better understanding these processes might lie in the blood of animals and the starch of plants, since these 'vital substances' are unique to each species (1913:1V). His volumes focus on description of starch found in agriculturally important plants, although examination of wild plants were included as well, 300 species in all.

Starch as an Archaeological Residue Farias Martins was one of the first to identify starch grains from archaeological contexts. Martins PhD dissertation examined sediments and macrobotanical remains from Peruvian archaeological sites (1976). Using LM and SEM, he found evidence of root crops in the form of starch grains, but other residues also were examined, including vessels, epidermal cells, and cork cells.

Scientific investigation of all types of residues, including starch, took off due to interest in stone tool use wear and functional analysis. As the edges of lithics were being examined microscopically, plant and animal residues became noted as items of research in their own right. This initial work focused on prehistoric residues that could be examined with light (optical) microscopy using medium to high magnifications (Loy and Nelson 1986:180). Frederick Briuer conducted the first study to indicate that microscopic plant remains could be removed from a substrate (stone tools) and can be used to help archaeologists understand the function of the artifact and activities of past peoples (1976). His lithic sample consisted of 37 tools from Arizona sites: 33 unwashed lithics from two rock shelters, and 4 washed tools from an open air site. Three analyses were conducted: 1) use wear with low magnifications (30x), 2) microscopic examination to identify in situ organic residues based on morphology and reaction with a series of botanical chemical reagents, and 3) crime lab analysis. The second and third approaches were not carried out by Briuer. Plant parts identified morphologically and with chemical staining 54 included: starch grains, stellate hairs, pollen, calcium oxalate crystals, raphides, cell walls (lignified, suberized and cutinized), cell lumen, tracheids, fiber tips, spiral vessels, and hair vessels. Even at this early stage of residue analysis, Briuer had the foresight to try to exclude noncultural natural phenomenon as causal factors for the presence of the residues. He did this in three ways, although these were not well-illustrated in his report. First, he examined the 37 chosen tools and confirmed use wear on 22 tools that had residues, giving these tools a stronger corroborative basis for being labelled 'used by people in the past' (1976:479). Second, he looked for similar residues with a comparison of on and off-site items as controls. Briuer found no similar residues on all of the off-site randomly selected rocks and macrobotanical remains tested. Many non-cultural items from the archaeological sites were examined for residues as well, with 20 tested with chemical reagents. Only 3 (1 bone, 2 plant twigs) gave positive reactions showing plant residues were present (1976:480). This showed that it is possible (but in this case infrequent) for non-cultural items to contain plant tissue residues. Finally, Briuer found the location of in situ residues to be associated near worked tool edges, which also provided support for the conclusion that the residues resulted from human use activities (1976:478).

A few years later, Shafer and Holloway (1979) followed Briuer and did residue analysis in conjunction with use wear of 11 Archaic chert flake tools from Hinds Cave, southwest Texas. They sought animal and plant residues to suggest tool functions. Starch grains were found on 7 tools. Shafer and Holloway also found phytoliths, pollen, tracheids, calcium oxalate crystals, plant fibers, charcoal, rodent and other hairs, and epidermal cellular fragments from lechuguilla (Agave lechuguilla), Agave (Agave sp.), sotol (Dasylirion sp.), and yucca (Yucca sp.) (1979:395).

By the 1980s more work was being done with archaeological starch and other residues. Don Ugent et al. determined the taxonomic origin of macrobotanical remains at South American sites (Ugent et al. 1982; 1984; 1986). Fullagar used 55 residue and use wear analysis to better understand tool function in Oceania (1986; 1988; 1989). Loy and Nelson promoted stone tool residue analysis and outlined potential research directions for the discipline (1986). Loy investigated starch from stone tools in the 1990s (Loy et al. 1992; Loy 1994), although he will best be remembered for his pioneering work on blood residue identification (eg. Loy 1983).

Starch grain research in particular owes much of its archaeological development to workers in Oceania. Basic questions concerning taphonomy of starch and contamination issues have been studied and applied to archaeological problems. The 2006 book 'Ancient Starch Research' edited by Robin Torrence and Huw Barton remains the major text for archaeological starch studies. This book not only provides information on the usefulness of starch to archaeology, starch biology, identification features, taphonomy, and practical advice on lab methodology, but also summarizes a number of case studies. The case studies present work from Papua New Guinea, New Zealand, New Ireland, Egypt, South Africa, Louisiana (USA), Honduras, Peru, Chile, and other areas, however the book draws mostly from Australian archaeology.

Residue Analysis as a Way Forward

As with any choice of research technique, there are pros and cons. Residue analysis is valuable in its ability to capture direct evidence of past peoples at a fine scale, its potential to discover ancient food resources that would otherwise be unknown, and its ability to provide evidence of plant foods that typically do not preserve macroscopically. Some of the drawbacks include: risk of contamination of artifact residues if improperly collected or handled, time and cost investments, and requirement of specialist knowledge or experience. Since its inception, archaeological residue analysis has been considered a specialists realm. However, researchers wanting to incorporate residue analysis in their studies can do so with relative ease once a reference collection of modern and archaeological residues is established. Also, residue studies do not have to be extremely specific to be useful, 56 the specificity required depends on the research questions being posed. For example, many studies that examine starch residues from artifacts are only looking for one type of starch, usually corn (Zea mays). My work on identifying the actual remains of the plant foods being processed at White Rock Springs aims to contribute to creating a comprehensive list of plants identified to the lowest taxon possible.

Starch analysis has found a variety of applications in archaeology. Starch has been used to identify species of macrobotanical remains (eg. Ugent et al. 1982; 1984; 1986), site activity areas (Balme and Beck 2002; Henry et al. 2004), stone tool function (Perry 2004; Zarrillo and Kooyman 2006), hominin diet and behaviour (Hardy et al. 2008; Mercader 2009), starchy food species consumed (Henry et al. 2010; Horrocks et al. 2004b; Piperno and Dillehay 2008), geographic extent of dietary patterns (Boyd and Surette 2010), use of domesticates or cultivated plants (Horrocks and Bedford 2010; Parr and Carter 2003; Perry 2002; Piperno 2006a; Zarrillo et al. 2008), and examination of cooking/processing techniques as suggested by modified starch (eg. Henry et al. 2009; Messner and Schindler 2010; Samuel 1996).

Artifact residue analysis has been successfully carried out at archaeological sites around the world, a trend that will likely grow. Starch in particular seems to show good preservation and has provided evidence of foods that were unknown archaeologically. Furthermore, starch residues may be especially well-suited for the detection of plant underground storage organs. For example, work by Messner et al. identified starch of roots or tubers from lithic and ceramic artifacts from Late Woodland sites in the Upper Delaware valley, food sources that were previously unknown (Messner 2008; Messner et al. 2008:123).

The attempt to identify plant foods important to the diets of past Plateau peoples from microfossil evidence is just beginning. One work on the subject is a phytolith study by Kendall Eccleston. She prepared a reference collection that tested 88 57 plant species for phytolith production, characterized distinctive phytolith forms for various families, genera and species, then applied this comparative collection to archaeological soil samples from three areas of the Cottonwood-Divide Creek site 10IH1286, an upland occupation on the Columbia Plateau in Idaho (1999). Eccleston found diagnostic forms for Asteraceae, Poaceae, Polypodiaceae, Pinaceae, Cyperaceae and Rosaceae, with genus and species level forms found in Celtis reticulata, Pinus ponderosa and Larix occidentalis. The ethnographically edible roots of Apiaceae and Liliaceae both yielded phytolith forms that appear to overlap with the short cells produced in grasses. However, decorated bilobate and 'cooling tower' phytolith forms were used as tentative indicators of root crops since the panicoid grasses in Eccleston's comparative collection did not produce them (1999:154, 205). She found three soil samples from the site to contain extremely high concentrations of root crop phytoliths. Two of these samples were taken near a feature and pestle below a date of about 3500 years ago, showing that intensive root crop processing occurred from the onset of site occupations (1999:205).

Residue analysis can make a valuable contribution to Canadian Plateau archaeobotany. It is important to identify the kinds of root foods that were relied upon since geophytes provided the main source of carbohydrates for past peoples on the Canadian Plateau; they were crucial to fulfilling overall dietary requirements (Peacock 1998). No root foods have been identified at White Rock Springs, and residue analysis offers a unique opportunity to 'see' them in the archaeological record. This microscopic vantage point can be successfully employed to determine the tools used for root food processing. These tools have thus far eluded identification.

Starch granules are a powerful archaeological tool for three reasons: the grains are ubiquitous, show long-term persistence (at least up to 100,000 years), and can provide information on plants and/or complement macrobotanical analysis. Many, but not all, plants produce potentially thousands of storage starch granules in their storage organs, seeds and/or fruits. These were often targeted by ancient peoples 58 as food resources. These starch granules are released from the plant during processing and they thus spread on tools and surfaces when plants are processed and cooked. Starch granules have a high preservation potential across a broad range of site types (eg. tropical, temperate, arid, etc.). The granules have been shown to be durable and long-lived. Starch analysis can help fill in the knowledge gaps in site environments where macrobotanical preservation is limited or where botanical representation is believed to be biased.

Extracting residues from White Rock Springs lithic remains is justified as an appropriate method to gain insight about past diets due to the considerable obstacles in identifying macrobotanical root tissues. Microscopic traces of the plants that were processed and eaten are the most profitable manner in which to proceed in answering what kinds of root foods were used in the diet.

Residue analysis is furthering the archaeological rethinking of generic lithic categories such as 'hide scrapper', 'fish knife', or 'hammer' that denote function. As Lepofsky has noted, the function of an artifact understood by ethnographic analogy is not known with certainty until the residues on the tool are examined (2004:433), perhaps alongside use wear as well. As many studies have already shown, the ethnographic use of a tool and actual residues may not match. My hunch is that stone tools were used in more ways than previously believed, with multifunctionality being the rule rather than the exception. Residues from berries, meat, feathers, and wood may all be found on the same tool, and use wear may also indicate contact with many types of food and/or technological uses. 59

Theoretical Considerations in Residue Analysis

Human ecological perspective The land, water, atmosphere, plants and animals that peoples experienced and interacted with are important to our understanding about cultural traditions. They constitute the mileu within which cultural practices form and change. The physical and biological context of the Canadian Plateau area provides the environmental history needed to understand past lifeways in the Hat Creek valley.

Hayden has stated that ecological factors are the most profitable way to explore hunter-gatherer issues such as population density and demography, degrees of mobility, optimization of deviation from rational cost/benefit choices, diet and reasons for abandoning settlements (1992a:5). However, Hayden's intellectual commitment to the ecological perspective can be questioned on grounds that he mainly presents models which emphasize the actions of wealthy elite individuals in society, with their prestige objects, competitive feasting, creation of social debts and aggrandizing behaviours, all of which are primarily social (not ecological) factors driving cultural change. His view on cultural change in the mid-Fraser River region can be described as heavily weighted towards discovering what he terms "... fundamental social characteristics of traditional cultures ..." (1992b:525). Prentiss et al. have argued that changes in the environment are the primary force of cultural change, rather than the agency of high status elite individuals (2005a). For Prentiss, the development of economic inequality was not actively pursued by strategizing individuals, but likely occurred as an "unintended effect of other developments," (2011:30).

The human ecological perspective underpins this study. I see resource availability as a key factor that can limit peoples' options or instigate a change in behaviour, but these challenges are met with creative solutions and the choices made by decision-makers. In this way, environmental determinism can be avoided by respecting that humans are important actors that shape their own destiny even in 60 times of resource stress. I also believe that even though archaeological data are always biased, corrupt, patchy, incomplete, and influenced by taphonomic factors, there are real patterns of past lifeways that can be observed and made relevant to the present.

Women and Visibility Archaeologists have had a tendency to see diet and economy through animal remains and lithic remains, the latter assumed to be hunting tools and hence indicative of a male package of activities. Archaeologists have often taken no account of the plant aspects of the economy, at the expense of understanding women's activities. The roles of women in economy have generally been overlooked and there has been a failure to recognize and value women's subsistence contributions. There may be seasonally used sites on the Canadian Plateau that are specific to women's task groups carrying out plant processing. Woolstonecroft (2002:69) suggests this may have been occurring at EeRb-140, a location where berry drying and preserving was done before the crop was ready for storage in the village. Sites that document women's activities may be locations where earth ovens are concentrated at places that were key to resource procurement and plant processing sites. These were likely places where men, women and children gathered and camped and worked together. However, ethnograpically root earth oven cooking is exclusively associated with females (Dawson 1982:19; Steedman 1930:509; citations within Alexander 1992b:127). Peacock notes that root food production visible on the landscape as earth ovens and associated artifacts represents women's work and their contribution to subsistence (Peacock 1998:321).

Theory in Residue Analysis Theory in the study of ancient organic residues is underdeveloped. This might be expected in an area of archaeology that is relatively new and contingent on developments in the natural sciences and technology. Rather than a theoretical approach, residue analysis predominately remains a collection of techniques 61 drawing heavily from the natural sciences, especially biology and chemistry. Techniques include high and low power microscopy (for microfossils such as starch and phytoliths), gas chromatography (lipids), stable C and N isotopic values (C3 and C4 plants, protein), and infrared spectroscopy, among many others. There is a widespread perception that views residue analysis as a means to identify what human activities occurred, not why they occurred. This perception is in large part due to visibility of the literature, with publications in science-oriented journals the most commonly known. On the other hand, publications that focus on application of residue analysis results found in site reports and regional journals are less well- known. Attempts to address 'bigger issues' in anthropology by using residue analysis have admittedly been few, with some scholars pointing to agency as a theory to improve the situation, marrying social and scientific. Studies based in the natural sciences are being confronted by the inseparability of the physical and social environments. It is difficult to focus on humanistic interpretation of residue analysis results when there are so many methodological problems to consider as well as a lack of actualistic experimental data. Basic experiments to assess the reliability of archaeological residue analysis results are needed. Techniques are in no way standardized and different workers have various suggestions as to the best way to collect and prepare residues. Uniformitarianism is a foundational assumption of the preservation of residues. Taphonomic residue degradation factors such as the movement of water, pH, and the enzymatic action of fungi, macro and microorganisms are assumed to act in similar ways today as they did thousands of years ago, and this is common practice in taphonomical studies.

Michael Haslam has discussed the scale of residue analysis as related to interpretive power. He raises the issue "...of the theoretical appropriateness of asking broad questions of micro-scale techniques," (2006a:406). This is relevant to all residue analysts, as it is true that a disconnect easily occurs between the micro focus of residue analysis conducted on a (usually small) selection of tools and the artifacts' broader archaeological context. Residues from lithic tools provide only a snapshot of one or more event(s) of the past. The lithic piece may have been an 62 expedient tool used once or a curated tool used for a set of activities over a longer period, thus lithic residues may represent snapshots of variable size. Overall, the event(s) that led to residue deposition on the tool are short-term occurrences. Another scalar issue raised by Haslam is the problem of workers over-extending their interpretations based on residue data provided by small artifact sample sizes. Often only a few tools are analyzed in studies, and this may not be sufficient data upon which to base interpretations and conclusions. Published stone tool residue studies over a thirty year period from 1976-2006 were characterized by small sample sizes, with a mode of only three artifacts (2009:49). So does this mean results that are derived from three tools are questionable? I feel the issue that lies at the crux of the sample size conundrum is the consideration of the trade-off that must be made between quantity and quality of artifact analysis: more tools means results that are more representative of an assemblage and an increased validity of results, while fewer tools analyzed with greater sophistication of technique produces more detailed data. Which sample size approach should be adopted? The suitability of small or large sample size and the number of characteristics and/ or measurements per tool will be dependent on the researcher's questions.

Although rare, there have been some suggestions as to how to make residue analysis and use wear in archaeology more theoretically engaging, with agency, and landscape the major concepts posited thus far. Haslam contemplates the ways in which new theoretical approaches such as agency and narrative can be applied to stone tool residue analysis (2006a). To this end, he admirably aims to increase the humanistic value in an area of archaeology that is highly technical and based in a wide range of scientific methods. He considers agency to be the deliberate actions carried out by an agent who has at least some ability to act otherwise. Individual actors in the past left residues that record specific actions on stone tools, and Haslam suggests this provides a good way to incorporate agency into the theory of residue analysis. Residue analysis brings a sense of life to the artifacts and helps us imagine people of the past because the information is so detailed. For example, when the residue analyst concludes 'this chert scraper was used to 63 process starchy root foods at an earth oven site,' it provides a vivid and captivating vision of the person who used that tool in the past. This makes the archaeological record engaging for the public, something they can relate to, an important goal of archaeology. Agency becomes more tenable when residue analysis can be combined with an examination of lithic manufacturing patterns and use wear studies. When multiple types of analysis can be combined in this way, individual workers can be identified.

Haslam promotes studying single artifacts as a means of showcasing the individual's action (2006a:405). I remain doubtful of the merit in studying the residues of a single artifact to construct social concepts. Understanding the actions of individual tool users is only worthwhile if it is ultimately overlaid upon the general picture presented at the site, and then eventually to regional culture level. To my mind, a single artifact approach can most fruitfully be employed where a good deal of basic research has previously been conducted, and thus a framework is available to support interpretations.

The artifact can be conceptualized as its own site on which human actions are inscribed. Specific areas on a tool can reveal prehensile use wear, direction of tool use and type of action (circular boring, scraping, grinding), hafting residues, as well as the distribution of starch, muscle tissue and wood residues. The concept of discrete 'locales' of the lithic are somewhat analogous with the activity areas of a site identified by the location of seeds, bones, processing tools, hearth features and discarded materials. This perspective was pioneered by Tom Loy who considered the artifact as site (1993). Loy stated that the trace residues gleaned from the artifact can provide a wealth of information that can be as helpful as the data from the actual site (1993:44). He illustrated this point by using many techniques on a lithic assemblage from northern British Columbia, conducting in situ microscopic examination of tool residues, protein analysis, ancient DNA and collecting carbon from protein material to directly AMS date. These analyses showed that many of the tools contained multiple lines of evidence for bison 64 butchery, and were AMS dated to 2,180 ±160 years, forcing the reevaluation of the current picture of small groups of people hunting in the boreal forest (1993:57). The implication of Loy's work is that individual stone tools can make excellent subjects of study in their own right, their residues revealing a great deal about past peoples diets. For the 'artifact as site' perspective to be applied, residues must be collected using a targeted residue sampling technique on each artifact. This means that tiny pin point samples are taken from different areas of the lithic tool, for example, at the flaked working edge, at the end opposite the working edge where hafting might be suspected, in the center of the tool where the tool could have been held in the hand, etc. An wide-spectrum approach to residue recovery from artifacts was taken in this study to recover the most residues possible. This choice of total residue removal technique will be discussed more in the Procedures section of Chapter 5.

Residue and use wear analyses provide the techniques to move beyond traditional stylistic, typological, and attribute analysis to provide new detailed information. In practical terms however, these new methods from wide ranging disciplines can be time consuming, costly, and technically involved to conduct. The time and costs required to conduct residue and use wear analyses may be fixed. On the other hand, the need for researchers to become more skilled and learn new methods of analysis is certainly a challenge that archaeologists can meet.

The analytical scale of residue and use wear analyses is very small-at the level of microscopic remains from individual artifacts. Yet this information can be applied to higher levels of analysis to understand patterns of human actions and behaviours, a progressive widening of scope from the microscopic to large-scale synthesis. For nested levels of analysis to be done effectively and convincingly, however, large artifact sample sizes from many sites will be required. The lithic as a site concept fits into this bottom up strategy to capture and observe the specific actions of past peoples, the foods consumed and materials worked. Banks carried out a detailed high-power use wear analysis of Upper Paleolithic stone tools through four cultural periods from the Solutre site, France (2006). He considered each artifact to have 65 its own landscape on which human patterns of use can be discerned (2006:90). Banks emphasized that microscopic tool data provides the best information starting point with which to pass through increasing levels of synthesis to understand big patterns. The artifact as a site concept is also suited and can be applied to the residue analysis of pottery. For example, vessel locales such as tar glue repair sites, waterproofing sites, and food residue sites are all areas that are records of specific human behaviours and are able to be identified.

The designation of specific residues on tools has been criticized as being subjective. Some publications ask us to trust the analyst's identifications without adequate demonstration (visual or descriptive) of the characteristics that were used to identify each residue type. Furthermore, analyst levels of residue experience and skill differ. This criticism calls attention to the need for 1) standardization of terminology and descriptions, and 2) micrographs of reference material to be presented and made available, preferably accessible through online databases. This is an issue that will become better resolved over time with greater collaboration and exchanges of information between residue analysts. 66

CHAPTER 4: RESIDUES, ARCHAEOLOGY AND TAPHONOMY

Starch Biology

Plants have an energy reserve composed mainly of carbohydrates, or fats and oils. The most common carbohydrate storage molecules used by plants include starch, fructans (typically inulin), and sucrose (Messner 2008:60,78). Plants whose main store of carbohydrates is inulin -not starch- in the Hat Creek valley include Balsamroot (Balsamorhiza sagittata), Yellow Avalanche Lily (Erythronium grandiflorum), Nodding Onion (Allium cernuum), and Edible Thistle (Cirsium edule) (Peacock 2002:46). All of these food sources, except perhaps the thistle, were of very high importance to First Nations peoples in interior British Columbia, and all were regularly pit cooked (Peacock 2002).

Plants and animals use glucose, a monosaccharide (one sugar molecule unit, a simple sugar) as the principal source of energy in cellular metabolism. In green plants, glucose is stored for future energy needs in long molecule chains of amylose and amylopectin, which compose the polysaccharide starch (Raven et al. 2005:17-18; Simpson 2010:516). There are three types of starch found in plants: small temporary or transient starches found in the chloroplasts, large grains called statoliths found in the root cap cells used for perception of gravity, and storage starch-the starch archaeologists are primarily concerned with. Storage starch accumulates in storage organs including seeds, fruits, swollen stems, tubers, and roots, and to a lesser extent, in most other vegetative tissues of the plant (Copeland et al. 2009:1528). Starch grains from different locations may have different morphologies in the same plant. Environmental factors also affect size and frequency of starch grains. Plants that are stressed from lack of water or overgrazing produce fewer and smaller starch grains (Gott et al. 2006). Some plants undergo substantial seasonal changes in the level of storage starch available in their subterranean organs. Starch reserves are mobilised within the plant to meet metabolic demands during seasonal periods of regrowth. Maximum 67 starch accumulation is reached during summer or fall, and the least amount of starch is present six months prior, in the winter and spring. For example, Messner found that some reference collection species, including Wapato (Sagittaria latifolia) and False Solomons Seal (Smilacina racemosa) collected at different times of year contained high or no evidence of starch (2008:66-67). The seasonal oscillation of starch accumulation in some plants certainly has implications for collecting reference collection material, as it is clear that starch visibility in certain species depends on what time of year the plants were harvested and processed. This problem highlights researcher's need to gain an intimate knowledge of individual plant species biology.

Researchers studying archaeological starch should have a basic understanding of how plants make starch. In the following section, a simplified outline of starch biosynthesis is presented.

The light reactions of photosynthesis are the primary way in which green plants capture some of the sun's energy and convert it into food energy for cellular processes. Phytosynthesis can be described as 6CO2 + 6H2O —» C6H12O6 + 6O2, with glucose and oxygen being the final products. Green plants need to store much of the energy collected from the sun for use months or years later.

When the chloroplasts are carrying out photosynthesis at a rate faster than sugar biosynthesis, up to one half of recently assimilated carbon dioxide is polymerized into insoluble temporary or transient starch granules in the cholorplast stroma between thylakoids (Badenhuizen 1969:96; Mautseth 2009:51; Weber 2006:275). During the dark period at night, transient starch is hydrolyzed by three major enzymes (endoamylase (p-amylase), exoamylase (a-amylase), and plastidal disporportionating enzyme (D-enzyme)) back to a soluble sugar and exported from the chloroplasts to the sink organs and leaves, providing the plant with a constant supply of carbon (Rost et al. 2006:134; Smith 2010:1; Weber 2006:278). The plant usually uses sucrose (a disaccharide form of sugar composed of glucose and 68 fructose) to transport carbohydrates from one tissue to another (Bidlack and Jansky 2011:22).

Long-term storage of energy occurs in non-photosynthesizing sink organs of the plant (mostly in roots, but also to a limited extent in the cortex and pith cells of bark, wood and the endosperm of seeds) and is accomplished by packing glucose polymers into ordered starch granules inside plastids (plant cell organelles) that can be used later for food or metabolized as energy. Storage starch biosynthesis at the amyloplast begins with sucrose imported to the amyloplast in the seed, fruit, tuber or storage root. Sucrose is hydrolyzed into a-glucose (ring form) and (3- fructose (ring form). Enzymes polymerize a-glucose monomers into short chains called glucan molecules. These short chains are then joined by enzymes to make longer molecules in two different forms: amylose and amylopectin. Amylose molecules contain -1000 a-glucose in a relatively unbranched long chain (about one branch every 1000 a-glucose) that bonds to itself in a single helix shape. Amylopectin molecules contain 1000 to 6000 or more a-glucose monomers with short chains of about 8 to 12 a-glucose branching off of the main chain at intervals of about 12 to 25 a-glucose monomers (Martin and Smith 1995:971; Raven et al. 2005:17). Inside the amyloplast cell, many amylose and amylopectin molecules cluster together into starch grains. Starch grains normally contain about 20-30% amylose and 70-80% amylopectin (Belitz et al. 2004:318), water accounting for 10-15% of starch by weight (Tester 1997:163).

Starch structure varies genetically, developmentally and environmentally. Starch also varies from one organ to another in the plant, between species, and between cultivars within species (Kaur et al. 2002; Smith et al. 1997:74). Starch grains are primarily composed of amylose and amylopectin, but also contain small amounts of water, lipid, protein and phosphorus, which affect properties of the grain such as swelling, solubility, gelatinization, light transmittance and flavour (Galliard and Bowler 1987:65; Martin and Smith 1995:973; Gallant et al. 1997; Singh et al. 2003; Tester et al. 2004). The overall morphology of the grain may also affect these 69 properties. Starch grains are semicrystalline with varying levels of crystallinity based on the ratio of amylopectin to amylose in the grain. Crystallinity and birefringence is associated with amylopectin, while amorphous regions are mainly amylose (BeMiller and Whistler 1996:192; Copeland et al. 2009:1527; Singh et al. 2002:220). The starch grain is composed of three different regions: amorphous lamellae (amylopectin branch points), crystalline lamellae (chains of amylopectin joined together in double helix formation), and amorphous growth rings. The internal structure of the starch grain has been described as made up of spherical 'blocklets' of 20 to 500 nm in the amylopectin lamellae (Gallant et al. 1997:179; Tang et al. 2006:555). The blocklet model was previously described as the 'cluster model': amylopectin clusters are organized into short chains that are connected and regularly spaced (Zobel 1988:47). At the finest level of structure, starch grains appear to be organized into alternating semi-crystalline and amorphous growth rings (Belitz et al. 2004:318; Sujka and Jamroz 2007:107). Currently, it is unknown where amylose and amylopectin are packed in starch granules, but there are at least three competing models to elucidate the internal structure and locations of the components that make up the grain (Tester 1997:169; Bello-Perez et al. 2010:34-35).

Raw, native starch granules are insoluble in water at room temperature, and highly resistant to enzymatic hydrolysis (Colonna et al. 1987:79). The physical properties and crystallinity of starch grains are affected by the ratio of amylose to amylopectin (Gott et al. 2006:42), which also defines the rate and extent of enzymatic hydrolysis (Buleon et al. 1998:97).

Starch Features useful to Archaeology

Previously, it was thought that some destructive chemical test might be needed to identify starch grains to species (Hall et al. 1989:148). We now know it is unnecessary to sacrifice any part of the archaeological residue sample. Destructive techniques are also undesirable when only trace amounts of residue are present 70 on lithic tools. Starch grains can be quickly and easily identified with the use of a polarizing filter on a transmitted light or optical microscope (LM). Many starch grain characteristics can be accurately viewed with LM at magnifications of 400x or higher. Morphology and other properties of the grain are under genetic control unique to each plant species, and it is for this reason that starch can be identified with consistency.

Starch grains are clear, usually colourless and have a mean refractive index of 1.5, which is somewhat affected by moisture content (Blanshard 1987:19). The most important characteristics of starch grain morphology are: birefringence, the extinction cross, three dimensional shape, hilum, lamellae, fissures, simple versus compound granules, and the presence or absence of facets.

Starch grains appear to stand out as bright white spots (often with an 'X' visible) against a dark background when viewed with cross polarized light, this characteristic known as birefringence. Birefringence is one of the most useful ways to identify starch grains in a preparation without the use of stains.

Native intact and partially gealatinized starches display an extinction or Maltese cross under polarised light (Fig. 4-1). The arms of the cross will rotate as the polarising lens is rotated. During microscope slide transect scanning, it is easiest to identify starch while using the polarising lens. This is because all native and even partially damaged starches are birefringent and easily spotted by a characteristic dark 'X' against the white area of the rest of the grain. If the starch grain becomes completely gelatinized, its semicrystalline structure is broken and the extinction cross is obliterated. 71

Fig. 4-1. Light micrograph of storage starch grains from White Potato (Solarium tuberosum) under cross polarized light, 250x magnification. Note extinction crosses.

Many starch grain shapes can originate from the same part of the plant, such as the grains illustrated in Figure 4-2 from the tuber of White Potato (Solarium tuberosum). It is important to record the shape of the grain in three dimensions, as it has taxonomic discrimination value. Slide preparations of starch must be fixed in a medium that does not harden, allowing the grains to move when the slide is gently probed during microscopic examination. Grains often appear to have a simple shape when only viewed in one plane, in two dimensions, but may reveal a more elaborate and distinct form when rotated.

Fig. 4-2. Different shapes of storage starch grains from White Potato, 250x magnification. The large grains display eccentrically placed hila and lamellae lines. 72

Fissures appear as cracked lines of varying length, depth, and width. Comparison of scanning electron microscopic (SEM) and LM photographs of the same starch grain have shown that fissures usually occur inside the starch grain, although in LM they may appear to occur across the surface of the grain (Hall and Sayre 1973:121; Fannon et al. 1992:285). The hilum is considered the centre of the grain and may be visible (open) or invisible (closed) in normal transmitted light. The hilum is commonly uncoloured, but may be pink, brown, gray, etc. Lamellae are growth rings of the starch grain believed to initiate at the hilum and grow in centric layers by apposition. In normal light, lamellae appear similar to tree rings radiating from the hilum. Lamellae may or may not be visible on a starch grain. For example, lamellae on large potato starch grains tend to be visible and lamellae on small grains such as rice are not visible with LM (Simao and Cordenunsi 2010:23). Starch grains may occur singly (simple grain) or as aggregates (when growing together, called compound grains). When grains grow closely together, they push up against each other, leaving the morphology of the grain shaped with facets.

Torrence et al. have used computer programs to test the reliability of computer- generated taxonomic assignment of starch grain photographs (2004). The photographs used were of known species from an Australian reference collection with taxonomic assignment conducted in two grain orientation views: end-on (also called centric, polar, or face view), side-on (also called eccentric or oblique view). Morphological and numeric variables were used in the computer analysis including measurements such as shape, presence/absence of lamellae, area of the polarization cross, and maximum angle of the cross. Their results are encouraging given that computer programs for starch identification could be used by archaeological researchers with no starch analysis experience. However, they do stress that it is best to take a starch population or assemblage approach to each species identification because multiple morphologies can be present (2004:530). It is important to determine which forms in the starch assemblage carry the most distinct features and are therefore most useful to archaeology. Kooyman and Kooyman (2010) have noted this may be problematic due to the fact that the most 73 distinctive starch grains in an assemblage may be relatively rare in the plant, and thus unlikely to be recovered and identified from archaeological materials. This means that a variety of starch grain forms from the species assemblage may need to be used in combination in order for a successful identification to be made.

Starch Look-alikes

Under polarised light, there are some objects that display an extinction cross very similar to starch. These include mineral and coprolitic spherulites, bordered pits, spiral wall thickenings, air bubbles, spores, pollen, coccoliths, some diatom frustules, zooplankton with calcareous shells including bivalve veligers (larvae), and ostracods. Misidentification of the item in question is usually easily avoided when viewed in normal unpolarised light. Spherulites are small (5-25 pm) rounded aggregates of radiating crystals. Like starch grains, they also posess lamellae and amorphous regions and display birefringence and an extinction cross when viewed with a polarizing lens. Coprolitic or faecal spherulites from animal dung are often recovered from archaeological sediments (Canti 1998:435). Canti (1999) showed that faecal spherulites occur in the highest concentration from ruminants grazing in alkaline (>7pH) soils, but they can also be found in ominvore faeces. The extinction cross observed in spherulites is fixed and does not rotate when the polarising lens is rotated, thus confusion with starch can be averted. Bordered pits are small porous openings found on tracheids and vessel elements in which the secondary wall arches over the pit membrane (Raven et al. 2005:592, 673). Bordered pits display an incomplete extinction cross that will rotate, but in normal light, bordered pits are seen to occur embedded in tracheids and vessel elements. Cellulose annular wall thickenings from tracheids and vessel elements appear as damaged or modified starch grains in both normal and cross polarized light (Torrence and Barton 2006: Plate 50). Haslam (2006b) has warned that conidia fungal spores, common to soils, may easily be mistaken for starch. Conidia extinction crosses rotate, and are less than 5um in size. Fungal hyphae with attached spores can appear as a 'string of pearls' that appear to have extinction crosses in cross 74 polarized light (Torrence and Barton 2006: Plate 48). In a two year taphonomic experiment by Barton, fungal bodies were found surrounding sweet potato starch in tool residue extracts from flakes from both buried and surface contexts (2009:133,135). These appear to be examples of the 'string of pearls' look of fungal hyphae with spores, although this was not explicitly stated.

The following kinds of 'non starch' are more of a concern when artifacts are retrieved from sites near water bodies, and are included here for completeness. Some phytoplankton and zooplankton with crystalline calcareous protective shells exhibit a maltese cross. Coccoliths are the calcium-rich plates or structures that form the protective test (shell) around unicellular planktonic algae called coccolithophores (Loy 2006:124). Bivalve veligers display the maltese cross in cross polarized light, however this cross is not visible when viewed 'on edge', i.e. when rotated (Johnson 1995:142). Ostracods and echinoderm larvae also display the maltese cross (Johnson 1995:142, 144). Spicules from sponges and echinoderms such as sea urchins also display birefringence, although these are unlikely to be encountered in tool residues from non-aqueous site environments and are distinctly linear, unlike starch grains.

Microscopic Indicators of Human Activity

There are a number of residue types that can be identified with light microscopy and used as indicators of human activities in the past. These include but are not limited to: starch, phytoliths, pollen, calcium oxalate crystals, wood tissues, resins or hafting residues, diatoms, fungal hyphae, animal tissues, vertebrate proteins, hair, feathers, parasite eggs, fatty acids, amino acids, and ancient DNA. Only those residues which were found in the tool extracts and are relevant this study are described here. 75

Plant Micro-remains Vascular Tissues Identifying wood and herbaceous plant material microscopically in residues from stone tools can provide good evidence for prehistoric woodworking and/or plant processing activities. These vascular tissues can also help identify the tools that were employed. When examining vascular tissue residues, the individual cells are often highly fragmentary and distinctive anatomical features may be obscured. There are two types of tracheary elements that conduct the water and minerals of the xylem: tracheids and vessel elements. These are the most common and important vascular tissue residues for archaeologists to identify. During the evolution of plant vascular tissue, xylem conducting trachied cells were the first to evolve and are found in the seedless vascular plants, seed-bearing gymnosperms, and flowering angiosperms (Raven et al. 2005:516). The vessel element, a more derived and specialized cell type, evolved in all angiosperms but only in the order Gnetales (tropical evergreen trees) group of the gymnosperms. The gymnosperms (including all conifers) have trachieds as the only xylem conducting cells. When vessel elements are observed in archaeological residues, they are always from angiosperm plants.

Tracheids and vessel elements have inner secondary walls that are lignified and strengthened by annular thickening, helical thickening (spiral thickening), scalariform thickening, and/or reticulate thickening (Bidlack and Jansky 2011:58; Mauseth 2009:105-106; Raven et al. 2005:517; Simpson 2010:75). These morphological traits are often visible with light microscopy. Cells of the rays, the horizontally oriented xylem of the plant, may also be encountered in residues, including ray parenchyma cells and also ray tracheids.

There are some helpful guides available for wood identification (cf. Core et al. 1976; Friedman 1978; Hoadley 1990; Panshin and de Zeeuw 1980). However, these guides are not designed for the examination of microscopic archaeological wood residues that are often degraded. Wood identification guides focus on bulk 76 wood classification, and generally consider large pieces of wood that can be sectioned in several views, not the single and often fragmentary tracheids, vessel elements, fibers, etc. that are typically found in residue extractions. Nevertheless, they provide good comparative photographs and descriptions of morphologies that can sometimes be identified in wood residues. Wood guide books show that there are morphological microscopic differences between softwoods (conifers, gymnosperms) and hardwoods (angiosperm trees), and these differences can be identified in wood residues.

The wood of softwoods (coniferous gymnosperm trees) is mostly composed of tracheids, whereas tracheids are less frequent in hardwoods and can be challenging to differentiate from the parenchyma (Hoadley 1990:14,28). The tracheids of all conifer softwoods have bordered pits, and this is one of the most easily identifiable and commonly encountered archaeological residues (Hardy and Garufi 1998:181). Hardwoods also posses bordered pits, but those of softwoods are much larger, more common, and structurally complex. The softwoods have no fibers or vessel elements and may have resin canals. Cross-field pitting occurs where longitudinal tracheids and ray parenchyma (may also include ray tracheids in the ray) intersect. The type of cross-field pitting seen can be diagnostic to conifer group (Friedman 1978:26; Hoadley 1990:26; Panshin and de Zeeuw 1980:150). Cross-fields are also visible in eudicots (Brian Kooyman, pers. comm.). One difficulty that may be encountered with trying to use diagnostic cross field pitting type for wood identification is that guides apply to the large thin-walled cells of the earlywood formed each spring, not the latewood. Latewood tracheids are thick- walled and narrow in diameter, thus the cross-fields with ray parenchyma cells are narrow (Hoadley 1990:27). Wood tissues with cross-field pitting that show up in residue extracts may be latewood, and therefore unidentifiable by cross-field pitting guides. It should also be noted that cross-field pitting types are altered by drying and charcoalification, and this can confuse identification of archaeological and fossil woods (Gerards et al. 2007). 77

From my limited experience, the most distinctive features of conifer softwood tissue to look for while scanning residue extracts under a microscope are: 1) large bordered pits on tracheids which display strong birefringent maltese crosses under the polarizer, 2) fragments of distinctive rectangle checker look of cross-field pitting of longitudinal tracheids and ray parenchyma, and 3) spiral thickening around tracheids (this feature is rarely encountered).

Hardwood (angiosperm trees) tissue from the tree is mostly composed of libriform fibers. Libriform fibers are long, narrow cells with moderate to thick walls that have small simple pits or borders too narrow to be apparent with transmitted light microscopy (Frost 1929:261; Panshin and de Zeeuw 1980:182-183). Tracheids occur less frequently in hardwoods (Hoadley 1990:29). Vessel elements are the largest cells in hardwoods (Hoadley 1990:31) and have relatively large diameters. Vessel elements have bordered, half-bordered or simple pits and have either simple or scalariform perforation plates at their ends (Panshin and de Zeeuw 1980:175; Raven et al. 2005:517). Spiral or helical thickening in vascular tissues is much more commonly observed in temperate zone hardwoods than in tropical hardwoods or coniferous softwoods (Core et al. 1976:76).

Features that I have used to identify hardwood fragments while scanning microscope residue extracts include: 1) intervessel pitting on the lateral walls that join vessel elements (pitting is dense and often 'honeycomb' in appearance, pits are bordered and placed close together on the vessel element, in contrast with the more widely-spaced bordered pits present on softwood tracheids), 2) vessel elements tend to be wider and shorter than hardwood or softwood tracheids, 3) spiral thickening may or may not be present on vessel elements, and may also be present on fibers (Core et al. 1976:78), 4) long slit like bordered pits on long and thin libriform (xylary) fibers and fiber tracheids (described in Panshin and de Zeeuw 1980:183). 78

Identification of hardwood tissue is complicated by herbaceous plants, which have tracheids and vessel elements of the vascular tissue that can appear similar to the hardwood trees (Brian Kooyman, pers. comm.). This fact makes differentiation of hardwood versus herbaceous plant in archaeological residues very difficult, especially since the residues are fragmentary and all the features of the vascular tissue cannot be viewed.

Microscopic wood residues were identified on experimental stone tools by Hardy and Garufi (1998). They found that most wood residues examined in situ on tool surfaces had no visible diagnostic anatomical features (1998:180). Their difficulty in identifying features of wood residues may have been in part due to use of in situ viewing. Identification of vascular tissues is probably improved by removing the residues from the tool surfaces for viewing on microscope slides. Wood residues in the form of distinctive phytoliths and fibers were found by Dominguez-Rodrigo et al. in situ on three Acheulian handaxes attributed to early humans in Peninj, Tanzania (2001:297). The authors suggest the handaxes were likely used for chopping wood and suspect the wood is Acacia.

Phytoliths Phytoliths, or 'plant stones' are hard microscopic bodies that form in plants that contain calcium, silica or other crystalline-forming minerals. Silica-based phytoliths are the main type of phytolith archaeologists study. Plants uptake dissolved silica in water that is deposited in cell walls, cell lumens, or in extracellular locations, leaving behind solid silica (Si02.nH20) when the organic part of the plant disintegrates. Most of the dissolved silica taken up by the plant precipitates out at the site of transpiration -the leaves- where most phytoliths are formed, although they can also be found in the stem. The term 'phytoliths' includes amorphous silicified plant material, articulated epidermal plant cells, guard cells of the stomata, spongy mesophyll cells, silicified tracheids and vessel elements, single cell hairs and multicellular trichomes with base cells. Ferns, horsetails, angiosperms and gymnosperms all contain silicified cells (Piperno 2006b:7; Sangster et al. 2001:85). 79

Phytoliths can be diagnostic to family, genus and sometimes species. Phytoliths are very resistant to degradation, and thus are useful candidates for archaeological reconstruction of ancient plant use (Pearsall 2000:355). Phytoliths are likely to be present in archaeological soils and sediments, hence non-cultural phytoliths may be present in lithic residue preparations unless steps in the laboratory are taken to remove them. Phytoliths are used as indicators of what foods were eaten by people in the past, which tools were used for plant processing, spatial analysis in a site (eg. activity areas), and paleoenvironmental reconstruction. Plant processing stone tools can develop a bright lustrous sheen, gloss, or polish where phytoliths or amorphous silica has accumulated on the tool working edges (eg. Hardy 2004:554). Phytoliths distinctive to various taxa levels have been recovered from obsidian artifacts (Kealhofer et al. 1999:542), stone grinding implements (Pearsall et al. 2004:432) and pottery (Thompson 2006:86).

Pollen In the context of archaeological investigations, plant pollen is usually used as an indicator of paleoenvironment and paleovegetation. Palynological analysis has also been used as a tool to track natural climatic shifts through time, or the timing of human impacts such as deforestation on the local landscape. However, pollen found on artifacts has also been identified as a residue resulting from human use (Linda Scott Cummings, pers. comm.). For example, Dickson found lime tree (Tilia cordata), perhaps meadowsweet (Filipendula ulmaria) and other species of pollen in a Bronze Age Beaker vessel. The pollen found in the ceramic residues was interpreted by Dickson as a drink containing honey which could have been mead or sweetened ale (1978). Palynology has also been applied to locating the origin of clay artifacts, akin to its use in forensics for tracing the locations that came into contact with murder victim's bodies. Hu et al. examined the pollen and spores extracted from a terracotta warrior and a horse from the Qin Shihuang Mausoleum which are 2200 years old, aiming to determine the construction provenance (2007). They compared the pollen and spore profiles of the terracotta figures to the Qin Dynasty aged soil samples inside the mausoleum. Hu et al. concluded that the the 80 horse was produced close to the mausoleum since arboreal pollen grains dominated the profiles from the horse and soil samples. Conversely, they determined that the soldier figure had been produced away from the site, as its pollen profile was dominated by herbaceous plants that did not match the site soils (2007:1155).

Moore et al. provides a good key to pollen and spore types that is well illustrated with SEM micrographs (1991:86). Owens and Simpson also provide helpful descriptions and accompanying microgrpahs of conifers native to British Columbia (1986). Comparing LM micrographs of archaeological pollen to the clear SEM micrographs published in books and journals can prove challenging. However, if multiple LM pollen grain micrographs at various depths of field are taken, chances of successful identification are improved.

Fungal Micro-remains Hyphae and Spores With light microscopy, fungal hyphae usually appear as smooth, branched, and slightly wavy thread-like filaments that may or may not have cross walls. Fungal hyphae have been identified in residue extracts from stone tools since the 1980s. Loy and Nelson commonly observed fungal hyphae and sometimes root hairs in their residue extracts taken from lithics. They consider these remains to be soil contaminants (1986:184). Currently, fungal hyphae and spores encountered in artifact residues are regarded as contaminant material originating from the soil (Lombard and Wadley 2007a: 162), and most soils where artifacts are buried are likely to contain fungi. However, various fungi and lichens were eaten and used in the past and it is therefore plausible that they represent use residues on stone tools or other artifacts. Alternatively, the types of fungal bodies found on stone tool surfaces may be indicative of the soil environmental context. To my knowledge, no experimental work has been carried out on artifacts to understand the taphonomy and preservation of fungal use residues. Identification of fungi and lichens (symbiotic organisms composed of a fungal partner plus an algal or 81 cyanobacterium component) in archaeological residues may be hampered by the need to obtain reproductive structures and/or spores to determine the fungal taxa.

Most spores encountered in residue extracts likely originate from fungi, although it is possible to also encounter spores from ferns, bacteria, and algae. The author is not familiar with spore identification and this was not seriously attempted.

Animal Micro-remains Cells and Tissues Animal residues including cells and tissues have less organized structure than microbotanical remains and are difficult to identify morphologically with LM. A particular problem for animal cell identification is that they lack a rigid wall like plant cells that better retains cell form. Wadley et al. (2004:1499), and Lombard and Wadley (2007a:156) have noted that plant and animal residues can be confused sometimes, for example plant cellulose appears similar to some faunal tissues because both are birefringent in cross polarized light. Due to the extreme difficulty of identification of animal cells and tissues by morphological means, they might have been missed as a residue category. The author feels immunological or chemical techniques are probably necessary to successfully identify microscopic animal residues (discussed in the Summary of Results section of Chapter 6: Results).

Feathers Feather barbules can be found in residue extractions. Researchers have demonstrated that plumulaceous (downy) feather barbs have characters that can be diagnostic to bird order, family and even species (Dove and Peurach 2002:51). Dove and Koch (2010) provide an excellent beginners guide to microscopic feather identification for forensic scientists. This source was consulted to help identify the feather fragment recovered in this thesis research. Each bird feather has two kinds of barbs: smooth interlocking pennaceous barbs, and the non-interlocking 82 plumulaceous barbs (downy barbs) that help insulate the bird's body (Dove and Koch 2010:21).

Residue Taphonomy

Residue taphonomy with regard to archaeology is a new area of study and most information available draws from the biological sciences and food sciences. All organic matter in plants and animals eventually decomposes into carbon dioxide and water (Jenkinson 1981:506). Indeed, this inevitable organic breakdown applies to potential food residues and is hastened by the metabolic activity of decomposers: fungi, protozoa, bacteria, and invertebrate animals such as worms, arthropods and molluscs (Beck 1989:34). There are two significant taphonomic issues that require development in residue analysis: 1) understanding residue preservation and modification, and 2) understanding contamination sources. The conditions under which ancient organic residues will preserve in sediments and on different types of substrates (eg. ceramics, wood, fire cracked rock, obsidian, basalt, and human teeth) is foundational to the residue analyst's assumptions and conclusions. It is unknown if residue saturation of substrate binding sites occurs during the first contact with a food, or the last. It is also not known if older residues are continually replaced by newer ones, or if the residues represent the total mix of all foods that have come in contact with the ceramic pot, lithic tool, groundstone bowl, wooden spoon, etc. (Barnard et al. 2007). Basic experimental work that documents how decomposition of specific residues proceeds in different soil environments is also necessary to bring more credence to residue analysis. The second major issue that requires attention is identification and control of sources of contamination that can enter authentic archaeological residues. Contamination of artifact residues may originate from the soil, wind, plants, animals and humans. According to Locard's exchange principle, when any two items come into contact, there will be a transfer of material (Bruier 1976:478). Assuming Locard's theory is correct, this concept is truly pertinent to residue analysis research design and 83 contamination considerations. In this section, I will discuss contexts of residue preservation, and the contamination of archaeological residues.

Residue Preservation Stone Tools There is good reason to believe that starch grains and other residues are physically protected and can remain intact in stone tool use wear features such as striations, pits, microcracks, and depressions in porous or coarse grained rock types. Cracks and crevices in the stone tool surface may provide a protected environment that harbours microscopic residues. A study conducted by Shanks et al. (2001) demonstrated that usable archaeological residues, such as proteins and DNA from white blood cells, although invisible, can be present beneath the stone tool surface in microcracks even after washing with water. They conducted an experiment with 45 obsidian microblades, soaking some tools in cow blood containing 1) leukocytes (protein) labeled with 5-carboxyfluorescein diacetate succinimidyl ester (CFSE), and 2) DNA stained with 4,6-diamidion-2-phenylindole (DAPI). Another set of tools was soaked in cow blood that was labelled with protein-staining CFSE only (this set of tools was later stained with DAPI in-situ to see DNA). Using confocal microscopy, the lablelled DNA and leukocytes were indeed found in microcracks of the microblades. The labelled DNA from cow blood residues penetrated to a median of 44 pm into microcracks (91 microcracks examined total). Shanks et al. also showed that fluorescent-labelled latex beads of sizes 10 pm entered microcracks easily, and the beads that were 20 pm were not retained in the microcracks after washing the microblades. Since microcracks are very tiny and narrow, residues of 10 pm enter into cracks by capillary action (2001:968).

Organic residues that dry quickly seem to facilitate long-term preservation (Barton 2009). In this desiccated state, the residue is hydrophobic and less vulnerable to hydrolysis. A protected micro environment in the tool surface may also shield starch from the enzymatic action of microorganisms. Kooyman et al. showed that 84 blood proteins can be detected on archaeological stone tools at least as old as 5600 years by using cross-over immunoelectrophoresis (CIEP) (1992). These protein residues remain intact through normal soil chemical and biological processes, artifact washing and cataloging (1992:268). Barton (2007) and Croft (2009) also have shown that it is possible to extract food plant starch grains from museum-stored archaeological tools of wood and ground stone. This starch is present in spite of possible post-excavation or surface contamination and museum curation procedures such as washing, repeated handling, labeling, etc. Many studies have been carried out on materials that have been washed. "Washing, labeling and sub-optimal curation methods do not, of necessity, destroy the potential for residue analysis..." (Loy 1994:96). The above authors agree that it is possible for ancient use-related residues to preserve on lithic tools in spite of natural taphonomic factors and artifact handling and preparation that occurs during curation.

Starch grains are a good line of evidence to determine plant and tool use on the Canadian Plateau due to their resistance to decay, sometimes persisting for thousands of years (Loy et al. 1992; Piperno and Hoist 1998; Piperno et al. 2000, 2004; citations within Barton 2007). There have been reports of very old starch found on stone tools, lending support to the belief that starch can preserve for many thousands of years. Van Peer et al. reported phytoliths and starch grains were recovered from two quartzite cobble tools found at a Middle Stone Age site in northern Sudan (2003). Based on optical stimulated luminescence (OSL) dates of the site sediments, the tools are at least 180,000 years old, the authors suggesting that grass seed processing is responsible for the siliceous and starchy residues (Van Peer et al. 2003:191-192). Mozambican stone tools (n= 58) from early Homo sapiens deposits dating from 42,000-105,000 years ago were found to contain grass seed starches, 89% of which were Sorghum spp. (Mercader 2009:1681). Starch with an age of >37,170 BP from an Australian stone tool was identified from Ngarrabullgan Cave in Queensland, Australia (Fullagarand David 1997:141). Starch has been found on tools from Upper Paleolithic sites in Italy, Russia and the 85

Czech Republic. These sites have been AMS dated to 30,000 calibrated BC (Revedin et al. 2010:18815).

Soils A diverse range of soil organisms break down organic residues such as proteins, lipids and carbohydrates in the soil. Starch particularly does not preserve well in sediments. When compared with other saccharides, the rate of starch decomposition in the soil is slower than easily digested glucose, but more rapid than cellulose and lignin decomposition (Greenland and Oades 1975:220-201). Haslam identifies the major factors of starch degradation in the soil as 1) properties of the soil itself including pH, temperature (freezing or fermenting soil gelatinization damage), texture and moisture, and 2) biotic elements present in the soil such as enzymes, bacteria, fungi and earthworms (Haslam 2004:1718). He emphasized that enzymes from bacteria and fungi are the most influential contributors to starch digestion and destruction/obliteration. On the other hand, there are conditions under which starch can be preserved in the soil, as Haslam goes on to state: "In the soil protective mechanisms do exist, for example, starches may be protected from enzymatic attack by the presence of soil aggregates, clays, or heavy metals" (Haslam 2004:1725).

Michael Therin performed experimental work to assess the degree of downward movement of corn starch in sediments. Therin simulated starch movement through 3 different sand particle size classes under two high irrigation regimes (representing rain forest environments) for a period of two months (1998). The vast majority of starch grains were recovered from the A horizon of each 'soil profile', on which the starch had originally been placed. The very high starch counts in the A horizon were found under all sand size classes and levels of irrigation. The results indicate that in short-term high rainfall conditions lacking the enzymatic action of fungal hyphae, microorganisms, detritivores such as burrowing insects and annelids, most starch granules do not percolate down the soil/sediment column. Additionally, the sand used in the experiments was a homogeneous texture, and 86 clay and heavy metals (which may trap starch particles) were absent, so the results may not be applicable to natural soils. Therin's work tested the movement of starch in a downward direction, however it was unknown whether starch could move laterally or upwards.

Haslam designed a cruciform shaped apparatus in order to test the idea that starch could potentially move in three dimensions (2008). Haslam used only a sand of 'medium' particle size for his experiment with similarly high amounts of simulated rainfall as those used by Therin, for 30 days. Haslam's findings were in agreement with those of Therin in that only a small percentage of the original corn starch had moved within the sediment matrix at the end of the experiment (2008:98). He found that although the soil sample taken 2 cm below where the corn starch was placed had the highest number of starch grains (n=44), starch movement did occur, with upward (n=1) and lateral (n=11) movement noted. One of the main lessons of Haslam's experiment is that starch can move away from its original deposition point, shown by starch grains being found as much as 12 cm away (2008:99). Therefore, when starch analysis is applied to archaeological sediments from cored or spatial activity areas, starch must be found in large concentrations to provide a reasonable basis for the inference of human activities. Two caveats were identified by Haslam for his experiment. Firstly, rainfall rates used in the study represented tropical rain forest environments, which may not provide a good index for assessing starch movement in other climates. Secondly, archaeological time frames cannot be replicated to understand the movement of starch.

Ancient DNA (aDNA) has also been shown to migrate downwards in non-frozen soil contexts, as tests by Haile et al. (2007) from two New Zealand rockshelters indicates. This study found modern domestic sheep DNA was present in the soil profile at a location lower than the European layer, indicating downward migration. However, no upward movement of ancient Moa (a group of extinct large flightless birds endemic to New Zealand) aDNA was found. 87

Special Contexts Direct evidence of foods eaten by past peoples can be encountered in special contexts including stomach contents, dental calculus and paleofeces. The successful extraction of dental calculus from fossil teeth is an exciting new path to understand the botanical aspects of past diet and behaviour. Researchers are using this method to find direct microscopic evidence of the consumption of plant foods that contain starch and/or phytoliths. Dental calculus is the precipitated mineralization of calcium phosphate deposited and built up on the surface of the teeth, the formation of which traps microscopic food particles. Although accretion rates of dental calculus are currently unknown, the deposits probably represent at least the last several years of the individual's diet. Henry and Piperno showed that extracting dental calculus from human teeth requires a very small amount of calculus and can be carried out nondestructively (2008). They examined one tooth belonging to each of five individuals from the middle Holocene site of Tell al-Raqa'i, Syria, and found 4700 BP starch of barley and possibly oats. In addition they recovered unidentified (but non-grass seed) starch, this comprising about 88% of the total starch from the five sampled teeth. This is a surprising result about the past peoples of this area, since a heavy dietary reliance on cereals is the current accepted belief. Starch has also been recovered from ancient Peruvians dated between 8210 and 6970 14C years BP, including species of squash, peanuts, and beans (Piperno and Dillehay 2008:19622). Cultivated forms of squash and peanut starch were also identified in the assemblage. Nenderthal diets also appear to have contained starchy foods. Direct evidence of plant food consumption has been seen in dental calculus from two neanderthals from Shanidar Cave, Iraq and one neanderthal from Spy Cave, Belgium (Henry et al. 2010). Henry et al. (2010) found date palm, legume, and grass seed starch in the calculus of these individuals and detected distinctive damage patterns on some starch grains that indicated that the plant foods were cooked.

Along with dental calculus, human paleofeces provide direct evidence of plant foods consumed, albeit over a shorter time span (representing a snapshot of the 88 diet in 1-3 meals, instead of several years). Studies by Horrocks (2005) have shown that starch grains and phytoliths can be recovered from desiccated human coprolites. Horrocks et al. studied coprolites from several sites in New Zealand (2004a). Many starch grains appeared to be unmodified and were identified to species, including sweet potato and bracken fern. Birefringence was maintained in the starch grains, however the extinction cross was lost in many, perhaps a result of cooking, but perhaps from the action of gastrointestinal enzymes. This finding is relevant to the taphonomy of starch, since we now know that at least a portion of starch that is consumed is not digested in the gastrointestinal tract and is passed into the feces. However, Sajilata et al. (2006:12) state that starch is rarely found in human and animal feces, starch being absorbed in the small intestine, and fermented by microflora in the large intestine. Coprolites can contain protein residues of the animal of origin as well as the animals consumed. Newman et al. conducted experiments with zoo animal scats that established that it is possible to identify the proteins of meats that have passed through the digestive system with cross-over Immunoelectrophoresis (CIEP) (1993:95). Newman et al. applied the CIEP technique to coprolite samples from archaeological sites and found six of seven tested positive for human and four tested positive for pronghorn (1993:96).

Modified Starches Any assemblage of starch granules recovered from archaeological tools or pottery will likely contain at least some modified starches- starches that have undergone physical changes in their structure caused by natural taphomomic or human agents. Normal raw or 'native' starch grains may become damaged when exposed to factors such as dehydration (Marousis and Saravacos 1990), moisture, freeze/ thaw (Krok et al. 2000; Szymonska et al. 2003), heat (Henry et al. 2009), high pressure (Tomasik 2002, cited in Sujka and Jamroz 2007:112), acids (Hoover 2000; Niemann and Whistler 1992), mechanical grinding such as milling, or enzymatic degradation from fungi and bacteria (Sujka and Jamroz 2007). Once the structural integrity of a native starch grain has been modified, it may loose the basic optical characteristics of starch grains and thus may not be identified by 89 microscopic analysis (see Figs. 4-3 and 4-4). Modified starch can also fail to effectively take up the staining agent I2-KI, commonly used to test archaeological samples for starch. However, Congo red stain can be used to identify damaged starch grains that are gelatinized or fragmentary (Lamb and Loy 2005; Sterling 1968).

Fig. 4-3. This large ruptured starch grain from Fig. 4-4. Note the complete loss of birefringence water soaked macaroni pasta has been and the extinction cross of the of the large damaged, probably modified by physical ruptured grain when viewed in cross polarized milling, water and temperature changes, pH light, 400x, image digitally lightened. changes, enzymatic attack, or other methods of commercial processing. Tearing of the grain edge is apparent in normal light, 400x magnification.

The morphology and chemical properties of native starch granules vary by botanical origin. This variation in starch grains means each type faces different susceptibilities to physical changes. Factors that affect susceptibility to physical change include: size (large grains are preferentially altered/destroyed), overall shape in three dimensions, topography of granule surface, presence of pores or channels (high porosity gives more surface area for microbial enzymes to attack the grain), presence of nonstarch constituents (eg. proteins, lipids), moisture content, ratio of amylose to amylopectin in the grain, amylose and amylopectin fine structure, and arrangement of crystalline and amorphous regions (Huber and BeMiller 2010:149). Starch grain size, surface topography, and porosity are known to greatly affect starch grain preservation. For example, Messner and Schindler 90 conducted sets of laboratory and actualistic experiments cooking Arrow Arum (Peltandra virginica). They found that "The grains most resistant to the [earth oven] cooking process were the smaller spherical to hemispherical forms which averaged 8um in size. Each of these grains exhibits a depressed central area. Since this trait is not present in the uncooked rhizome starch it is likely the result of heat damage,". They go on to note that these damaged grains may not be taxonomically valuable for identification (2010:332). Natural processes as well as mechanical treatment of starch grains creates surface pores.

Modified starches are archaeologically significant because they can signal how foods were processed in the past (Babot 2003). There are a number of studies that have shown particular modifications to starch grains result from specific processing techniques. Some of these include: the fermentation of wheat starch to make beer, causing more and deeper holes to appear on the grain surface (Samuel 1996); milling, which causes a rough appearance, cracks, and a partial loss of the polarization loss from pulverization (Niemann and Whistler 1992); grinding, which causes fragmentary grains (Babot and Apella 2003); boiling with water, which can causes swelling, distortion, and loss of birefringence (Henry et al. 2009). The survival of cooked starch grains is rather remarkable since cooking makes starch more susceptible to organic decay (Samuel 2006). In spite of this, Barton found that heat damaged starch can survive a long time on artifacts (2007), and resistant starch has been recovered from carbonized material (Cortella and Pochettino 1994; Ugent et al. 1986). Heat by itself does not necessarily destroy starch grains, but heat plus water or a moisture rich cooking microenvironment results in a great deal of gelatinization (Messner and Schindler 2010:333)

It seems logical that phytoliths, starch grains, pollen, and animal residues may be modified or damaged by natural taphonomic factors, just as faunal bones are modified by factors such as sun bleaching, wind erosion, scavenging, changes in pH, etc. Starch that archaeologists recover may have undergone partial digestion by insects, earthworms, and bacteria and fungi, which can cause loss of extinction 91 cross, increase or introduction of pores to the grain surface, and grains which are fragmentary. Freeze/thaw action in the sediments may act to physically erode phytoliths, pollen and starch. Potato starch grains that have been frozen in soils, dehydrated by air, or dehydrated by heat display a flattened relief, loss of lamellae, loss of extinction cross, and regions of the grains that have lost birefringence (Babot 2003:74). Although the research in this area is very scarce, it appears that modifications to the residue can occur by natural factors. It is therefore important that residue analysts do not assume that damaged residues are modified due to human actions such as milling, cooking, or fermenting.

Contamination of Archaeological Material Potential contaminants are of concern to any residue analysis. Wadley et al. (2004) and Lombard and Wadley (2007a) designed a series of four blind tests wherein the residue analyst was asked to microscopically (LM) identify plant and animal residues on stone tools prepared for the experiments. Lombard and Wadley found that"... not all residues on stone tools are associated with the use of the tool," (2007b: 19). Multiple lines of microscopic evidence may be a key to securely identifying authentic residues from human activities. They have pointed out that if a particular residue is an authentic indicator of human tool use, we should expect it to be found in combination with other related residues. Identification of plant processing is strengthened by a variety of plant residue types found together: wood, bark cells, resin or plant exudates, fiber and starch grains (2007a:159). These associated residues provide more reliable evidence for human use of lithics than one residue type alone.

There are three stages at which sources of modern starches could be introduced to the archaeological material: 1) during excavations, 2) during handling of lithics for washing, cataloging and analysis, and 3) during the extraction of residues from the tool and preparation of microscope slides. Care should be taken to reduce contamination of authentic use-related residues at every stage of artifact handling, from excavation to lab work to examination. These precautionary steps increase 92 the validity and confidence in the data obtained. Linda Scott Cummings has designed protocols for archaeological researchers aiming to recover pollen, phytoliths, starch, AMS radiocarbon, macrofloral samples or conduct Fourier transform infrared spectroscopy (FTIR) (2007). Sources of on-site contamination can include: backdirt/backfill piles, cigarettes, cigars, and chewing tobacco, dogs that shed hair and pollen, pollen from weedy plants, lunch garbage and vegetables.

Various measures can be taken to reduce contamination of archaeological residues present on artifacts and in microfossil soil samples. Before excavation, Lints and Surette (2010) recommend that a site survey be conducted to identify potential source regions of contamination, such as agricultural fields, and planned excavation unit proximity to waste disposal areas that could have leached into underlying archaeological deposits. Surface samples (modern controls) should always be taken before clearing vegetation or excavating the site (Scott Cummings 2007). Soil and sediment samples from the White Rock Springs site and offsite locations were taken to form a comparative basis for microfossil finds and will be processed in winter 2011/2012. During excavations, microfossil soil samples should be collected on days when wind is low. If this is not possible, one can shield fresh cut dirt where samples are being taken. Soil samples should be taken from archaeological features immediately, as the surface become contaminated with modern pollen within 10-30 minutes (or less) (Scott Cummings 2007). The presence of dogs on site or at camp should be noted and samples of their hair should be examined as possible contaminants. Lunch breaks should always be held away from open excavations and the site and workers hands should be washed after eating or smoke breaks (Scott Cummings 2007). Each artifact that may be involved in a residue analysis should be bagged immediately in its own new zip lock bag, along with the dirt adhering to the artifact. Any artifact cards should be placed in their own zip lock bag. Then, the bagged artifact and its bagged card can be placed together in another new zip lock bag (a total of three zip lock bags per artifact). This treatment of artifacts during excavation is in agreement with Newman and Julig's protocols for protein residue analysis 93

(1989:120-121). Barton and White also state (1993:171): "Ideally, residue analysis should be conducted on artifacts removed from the ground with as much adhering matrix as possible, using sterile plastic gloves, immediately bagged, and subsequently handled as little as possible." Tape, ink, glue, nail polish and white out for cataloging artifacts should not come into contact with the artifacts and not be used for cataloging. After excavations, artifacts should not be prepared for curation, as residue analysis should be carried out prior to any other research or tool handling (Hall et al. 1989:148). Washing any artifact that may be selected for residue analysis should be avoided. Lints and Surrette (2010) state that "Washing artifacts after they are removed from archaeological context is usually completed to either aid in the identification of such artifacts or for cosmetic purposes. This activity, especially concerning lithics, ceramics, and bone will greatly disturb or completely remove archaeological residue adhering to these particular artifacts (Livarda and Kotzamani 2006)." I agree that washing artifacts likely removes at least a portion of the archaeological residues that may be present. Removing artifacts their original bags or direct handling should be avoided until all residue extractions have been completed. If handling artifacts before residue extractions is necessary, the worker should wear powder free gloves that have been tested for starch and other modern contaminates. A different pair of gloves should be worn for each artifact handled to prevent cross contamination. A summary of excavation and laboratory artifact handling protocols suitable for residue analysis can be found in the Recommendations section of Chapter 7: Conclusions. 94

CHAPTER 5: THE TOOLS AND PROCEDURES

Description of Lithic Sample and Handling History

A sample of 106 stone tools were selected to represent all major tool types present in the White Rock Springs lithic assemblage. Tools were not checked for the presence of visible residue plaques or obvious use wear. Stone tool types included: utilized and retouched flakes, microblades, bifaces, unifaces, a macroblade, cobbles with striations, and a broken hand maul. Some of these artifacts are fragments of the complete tool. In descending order, tool material types consisted of basalt, chert, and chalcedony, with a few currently unidentified rock types.

The stone tools used in this study all date within the Late Period (ca. 4500-200 BP). Selection of lithics for residue analysis was random since residue plaques or use wearuse wear were typically not visible with the naked eye on the tools. Those tool types that were expected to contain the most plant residues were not specifically targeted for residue extraction. The main goal was maximum recovery of residues from as many lithic tools as possible. The sample represents all major stone tool classes in the total lithic assemblage from White Rock Springs. The majority of the excavated lithic remains recovered from White Rock Springs are from an area designated 'Campsite 2' (CS2). This predominance is reflected in the lithic sample, 76% of which originates from CS2 . There were also some lithic tools found at 'Campsite 1', and 19% of the lithic sample is from this area. These two areas were designated as campsites since a variety of tools and faunal remains were recovered in these locales of the site. Only 5% of the sample came from areas outside of Campsites 1 and 2, from cultural features (CF). 95

Tools bv Excavation Area (n=1Q6) CS2= 81 (76%) CS1= 20 (19%) CF18= 2 (2%) CF9= 1 (1%) Surface CF17/CF18= 1 (1%) Surface CF9= 1 (1%)

The tools examined in this study were excavated from White Rock Spring in a way suitable for typological/attribute lithic analysis. At the time of recovery, no effort was taken to ensure minimal contact with hands or excavation tools, although formal tools were bagged separately in zip lock bags as they were discovered. Sediment adhering to tool surfaces or surrounding soil layers was not specifically collected. During excavations hands were not washed after meals, so modern foods could have come into contact with the lithics as they were removed from the ground.

The 106 stone tools used were subject to lithic attribute analysis by Dr. David Pokotylo, research assistant Kyla Hynes and student volunteers at the University of British Columbia, Vancouver. Tools were not treated in a way designed to protect the residues, nor handled with care to prevent contamination at any stage. Washing, cataloging and initial typological/attribute lithic analysis was conducted on the lithics prior to residue extraction.

All lithics were washed with bare hands in tubs of regular tap water, sometimes using mesh screens. No brushes or other tools were employed to remove dirt. Lithics were air dried on paper towel-lined trays. No outside food was allowed in the lab where lithics were cleaned. The washing process may have removed a portion of archaeological residues on the lithics. Obviously, it is not desirable to lose any microscopic information that is available in trace amounts. Each lithic was examined with bare hands and characterized in a database designed by Dr. Pokotylo. Various attributes were recorded such as material type, length, width, 96 edge angle type, presence and number of platforms, among others. Bare hands that contain modern food remnants could have come into contact with lithics as they were handled. Open windows, heating or cooling vents or any other form of mechanical ventilation can introduce airborne starch and dust particles to artifacts that are not bagged.

Steps were taken to prevent modern starch contamination during the residue extraction process. These are outlined in the summary of the residue extraction procedure below, and in more detail in Appendix A.

Procedures

Residue Extraction There are several ways to remove residues from lithic tools, depending on the goals of the study. The targeted or spot approach of removing visible residue plaques by scraping with a sterile metal instrument (eg. Boyd et al. 2008) or placing a very small amount of liquid on a specific area then aspirating the dissolved residue, as performed by Barton (2007), were not practical for this study. Due to the very small concentrations of possible starch present on each tool (which were not visible), it was preferable to remove all residues (eg. Fullagar et al. 2008:161). The targeted approach to starch extraction makes sense only when clear, relatively large residue plaques can be seen on the tools being tested. Many of the tools being tested likely contain starch in frequencies too low and spread out over a wide area of the tool surface to be easily visible and pinpointed with a pipette tip. Therefore, a total immersion approach was taken based on Barton et al. (1998) and Zarrillo (2010), with some modification. The tool surface of interest was immersed in a dilute ammonia solution to dislodge all residues. Since the tools were previously washed at UBC, dirt was not abundant on their surfaces prior to residue extraction, and it was assumed that dirt adhering to the lithics would be minimal. 97

The following is a general summary of procedures used to extract and prepare residues from lithic tools for transmitted light microscope viewing. For step-by-step lab procedures used, refer to Appendix A. This procedure of starch residue extraction and slide mounting is a modification of protocol used by Sonia Zarrillo (PhD candidate, University of Calgary) (2010). Residue extractions and slide mounting took place in the Latin American Research Laboratory from May 12-27, 2010, where Sonia Zarrillo conducts her starch research at the University of Calgary.

Precautions were taken to prevent and avoid the introduction of contaminant starch, dust, woody or herbaceous fragments, etc. during residue extraction. No make-up, hair products, body powders or starch-ironed shirts that could introduce starch to the lab were worn. During residue extractions and slide preparations, a bright turquoise coloured scrub top was worn so any fibres introduced to the samples could be identified as unnatural. Three types of non-powdered gloves were tested for modern starch. Disposamed brand non-powdered gloves were found to contain maize starch and VWR brand non-powdered gloves contained starch. Mr. Clean poly gloves were found to contain no starch and were used in any direct handling of lithic tools, with a different pair being used for each tool to prevent cross contamination. A clean glass slide was laid out each week to test for airborne and counter top sources of contaminant starch. All plastic containers, glass microscope slides, plastic pipettes and pipette holders used in the extractions were vinegar pressure cooked at 15 PSI for at least 30 minutes to ensure any possible starch was gelatinized and then washed away with deionized water. At the beginning of each tool set residue extraction, all work surfaces were wiped down with starch-free cotton cloths and water and then covered with plastic wrap. Each tool had a labelled poly glove and plastic pipette for use with that tool only.

Eleven tools received a jet bath, a pre-cleaning wash treatment in which the artifacts were exposed to a small stream of deionized water from a squeeze bottle prior to residue extraction (results in Appendix B: All Residues from Lithic Sample). 98

Testing the jet bath treatment on these tools was carried out to determine if jet baths would: 1) remove contaminate starch that could be present on the tools, and 2) remove authentic residues. It was determined that an initial cleaning wash of artifacts by jet bath should not be carried out. With regard to 1), the assumption that jet baths should remove all modern surface contamination because it is not 'stuck on' the tools was incorrect. The starch grains were not formally characterized or photographed, and thus the identity of starch grains from the tool jet baths cannot be confirmed with certainty. Lab notes indicate that two polygonal grains found were probably from corn. Descriptions and drawings of the other starch grains removed in the jet baths do not appear to fit the modern starch contaminants of corn, wheat, oats, rice, potato, millet, sorghum, lentil, green pea, chick pea or mung bean, and are probably human use-related starch residues. Thus, there were some tools that received a jet bath with no contaminant starch identified, however did have archaeological starch. From these tests, it was reasoned that the use of jet baths: 1) cannot remove all contaminate starch from tools (if present), and 2) can remove authentic human use-related starch residues from tools. Thus, the author does not recommend carrying out a jet bath wash as a pre-treatment for residue analysis.

To remove residues, each lithic was placed with a gloved hand in a sonic bath for 30 minutes with 5% ammonia. For tools with about a max diameter of less than 6 cm, ammonia completely covered the lithic. For larger tools where it was not possible to completely immerse the tool in ammonia, at least the edge or surface likely to have been the working edge or working surface was covered, if not more of the tool. When sonication was complete, each lithic was held with a gloved hand and lightly washed in a deionized jet bath. This jet bath was performed to remove ammonia from the tool and dislodge any extra residue adhering on the lithic surface, ensuring that such residue was included in the sample. The liquid and residue was transferred from each container with a plastic pipette to a centrifuge tube(s). A single centrifuge tube was usually sufficient to carry all residue extracted from most tools. However, some larger tools required 2 centrifuge tubes to carry all 99 residues. The residues were centrifuged 4 times for 5 minutes at 3,000 RPM, with ammonia progressively being removed from the sample and replaced with deionized water after each spin. Deionized water was removed after centrifuge spins using a designated pipette only used with that specific tool.

Slide Preparation Two microscope slides were made for each residue sample. DNA pipettes were used to aspirate 20 pL of wet residue sediment for deposit on slide 1 and slide 2, with a new plastic pipette tip being used for each residue sample. 20 pL of pure glycerol (refractive index 1.475) was used as the mounting medium on each slide. The glycerol and the wet residue sediment was mixed with a pipette tip to evenly distribute the mounting medium on the slide. Residues were mixed with glycerol while still wet, since lamellae or growth rings are difficult to decipher if grains are completely dehydrated (Sterling 1968:143; Tester 1997:166). Since the residue sediment contained a small amount of deionized water (refractive index of 1.333), the median refractive index of the prepared slides is ca. 1.4. Each slide was sealed with a square glass slide cover, using toothpicks to spread clear nail polish at slide cover edges. A new toothpick was used for each insertion into the nail polish bottle to prevent cross-slide contamination. Slides were dried in an oven at 40-45°C for about 30 minutes to harden the slide cover seals.

Starch Grain Identification Only techniques that were nondestructive to the starch grains were employed for identification. Stains can be applied to residues on microscope slides to identify starch, as starch will take up colour. The most widely used dye technique to identify native starch is iodine-potassium iodide (I2-KI) as outlined by Loy (1994). Use of I2- Kl causes the starch grains to lose birefringence (Hall et al. 1989:146). I did not use dyes to identify starch grains but relied chiefly on starch grain birefringence and the presence and rotation of the extinction cross, although a clear 'X' was sometimes completely lacking. No dyes were used because dyes alter natural starch grain colour which may be of taxonomic discrimination value, and because 100 dyes may also interfere with the identification of other residues. When oxidising chemical reagents such as chloral hydrate-iodine, pyrogallic acid, ferric chloride, sulphuric acid, and hydrogen peroxide are applied to samples of starch, the reaction to each is distinctive. These chemicals can be used to observe characteristics such as the extent of swelling, origin of reaction and digestion time, however they destroy starch grain structure (Hall et al. 1989:144). These destructive chemical techniques were avoided. The large quantities of starch needed to be sacrificed in order to conduct chemical tests were not expected to be recovered from the sample of stone tools. Most of these chemicals also require special handling due to toxicity, hence they were not employed.

Some starches appear too much like other species to be differentiated even between genera, thus discrimination is not possible. This means this study is biased towards starches that are easily identified and not representative of the complete suite of starches that may be present on each lithic tool. Transitory starch was not targeted. Transitory starches of the leaf are typically very small, usually falling in the 2-5 pm range, however they still overlap with some storage starch sizes.

A starch slide reference collection of plant species from British Columbia and Alberta including microscopic digital photographs was consulted. Over 300 slides from British Columbian plants collected by Brian and Patrick Kooyman were prepared by Patrick Kooyman in November 2009.1 added 18 plant species to this collection in January 2010. Subsequently, Patrick Kooyman processed further plant starch slides over the period from October 2010 to May 2011. Each part of the plant that could contain food starch was selected and prepared, such as the root, rhizome, base of stem, berries and seeds. This BC/AB starch reference collection is a work in progress, with new species continually being added, as well as samples from different ecological areas being made to account for intraspecific starch variation. Characterization of the range of starch grain forms, and the nature 101 of familial, generic, and specific diagnostic types, is being undertaken by Brian Kooyman.

Archaeological starch work being carried out by Brian and Patrick Kooyman at the University of Calgary is foundational to assessing the potential of starch residue analysis in British Columbia and Alberta. They are the first to conduct a systematic study to determine which plant species are likely to be recognized from archaeological samples and be useful for understanding ancient human-plant interactions. Kooyman and Kooyman (2010) described starch patterns by family and by shape according to their observations of the BC/AB reference collection material. I have used their starch characteristics to describe starch grains wherever possible. Grain shape designations include cuneiform, bell (wavy or not base, straight or curved 'bowl' sides), polygon, disc, sphere (near sphere, slightly irregular sphere), oval (symmetric, elongate, egg-shaped, pear-shaped), and grain sizes include very small (<4.5 pm), small (4.5 pm-<5.5 pm), medium (5.5-10 pm), large (>10-17 pm), very large (>17-25.5 pm) and huge (>25.5 pm) grain sizes. The shape designations of Kooyman and Kooyman were based on two dimensional micrographs, with follow-up confirmation of the three dimensional form based on re-examination of the original prepared slides. It should be noted that Kooyman and Kooyman's starch reference collection and comparative analysis are works in progress and were not complete at the time of writing this thesis. Once their analysis is complete, the ability to identify starch grains from archaeological contexts in the Canadian Plateau will be much improved.

Anyone who has attempted starch grain identification knows that many starch grain characteristics are subtle and can be hard to observe. Morphological categories may grade into each other and may not be as distinct when archaeological material is being observed. This issue may be compounded in cases where grains are very small and/or are difficult to rotate while viewing, which can make features challenging to see. It is important to note that qualitative morphological traits are interpreted subjectively and different observers may have subtle disagreements in 102 the characterization of a starch grain. For example, what one worker would describe as 'sub-round' another worker may describe as 'oval'. Obviously, slight differences in description or opinion makes comparability between researchers work problematic. This is frustrating but working out appropriate traits and terminology is part of the development of starch identification research. There is no standardized guide for identification of key features of starch in Canada. The creation of such a comprehensive guide allowing all workers to use similar and well-illustrated descriptive terms will be beneficial.

For every tool and its residue extraction, two slides were prepared but only one for each tool was examined. If one of the two slides appeared to be a better preparation (fewer bubbles, better slide cover seal, etc.), it was the slide chosen for analysis. Scanning was comprised of ten transects across each slide at 250x magnification with a Leitz transmitted light microscope with polarizing lens. The goal of residue examination was recovery of specific information about starch residues that could indicate foods prepared and general information about other residues. Therefore, plant tissue fragments, phytoliths and fungal hyphal masses were not systematically photographed but rather noted as present/absent on each of slide. Photography of starch grains and other residues was done with a 5 megapixel Motic camera and software. No micrographs were cropped, resized or scaled, and the zoom feature of the camera was not used. When a starch grain was encountered, various characteristics were recorded. The following is the list of characteristics recorded for each starch grain found in the lithic residues:

Granule Form: simple or compound Extinction Cross: distinct, intermediate, or fuzzy. This characteristic was recorded in polar or flat view (not oblique side view) of each starch grain. Basic Shape in Three Dimensions: cuneiform, bell (wavy or not base, straight or curved 'bowl' sides), polygon, disc, sphere (near sphere, slightly irregular sphere), oval (symmetric, elongate, egg-shaped, pear-shaped) 103

Hilum: open or closed (ie. not visible). The hilum is designated as open if the hilum appears as 1) a solid dot, 2) a faint barely visible bit of colour or darkening, or 3) a ring with clear core or empty centre (N.B. this condition has been referred to as 'vacuoles' by Torrence 2006:129). From my observations and micrographs, it seems that in some cases, the hilum can appear as as solid dot as well as a ring with an empty centre, depending on the plane of focus (eg. starch 15 from tool 1560: compare the hilum as a pink solid dot to the hilum as a ring appearance in photos taken in different planes of focus, Figs 5-1 and 5-2). This same observation must have also been made by Zarrillo and Kooyman, although this was not stated (2006). In their study, the appearances of the hila differ between micrographs of reference collection Amelanchier alnifolia and the archaeological starch that was identified as A. alnifolia a grinding tool. In reference collection photos presented in Zarrillo and Kooyman, the hilum appears as a ring, and appears a solid dot in the A. alnifolia archaeological starch (2006:487-489). Hilum Placement: centric (hilum equidistant from grain edges) or eccentric (hilum not central). If the hilum placement could not be seen in normal light, its position was estimated based on the placement of lamellae and extinction cross in cross polarized light) Hilum Area: raised bump, depressed (lightened or darkened), or N/A (hilum flat) Fissures: present or absent; straight, arced, stellate, Y or X-shaped, or irregular. The location of fissures was sometimes noted as appearing on the edge of the grain, across the hilum, or radiating from hilum. Fissures are narrow, sharp-edged splits in the grain that appear as linear depressions that are slightly to markedly jagged. Using LM alone, the researcher is unable to determine whether the fissures observed on the starch grains exist on the outside surface of the grains or are in fact cracks inside the grains. A comparison of LM (which views and focuses through the whole grain) and SEM (which views only the outside surface of the grain) micrographs of the same starch grain is required to establish the location of fissures. Groove: present or absent. A groove is a smooth linear trench-like depression on the surface of the grain 104

Lamellae: present (visible) or absent (invisible) with LM at 400x magnification. Lamellae appear as concentric lines or rings when visible, (visibility of lamellae can depend on hydration of the grains in slide preparation medium) Facets: present or absent; (how many?, rounded or angled edges, banded or unbanded). Facets are the straight edges that make the grain look angular. No differentiation was made between facets that had sharp well-defined edges and corners, and those facets that had rounded corners (some authors refer to as 'pressure facets'). Surface Features: roughened or smooth, description of unique features (eg. raised ridges). Internal features of the grain can be mistaken for surface features, as demonstrated when LM and SEM micrographs of the same grain are compared (Hall and Sayre 1973:121). Examining the grain through planes of focus with LM can determine which features are internal or on the surface (Brian Kooyman, pers. comm.). Max Length: the longest diameter measurement that can be taken on the grain in flat polar view in micrometres (pm). The descriptive size classes of very small (<4.5 pm), small (4.5 pm-<5.5 pm), medium (5.5-10 pm), large (>10-17 pm), very large (>17-25.5 pm) and huge (>25.5 pm) by Kooyman and Kooyman (2010) were used. Taphonomy: deep fissuring, loss of extinction cross, very roughened grain surfaces, can be indications of starch which has undergone gelatinization from water and heat exposure, milling damage, fermentation, or other processing that modifies the physical nature starch grains Other Notes: any other features of note Fig. 5-1. In this view and plane of focus, the hilum Fig. 5-2. When rotated, the hilum of Starch 15 of Starch 15 (large grain in lower center) appears appears an open ring form. From Tool 1560, a solid dot. From Tool 1560, 400x. 400x.

A/on Starch Residues Residues in addition to starch were identified, although recorded in lesser detail on a presence/absence basis, items of interest were always photographed in normal and cross polarized light and any striking features or observations were noted. These other residues, including phytoliths, tracheids, vessel elements, fungal hyphae, pollen, mammal hair and feather parts, were identified with the aid of published micrographs and descriptions, consultation with knowledgeable professors, as well as a small modern reference collection. Non starch residues were examined to gain a baseline impression of what kinds of biotic microscopic items can be found in lithic residue samples from a Canadian Plateau archaeological site. Each type of residue found on the lithic tools will be addressed in the following chapter. 106

CHAPTER 6: RESULTS

Plant, fungal and animal microscopic residues found in the lithic sample are presented below. All residue types found on each tool (such as phytoliths, tracheids, vessel elements, fungal hyphae, hair, modern threads, etc.) were collected on a present/absent basis and can be found in Appendix B: All Residues from Lithic Sample. Since starch was a major focus of this study, the starch residues found will be dealt with in some detail. Starch grain data can be found in Appendix C: Starch Morphotypes.

Plant Micro-remains

Starch Overview and General Remarks The total number of starch grains recovered and characterized from the residue analysis was 151 grains from 34 lithic tools (7 acute-angled utilized flakes, 4 flakes with steep-angled marginal unifacial retouch, 4 steep-angled utilized flakes, 2 biface fragments with longitudinal splits, 2 cobbles with linear striations, 2 proximal end microblade fragments, 2 flakes with acute-angled marginal unifacial retouch, 2 flakes with acute-angled marginal bifacial retouch, 2 flakes with steep-angled marginal bifacial retouch, 2 acute-angled biface end fragments, 1 acute-angled biface medial fragment, 1 medial portion microblade fragment, 1 steep-angled formed uniface, 1 broken maul with flat base). 66 of the 151 total of starch grains (44%) were identified as modern contaminate corn and natural contaminate wheatgrass starches. 34 out of 106 processed stone tools were found to contain one starch grain or more, or 32% of the lithic sample. The goal was to remove all residues from the surface of each stone tool. Amount of starch found per tool was counted in a controlled fashion, but this does not mean that starch grain counts from all tools are comparable, given that each tool is a different size, shape, and material type. For heuristic purposes, I have lumped detailed tool types into three major categories based on their working edges: steep, acute or groundstone 107 surface. I see (cf. Kooyman 2000) steep edge tools likely to be used in a scraping action, acute-angled likely to have been used for cutting, slicing or peeling, and cobble/groundstone tools likely to have been used in a pounding/grinding capacity, although these categories remain to be confirmed by use wear analysis.

Please note that all micrographs were taken with a digital camera and a light microscope, taken at 400x magnification unless otherwise noted. No micrographs were cropped. All micrographs shown here were resized to 7.5 cm width x 5.62 height cm, and were constrained so the original dimensions of the files were maintained.

Starch Data Collected Refer to Appendix C: Starch Morphotypes, for all starch grains found, characterized by lithic type.

Density of Starch by Tool A total of 151 starch grains (including starch grains identified as contaminate starch) were found in the residue extracts from 35 lithic tools. The following list outlines which tools had starch grains found in their residue extracts and in what quantities.

7 Acute-analed utilized flakes 6352 EX1 SL1 CS2= 1 6354 EX1 SL2 CS2= 1 4246 EX1 SL2 CS2= 1 10520 EX1 SL1 CS2= 1 3824 EX1 SL2 CS2= 3 2433 EX1 SL1 CS1= 1 19001 EX1 SL1 CS2= 1 Total Starch Grains= 9 4 Flakes with steep-angled marginal unifacial retouch 6114 EX1 SL1 CS2= 4 starch 9995 EX1 SL1 CS2= 12 1560 EX1 SL1 CS1= 18 5364 EX1 SL1 CS2= 13 Total Starch Grains= 48

4 Steep-angled utilized flakes 13577 EX1 SL2 CS2= 2 starch 10522 EX1 SL2 CS2=1 11355 EX1 SL1 CS2= 1 2178 EX1 SL1 CS1= 5 Total Starch Grains= 9

2 Biface fragments with longitudinal splits 11356 EX1 SL1 CS2= 2 starch 13575 EX1 SL1 CS2= 1 Total Starch Grains= 3

2 Cobbles with linear striations 3674 EX1 SL2 CF18= 4 starch 3699 EX1 SL2 CF9= 2 starch Total Starch Grains= 6

2 Proximal end microblade fragments 6377 EX1 SL1 CS2=1 starch 111368 EX1 SL1 CS2=1 Total Starch Grains=2 2 Flakes with acute-analed marginal unifacial retouch 13574 EX1 SL1 CS2= 1 starch 9991 EX1 SL2 CS2= 2 Total Starch Grains= 3

2 Flakes with acute-analed marginal bifacial retouch 3694 EX1 SL2 CF 18= 6 starch 1790 EX1 SL2 CS1= 38 starch Total Starch Grains= 44

2 Flakes with steep-angled marginal bifacial retouch 18997 EX1 SL1 CS2= 1 starch 3830 EX1 SL1 CS2= 6 Total Starch Grains= 7

2 Acute-angled biface end fragments 2417 EX1 SL1 CS1= 1 starch 13569 EX1 SL2 CS2= 1 Total Starch Grains= 2

1 Acute-angled biface medial fragment 1919 EX1 SL1 CS1= 1 starch Total Starch Grains= 1

1 Medial portion microblade fragment 11366 EX1 SL1 CS2= 2 Total Starch Grains= 2

1 Steep-angled formed uniface 17661 EX1 SL1 CS2= 1 starch Total Starch Grains= 1 110

1 Broken maul with flat base 3651 EX1 SL2 CF9= 13 starch Total Starch Grains= 13

1 Acute angle CS2 (tool type unknown) 9866 EX1 SL1 CS2= 2 starch Total Starch Grains= 2

Grand Total Starch Grains=s 151

Lumped Tool Categories: 20 Acute-angled tools= 68 starch grains (45%) 12 Steep-angled tools= 64 starch grains (42%) 3 Cobble/maul tools= 19 starch grains (13%)

A total of 85 starch grains that were not identified as contaminate starch were found from 26 lithic tools. The density of non-contaminate starch per tool is presented below. 26 of 106 tools processed (25%) contained non-contaminate starch grains in their residue extracts.

4 Acute-angled utilized flakes 6352 EX1 SL1 CS2= 1 starch 6354 EX1 SL2 CS2= 1 3824 EX1 SL2 CS2= 1 2433 EX1 SL1 CS1= 1 TOTAL STARCH=4 111

4 Flakes with steep-angled marginal unifacial retouch 6114 EX1 SL1 CS2= 4 starch 9995 EX1 SL1 CS2= 9 1560 EX1 SL1 CS1= 9 5364 EX1 SL1 CS2= 9 TOTAL STARCH=31

3 Steep-angled utilized flakes 13577 EX1 SL2 CS2= 2 starch 11355 EX1 SL1 CS2= 1 2178 EX1 SL1 CS1= 3 TOTAL STARCH= 6

2 Biface fragments with longitudinal splits 11356 EX1 SL1 CS2= 2 starch 13575 EX1 SL1 CS2= 1 TOTAL STARCH=3

1 Cobble with linear striations 3674 EX1 SL2 CF18= 3 starch TOTAL STARCH= 3

2 Proximal end microblade fragments 6377 EX1 SL1 CS2=1 starch 111368 EX1 SL1 CS2=1 TOTAL STARCH=2

2 Flakes with acute-anoled marginal unifacial retouch 13574 EX1 SL1 CS2= 1 starch 9991 EX1 SL2 CS2= 2 TOTAL STARCH= 3 112

1 Flake with acute-angled marginal bifacial retouch 1790 EX1 SL2 CS1= 14 starch TOTAL STARCH= 14

2 Flakes with steep-analed marginal bifacial retouch 18997 EX1 SL1 CS2= 1 starch 3830 EX1 SL1 CS2= 6 TOTAL STARCH=7

1 Acute-angled biface end fragment 13569 EX1 SL2 CS2= 1 TOTAL STARCH= 1

1 Medial portion microblade fragment 11366 EX1 SL1 CS2= 2 TOTAL STARCH=2

1 Steep-angled formed uniface 17661 EX1 SL1 CS2= 1 starch TOTAL STARCH= 1

1 Broken maul with flat base 3651 EX1 SL2 CF9= 7 starch TOTAL STARCH=7

1 Acute angle CS2 (tool type unknown1 9866 EX1 SL1 CS2= 1 starch TOTAL STARCH= 1 113

Grand Total Non-Contaminate Starch Grains- 85

Lumped Tool Categories: 14 Acute-angled tools= 30 starch grains (35%) 10 Steep-angled tools= 45 starch grains (53%) 2 Cobble/maul tools= 10 starch grains (12%)

Morphotypes Starch grains were grouped together according to morphological and size traits held in common. A morphotype group was designated if a number of starch grains had characteristics which were found to be the same. The morphotypes defined here are based on major morphological similarities, and there were sometimes small morphological differences between starch grains assigned to the same group. Slightly more inclusive groupings may be useful in helping to identify species, since many plants tend to possess a variety of starch forms. These small differences in morphology within each morphotype may represent the variation present in one species, or alternatively, each morphotype may encompass more than one species. In the following, each of the seven morphotype groups found will be described, illustrated, discussed and identified where possible.

Morphotype A 'Big Lenticular Disc with Equatorial Groove' (n=27 grains) Granule form: simple Extinction cross in polar flat view: fuzzy (can be very faint or weak), dark arms of X very wide and diffuse (Fig. 6-9). A few grains displayed a small extinction cross in the hilum area within the larger cross of the whole grain (these were always the grains that appear to have a 'donut' or ring around hilum area in normal light). Extinction cross in side edge view: arms of X usually much more visible Basic shape in 3D: lenticular or biconvex disc (grain must be probed to see disc shape), irregular lenticular, wedge shaped, slightly jelly bean shaped Hilum open/closed: closed Hilum colour: none 114

Hilum placement: centric Hilum area raised bump/depressed: usually N/A. It was observed that some grains have a distinct 'donut' or ring around the hilum area. In polar or face view, the hilum area of a grain can appear darkened, suggesting a raised bump or depression, or as an X within X under cross polarized light. Fissures: usually none, but can have small arc fissure or faint X fissures over hilum Grooves: equatorial groove seen in edge-on view when grain is rotated. An equatorial groove is a continuous groove that encircles the entire grain along its lateral edges, like the equator line around the earth. Lamellae: present or absent Facets: none Surface features: usually smooth, but fissures can make grain surface appear rough Max length (ranged: 17.1-34.2 pm (very large to huge) Max length (mean averaged: 24.2 pm Other notes: edges of some grains appear roughened or bumpy (modified starch?), deep Assuring could be due to processing (eg. grinding, cooking, etc.).

Moprhotype A grains were the most commonly encountered starch grain (n= 27), after polygon Morphotype G starch grains (n= 39). Morphotype A starch grains are lenticular disc shaped with a distinct continuous equatorial groove around the disc edge which can only be seen when the grain is rotated on its side in profile view. The grains are very large to huge in size with closed centric hila and no facets. In cross polarized light, the extinction cross is very faint, fuzzy and diffuse in polar flat view of the grain, but usually distinct when the grain is rotated on its side. Surface of the grains is usually smooth in appearance and faint fissures when rarely present are not deep. Some grains showed a 'donut' or ring around the central hilum area in normal light; this hilum area appears darkened or has its own extinction cross in cross polarized light. Examples of Moprhotype A grains are shown in Figures 6-1 to 6-9. Again, note that all micrographs were taken with a digital camera and a light microscope, taken at 400x magnification unless 115 otherwise noted. No micrographs were cropped. All micrographs were resized to 7.5 cm width x 5.62 cm height, and were constrained so the original dimensions of the files was maintained.

Fig. 6-1. Morphotype A starch grain appears as a flat disc in polar view. Max length of grain is 28.5 mti- Starch 1 from tool 1919, 400x.

Fig. 6-2. Morphotype A starch grain from an pjg. 6-3. The extinction cross of Morphotype A oblique side view. When the grain is probed, grains usually become visible when probed into the equatorial groove is visible. Starch 1 from the oblique side view in cross polarized light, tool 1919, 400x. Starch 1 from tool 1919, 400x. 116

Fig. 6-4. Morphotype A in polar view with notably roughened grain edges, max length 29.1 |jm. Starch 2 from tool 3694, 400x.

Fig. 6-5. Morphotype A in oblique side view Fig. 6-6. Same grain in cross polarized light with equatorial groove visible. Starch 2 from partial oblique view. tool 3694, 400x.

Fig. 6-7. Morphotype A starch grain 3 from tool 3694 with the 'donut' or ringed area around the hilum, max length 28.5 ^im, polar view, 400x. 117

Fig. 6-8. Morphotype A starch grain 3 from tool Fig. 6-9. Same grain showing the hilum area 3694 showing the 'donut' or ringed area around darkened in cross polarized light, 400x. the hilum, slightly rotated, 400x.

The equatorial groove along the edge of the biconvex grain is the most distinctive feature of the Morphotype A grain. Three species with this feature were reviewed as possible matches with Morphotype A: Bluestem Wheatgrass, wheat, and American Vetch. The morphological and size traits of Morphotype match well with Bluestem Wheatgrass (Pascopyrum smithii), also known as Western Wheatgrass, present in the BC/AB collection (B. Kooyman, pers. comm.). Bluestem Wheatgrass is found in the Hat Creek valley area, but Bluebunch Wheatgrass (Pseudoroegneria spicata) is common as well.

The Bluestem Wheatgrass (Pascopyrum smithii) seed starch from the BC/AB starch reference collection provided the closest match to the characteristics seen in Morphotype A starch grains. Bluestem Wheatgrass is different from all other grass species currently in the collection. Bluestem Wheatgrass starch was the only starch in the BC/AB reference collection that contained very large irregular lenticular shaped grains with equatorial grooves, major features of Morphotype A grains (see Figs. 3). At the present moment, there is only this one wheatgrass species in the collection, so it is possible that Morphotype A may be another wheatgrass species, to be determined by further work. 118

Fig. 6-10. Starch grains in polar view of Fig. 6-11. Same grains, note large grain rotated Bluestem Wheatgrass (Pascopyrum smithii) to show equatorial groove. Image by B. from the BC/AB reference collection, 400x. Kooyman. Image used with kind permission of B. Kooyman.

Fig. 6-12. Large Bluestem Wheatgrass starch grain rotated with equatorial groove, 400x. Image by B. Kooyman.

The majority of the reference collection sample of Bluestem Wheatgrass seed starch grains (n=18) were in the range of 17.5-20 |jm (very large size). The total range was 10-28.75 pm (large, very large, and huge), with an average size of 18.3 pm (B. Kooyman, pers. comm.). The sizes of Bluestem Wheatgrass grains overlap with those recorded for Morphotype A, although the Bluestem Wheatgrass grains seem to be a bit smaller than Morphotype A overall. This size difference might be explained by two possibilities. The Bluestem Wheatgrass grains in the BC/AB reference collection may have been harvested at a different time than those 119 present on the tools and thus may be smaller. A second possibility is that Bluestem Wheatgrass is a species close to, but not a perfect match with, Morphotype A.

Bluestem Wheatgrass seed starch grains display a variety of shapes. There appears to be a good deal of morphological variation in this species. However, many Bluestem Wheatgrass starch grains did have the classic lenticular shape with equatorial groove that was noted in Morphotype A grains. The Morphotype A group perhaps only accounts for this one cf. P smithii variant (lenticular, irregular lenticular) and not the others (eg. P. smithii grains that were pear shaped, wedge shaped, tear drop, etc.). It seems likely that Morphotype A represents P. smithii or a grass species closely related to P. smithii.

Why this grass is present on so many tools is an unsolved problem. The Morphotype A grains were recovered from 14 tools, including: 5 acute-angled utilized flakes, 3 flakes with steep-angled marginal unifacial retouch, 2 flakes with acute-angled marginal bifacial retouch, 2 cobbles with linear striations, 1 acute- angled biface medial fragment, and 1 broken maul base. These tools were collected from CS1, CS2, the toss zone of CF 18, and surface areas of the White Rock Springs site. Given that Morphotype A grains were found on many tool types from various site areas without apparent pattern, it can be inferred that generalized natural contamination is responsible for the presence of this group of starch grains.

The likeness in morphology between Morphotype A grains and some grains of Bluestem Wheatgrass starch is striking. This similarity in morphology and the fact that Bluestem Wheatgrass is relatively common in the Hat Creek valley area, make it likely that Morphotype A starches are of natural origin and were abundant in sediment surrounding tools. Nevertheless, it cannot be stated definitively that Bluestem Wheatgrass is the correct species identification, since Bluestem Wheatgrass, wheat, barley, rye and triticale A-type grains all have very similar size, lenticular shape and equatorial groove characteristics. Wheat, barley, rye, and triticale are modern crops that did not occur in the site area prehistorically. There 120 are probably also other unknown grass species that contain large lenticular grains with equatorial grooves, although there are none in the grass species currently in the BC/AB starch reference database. Morphotype A starch grains were not restricted to a few tool types as would be expected if they were associated with a task; the grains were present on six diverse tool types, supporting the inference that they are present due to natural contamination from the soils.

There are currently no parsimonious explanations as to why Bluestem Wheatgrass (or a grass species similar to Bluestem Wheatgrass) seed starch was present on so many tool surfaces. If Morphotype A starch grains are found in analysis of soil samples tests taking place in Winter/Spring 2012, it will indicate they are of relatively recent origin. It makes sense that the starch grains would have had to have been put on tools by breaking through the grass seeds. This brings to mind a grinding action with mortar and pestle type tools, not an assemblage that includes cutting and scraping tools. Tasks such as making grass strings on which to thread roots for storage, preparing grass for bedding, or for lining earth ovens or making mats would focus on cutting the stem part of the grass, not through the storage starch of the grass seed endosperm. Perhaps the wheatgrass was used in earth ovens and when heated, the seeds ruptured and starch from them subsequently became airborne and covered the surfaces of nearby tools. Another possibility is that Bluestem Wheatgrass (or perhaps Bluebunch Wheatgrass, Pseudoroegneria spicata) may have been a dominant species at the site in the past at the time of occupation. In this case, wheatgrass seeds would be dropped over the site after the plants flower. If people returned to the site in the next spring, they might have broken seeds open by trampling over the site. Thus endosperm starch from the seeds would be liberated into the soils and air. In this scenario, it is likely that this 'contamination' would adhere and become stuck to tool surfaces as they were being used by people for various activities. Hence, as tasks were being carried out by people, wheatgrass starch would join other residues in cracks or plaques and be in a protected microenvironment. This idea accounts for the presence of 121

Morphotype A starch being present on a variety of tool types, and also its preservation over many years (B. Kooyman, pers. comm.).

There is no mention of consumption or other use of Pascopyrum smithii ethnographically in the Canadian Plateau area. Although there is no existing post- contact documentation that shows P. smithii was used by past peoples on the Canadian Plateau, this lack of evidence does not mean its use in the past can be excluded. It might be suggested that wheatgrass seeds were a food source in the past, before the European introduction of wheat, corn, white potatoes and beans. The ethnographic record does not mention labour intensive wheatgrass seed processing; it might have been lost so early in the Contact period that no one remembered preparing or eating wheatgrass by the time of ethnographic documentation. Bluestem Wheatgrass makes excellent fodder, the hay is nutritious and stores well, and is eaten by horses (according to the Lakota and Montana Indian). The plant was also used in water to soak melon seeds to keep worms away (according to Keres Pubelo, Western) (Moerman 1998:379).

Clearly, more research is required to explain the presence of Bluestem Wheatgrass (P. smithii) on the White Rock Springs stone tools, as satisfactory interpretations are not available at this time. Upcoming soil analysis and use wear analysis of the tools examined in this study will undoubtedly shed light on the nature of starch deposition on artifacts. Taphonomic experiments with plant species local to the site area and experimentally knapped tools may help determine the pathways grass seed starch follow in order to adhere as residues on tool surfaces.

Wheat (Triticum aestivum cv) was also considered as a possible match with Morphotype A due to its prevalence in modern foodstuffs. The equatorial groove seen on Morphotype A grains is a feature present in some of the large A-type starch grains (but not the small B-type grains) of wheat (Triticum aestivum cv), barley (Hordeum vuigare cv), rye (Secale cereale cv), and triticale (x Triticosecale Wittmack) (Evers 1969:97, 1971:159; Fannon et al. 1992:284; Gott et al. 2006:40; 122

Kiseleva et al. 2005:79; Li et al. 2001:400; Li et al. 2011:57; Yoo and Jane 2002:299).

The following trait list of A-type wheat starch grains was created to compare with Morphotype A grains based on observations reported in cereal science and archaeological literature (Baldwin et al. 1997; Bowler et al. 1980; Chandrashekar et al. 1987; Evers 1969, 1971; Fannon etal. 1992; Gottetal. 2006; Henry et al. 2009; Jane et al. 1994; Kim and Huber 2008; Parker 1985; Peng et al. 1999; Sujka and Jamroz 2007; Whistler and Thornburg 1957; Yoo and Jane 2002), as well as observations of commercial wheat flour starch that the author made.

Wheat (Triticum aestivum) (Common names: Hard Red Winter Wheat, Bread Wheat, Common Wheat) large A-type Starch Grains Granule form: simple Extinction cross in polar flat view: intermediate Extinction cross in side edge view: information not available Basic shape in 3D: disc like or lenticular, biconvex, pancake appearance, not spherical Hilum open/closed: closed Hilum colour: none Hilum placement: centric Hilum area raised bump/depressed: usually N/A, but some A-type grains have a depression or raised bump in the center of each major surface Fissures: none Grooves: equatorial groove apparent on young starch grains from immature seeds. During development of the starch grain, the depth of the groove decreases as the thickness and diameter of starch grain increases. At maturity, 52 days after anthesis (flowering), the equatorial groove is seen as little more than a score with SEM (Evers 1971:159), and would thus be more difficult to decipher with LM. Equatorial groove observed on some commercial wheat starch grains. 123

Lamellae: absent, parching grains increases the definition of lamellae (Henry et al. 2009:921) Facets: none Surface features: surface smooth, pores sometimes found along equatorial groove Max length (range): information not available Max length (mean average): 10-35 (jm (Peng et al. 1999:375; Singh et al. 2003:223), 15-40 pm (Evers 1971:157), 12-24 pm (Stoddard 1999:145), 18-33 pm (based on four wheat varieties: waxy, Kanto 107, Centura, and commercial) (Yoo and Jane 2002:299). Other notes: non-birefringent isotropic region in the center of the grain composed of amorphous material, thus in cross polarized light appears as a central darkened hilum area (Chandrashekar et al. 1987:197).

While many of the above characteristics of commercial A-type wheat starch are similar to Morphotype A, the characteristics of Bluestem Wheatgrass were deemed the best fit available. It was determined that A-type wheat starch grains are relatively uniform in their lenticular biconvex shape and did not accurately reflect the variation in shape that was present in Morphotype A grains, which included lenticular, irregular lenticular, jelly bean shaped, wedge shaped, and tear drop shapes. Shape variation in grains was observed however, in Bluestem Wheatgrass, including the shapes listed above.

In cross polarized light and polar face view, some of the Morphotype A grains were darkened or lightened in the central hilum area, suggesting a depression or raised bump was present. In normal light, this bump/depression appears as a ring or 'donut' around the hilum area (see Figs. 6-7 and 6-8). Sometimes a small extinction cross in the hilum area within the larger cross of the whole grain could be seen (these were always the grains that appear to have a 'donut' or ring around the hilum area in normal light). A bump or depression of the hilum area is present in some A-type wheat starch grains (Evers 1969:97-98), the bump being especially evident in young immature grains (Evers 1971:158). However, this central bump or 124 depression look was also seen in some P. smithii wheatgrass starch grains. The P. smithii starch grains with depressions tend to the middle of the size range rather than the extremes (B. Kooyman, pers. comm.). A central hilum area differentiated from the rest of the grain has been described as a non-birefringent isotropic region composed of amorphous material for bread wheat (Triticum aestivum) and sorghum (Sorghum bicolor) cereals (Chandrashekar et al. 1987:197); this central hilum area may also be present in the starch grains of uncultivated wheatgrasses.

The Morphotype A starch grains observed often had damage features suggestive of gelatinization. When gelatinization of starch occurs, the grain usually swells and looses birefringence (Bowler et al. 1980:189). The polar flat view of Morphotype A grain extinction crosses were similar to the description and micrographs of Henry et al. (2009:918) starch processed by boiling for 10 minutes that is "much faded and with a very wide center." Starch can become modified by natural agents, or human actions from the past or present, but how wheatgrass (Pascopyrum smithii) starch identified on the tools came to be modified is unknown. Perhaps heat exposure in earth ovens caused a loss of grain birefringence. It should also be noted that modern processing techniques for starch products can involve heating in water, changes of pH, shearing, wet and dry milling and decanter centrifuging (Seib 1994:50-51), all of which modify the starch end product.

Wild or American Vetch (Vicia americana) seed starch grains from the BC/AB starch reference collection also showed some starch grains with what appear to be long and relatively deep fissures. The Wild Vetch was discounted as a possible identification of Morphotype A since 1) the vetch grains do not appear to be a true lenticular or biconvex shape, 2) they seem to have fissures rather than smooth grooves, and 3) rotations of the vetch grains show that the fissures are present on both sides/edges but do not completely encircle the grains. 125

Morphotype B (n- 2 grains) Granule form: simple Extinction cross in polar flat view: very faint and fuzzy Basic shape in 3D: sphere Hilum open/closed: open Hilum colour: pink Hilum placement: centric Hilum area raised bump/depressed: N/A Fissures: none Groove?: none Lamellae: not visible Facets: none Surface features; smooth Max length (ranged: 5.7-6.8 pm (small and medium size grains) Max length (mean averaged: 6.3 pm

Other notes: N/A

Morphotype B starch grains were extracted from actue-angled tools and represent species in the Rosaceae Family. For the Rosaceae Family, "Seed forms were predominantly small to very small irregular spheres, with the exception of an uncommon but unique very large reniform-like form in prickly rose. Seed starches are not only small, but the hilum is often red to pink in the centre or in the outer rim; such forms are not common in other Families, but a few species in the Liliaceae, Apiaceae, Ericaceae, and Geraniaceae, and Grossulariaceae also have such hila, primarily in seeds," (Kooyman and Kooyman 2010) (see Figs. 6-13 to 6-16). The extinction crosses of the two Morphotype B starches were very faint under cross polarized light (Figs. 6-14 and 6-16), however partial loss of birefringence can result from damage to the grains. The Rosaceae Family includes many fruits that were eaten by Canadian Plateau peoples, and Morphotype B starch grains provide evidence of cherry, berry, and/or rose hip processing. Species in this Family include: Pin Cherry (Prunus pensylvanica), Bitter Cherry (Prunus emarginata), 126

Choke Cherry (Prunus virginiana), Salmonberry (Rubus spectabilis), Thimbleberry (Rubus parviflorus), Blackcap Raspberry (Rubus leucodermis), Red Raspberry (Rubus idaeus subsp. strigosus), Wild Strawberry (Fragaria virginiana), Wood Strawberry (Fragaria vesca), and Saskatoon Berry (Ameianchier alnifolia), Prickly Rose (Rosa acicularis subsp. sayi), and Woods' Rose or Prairie Rose (Rosa woodsii subsp. ultramontana).

Fig. 6-13. This Morphotype B grain is a 6.8 pm Fig. 6-14. Same grain. Extinction cross is weak in sphere with an open pink hilum. Starch 34 from cross polarized light, 400x. tool 1790, 400x.

Fig. 6-15. Morphotype B grain is a slightly Fig. 6-16. Same grain showing a weak irregular sphere with a light red core in the extinction cross in cross polarized light, open hilum and a dark red rim and is 5.7 pm. 400x. From tool 9991, 400x. 127

Morphotype C 'Oval with Depressed Hilum Area' (n= 5 grains) Granule form: simple Extinction cross in polar flat view: fuzzy and hardly visible Basic shape in 3D: near sphere to slight oval Hilum open/closed: open Hilum colour: pink Hilum placement: centric Hilum area raised bump/depressed: depressed. Hilum can appear as a dark irregular solid, a ring or rim with a lighter centre, or as large depressed pink area, depending on how the grain is rotated Fissures: none Grooves: none Lamellae: 1 ring visible near grain edge (hint of lamellae near outer edge margin/ boundary) Facets: none Surface features: smooth Max length (ranged: 6-9.4 pm (medium size grains) Max length (mean average): 8.2 pm Other notes: medium size grains (5.5-10 pm), extinction cross low visibility could mean these are starches modified by taphonomic and/or human actions.

Morphotype C medium sized grains are too large and too ovate to be part of the Rosaceae Family. Morphotype C grains also poses depressions in the hilum area that are lacking in the Rosaceae Family. These grains differ from Morphotype B in that their shape is not a perfect sphere, but rather a sub-round to slightly ovate. At first view, the grains usually appear as perfect spheres, however, when rotated the grains show a slight oval shape. Depending on the plane of focus, the hilum can appear as a dark irregular solid, a ring or rim with a lighter centre, or as large depressed pink area. In all views the hilum appears large. The most distinctive feature of Morphotype C is the view of the grain with large depressed oval pink 128 hilum area (resembling nucleated red blood cells of frogs or birds) (see Figs. 6-17 to 6-19).

Fig. 6-17. Morphotype C oval grain with large pink depressed hilum area, 8.620 jjm. Starch 1 from tool 3674, 400x.

'•"•'if*•. , _ •

Fig. 6-18. (center, top) Morphotype C oval grain with depressed pink hilum area apparent, 8.8 |jm. Starch 1 from tool 13575, 400x. 129

Fig. 6-19. Morphotype C with large pink hilum area, 6 pm. Starch 11 from tool 9995, 400x.

Morphotype D 'Roughened Bowl Shaped Bell' (n= 3 grains) Granule form: simple Extinction cross in polar flat view: fuzzy to hardly visible Basic shape in 3D: bell gumdrop, bowl sides, straight base Hilum open/closed: closed Hilum colour: none Hilum placement: eccentric Hilum area raised bump/depressed: N/A Fissures; none Grooves: none Lamellae: absent Facets: none Surface features: roughened Max length (ranged: 11.4-22.8 pm (large to very large size grains) Max length (mean average V 16.2 pm Other notes: perhaps cratered surface of grain

Morphotype D starch grains are roughened bowl shaped bells that resemble gumdrop candies. The hilum is closed and placement appears eccentric in side view, centric in top down polar view. The grains occur singly and there are no 130 fissures, grooves, lamellae or facets (Figs. 6-20 and 6-21). The sample size is small but all grains are very roughened, perhaps even cratered. Whether this results from taphonomic damage that modified the starch or this is the natural surface topography of the native grains is not known. Starch 12 from tool 9995 (flake with steep-angled marginal unifacial retouch CS2) is particularly pitted in appearance (Fig. 6-21).

Fig. 6-20. Morphotype D starch 1 from tool 11356 is a bowl shaped bell with open hilum arid rough surface texture, 14.3 pm, 400x.

Fig. 6-21. Very large Morphotype D starch grain 12 from tool 9995, with a very rough, perhaps even cratered surface, 22.8 pm, 400x. 131

Morphotype E Reniform 'Reniform with Lamellae' (n= 2 grains) Granule form: simple Extinction cross in polar flat view: fuzzy Basic shape in 3D: reniform Hilum open/closed: closed Hilum colour: none Hilum placement: slightly eccentric Hilum area raised bump/depressed: N/A Fissures: 1 short fissure through hilum Grooves: absent Lamellae: present Facets; absent Surface features: smooth Max length (range): 25.9 pm (huge grains) Max length (mean average): 25.9 pm Other notes: N/A

Morphotype E is represented only by two grains from two tools. The most distinctive features of this morphotype is the kidney-bean or reniform shape, presence of easily visible lamellae, and a curved fissure running through the hilum area (Figs. 6-22 and 6-23). The extinction cross appears fuzzy or weakly defined under cross polarized light. Size of both starch grains was 25.9 pm. 132

Fig. 6-22. Huge Morphotype E reriiform with lamellae, 25.9 pm. Starch 12 from tool 1560, 400x.

Fig. 6-23. Huge Morphotype E reniform with lamellae, 25.9 pm. Starch 1 from tool 3824, 400x.

Morphotype F 'Irregular Roughened Sphere' (n= 9 grains) granule form; simple Extinction cross iri polar flat view: intermediate to fuzzy Basic shape ill 3D; irregular sphere Hilum open/closed: open Hilum colour; berry or pink Hilum placement: centric Hilum area raised bump/depressed: N/A Fissures: appear to be absent, although some grains are small and this may be difficult to see 133

Grooves: none Lamellae: absent Facets: absent Surface features: roughened Max length (range): 5.7-10 \im (medium grains) Max length (mean average): 8.1 pm Other notes: N/A

Morphotype F starch grains are roughened irregular medium size spheres with centric hila (Figs. 6-24 to 6-26). Edges of grain always appear rugged. Hilum shows pink or berry colour tinges and may appear as a soild dark dot or as an open clear white dot, depending on plane of focus. Hilum area is not raised or depressed.

Fig. 6-24. (center, top) Sphere shaped Morphotype F grain 6 with open pink hilum and rough edges, 7.1 pm. From tool 9995, 400x. 134

Fig. 6-25. Morphotype F sphere shaped grain Fig. 6-26. Same grain in cross polarized light, with rugged edges and open pink hilum. 400x. Starch 10 from tool 1560, 8.6 pm, 400x.

Morphotype G 'Polygon with Open Centric Hila, Fissures and Facets' (n- 39) Granule form: simple Extinction cross in polar flat view: distinct to intermediate Basic shape in 3D: polygon Hilum open/closed: open Hilum colour: pink, brown or no colour Hilum placement: centric or slightly eccentric Hilum area raised bump/depressed: N/A Fissures: present. Fissures can be stellate, X or Y shaped over the hilum. Fissures are typically irregular can be relatively deep (as suggested by how dark and wide they are). Fissures are sometimes long. Grooves: none Lamellae: absent Facets: present. Corners where facets join are rounded. Facet edges are not uniformly straight, being always slightly indented, bumpy or irregular in appearance. Surface features: almost always rough, bumpy irregular surface, edges of grains appear irregular Max length (ranged: 11.4 pm-28.5 |jm (large, very large and huge sized grains) 135

Max length (mean averaged 17.9 |jm Other notes: N/A

Moprhotype G polygon shaped grains were the most frequently observed starch grains (n= 39). The Morphotype G grains were recovered from 8 tools, including: 3 flakes with steep-angled marginal unifacial retouch, 2 steep-angled utilized flakes, 1 flake with -angled marginal bifacial retouch, 1 cobble with linear striations, and 1 acute-angled biface end fragment and pieces. These tools were collected from CS1, CS2, and surface areas of the White Rock Springs site. Morphotype G starch grains are polygons with at least one fissure and at least somewhat irregular or rough edges. The grains have centric or slightly eccentric open hila (Figs. 6-27 to 6-37).

Fig. 6-27. Morphotype G polygon, open hilum with Y fissure, 20 |jm. Starch 5 from tool 1790, 400x. 136

Fig. 6-28. Morphotype G starch 5 from 1790, Fig. 6-29. Same grain in cross polarized light, rotated with Y fissure still apparent, 400x. 400x.

Fig. 6-30. Morphotype G polygon with open hilum and rough edges, 20 pm. Starch 8 from tool 5364, 400x.

Fig. 6-31. Morphotype G from Starch 8 from Fig. 6-32. Same grain in cross polarized light, tool 5364, rotated with edges now appearing 400x. smooth, 400x. 137

Fig. 6-33. Morphotype G Starch 1 from tool Fig. 6-34. Same grain in cross polarized light, 10522, 17.1 |jm, 400x. 400x.

Fig. 6-35. Morphotype G polygon,14.3 [jm. Starch 8 from tool 1790, 400x.

Fig. 6-36. Morphotype G polygon, rotated with Fig. 6-37. Same grain in cross polarized light, Y fissure evident. Starch 8 from tool 1790, 400x. 400x. 138

Pearsall et al. (2004:430-431) defined the characteristics of of 9 varieties or subspecies of Zea mays L. starch from the kernels (seeds). They included species representative of four types of endosperm including pop and flint (hard endosperms), flour (soft endosperm), and sweet endosperm. Based on their work as well as Hall and Sayre (1973), Fannon et al. (1992), Piperno et al. (2000), Singh et al. (2003), Perry (2004), Zarrillo and Kooyman (2006), Boyd et al. (2006), and Boyd and Surette (2010), the following is a list of the general characteristics of maize starch, regardless of variety unless otherwise noted.

Corn (Zea mays) characteristics Granule form: simple Extinction cross in polar flat view: distinct to intermediate (observations from published micrographs, not described in text of consulted articles) Basic shape in 3D: varies from spherical to oval-spherical to polyhedral (edges showing multiple pressure facets), depending on how tightly packed granules were in the kernel. Hemispherical or 'vase-shaped' grains may also occur and are diagnostic of maize. Hilum open/closed: open Hilum colour: not described Hilum placement: centric or slightly eccentric Hilum area raised bump/depressed: not described Fissures: May be present or absent (fissures on corn starch polygons not always visible by LM). In the dried grain, a central cavity sometimes develops, with fissures radiating out from it in a linear, Y or X pattern. Some or all of this cracking/ Assuring seen in corn starch occurs inside the grain rather than surficially (Hall and Sayre 1973:121), and there are no published reports showing fissures on the surface of corn starch by SEM, even though they appear by light and transmission electron microscopy to extend to the surface (Fannon et al. 1992:285). Grooves: surface can appear grooved Lamellae: absent Facets: hard endosperm corns have 4 to 5 sided blocky angular polygon grains 139

Surface features: hard endosperm corns have rough, cratered or grooved surfaces. Soft endosperm corns are predominantly spherical and smooth Max length (ranged: 4-24 pm (very small, small, medium, large, very large grain sizes) Max length (mean average): range of means of different varieties from 8 ± 1.5 pm (medium) to 16 ± 4.9 pm (large and very large) Other notes: Zea mays grains have been noted to often have a distinct 'double border' visible near their edges. I think this is an unreliable and confusing characteristic to attempt to observe. Micrographs of all starch grains examined in this study seem to show a 'double border'. When originally observed, I called this phenomenon 'lipping at grain edge', 'hint of lamellae' or one 'ring' near grain edge. I have since realized the 'double border' appearance is probably an illusory refraction image or artifact of light microscopy and not a real morphological feature of the starch grains. Double borders or halo effects around starch grains are known as Becke lines and were observed on all starch grains in observed by Sandstedt (1955:19).

Thirty nine polygons that are likely hard endosperm varieties (eg. pop and flint varieties) of Zea mays were identified from 8 tools. The number of starch grains on these 8 tools was not distributed evenly. Some tools only had 1 starch with Morphotype G characteristics, while other tools had 2, 3, 8, or 22 Morphotype G grains identified, accounting for 8% to 100% of all starch grains found in scanned residue extracts from the tools. Starch grains were identified as hard endosperm varieties of Zea mays based on the following characteristics: simple grain, blocky polygon shape, distinct to intermediate extinction cross, open hilum that is centric or slightly eccentric, must have 1 or more fissures (Y shape common), and no lamellae visible. I did not include a rough cratered or grooved surface as a trait necessary for the grain to be classified as a hard endosperm maize since this surface feature does not seem to be a requirement (based on flint maize micrographs presented in Zarrillo and Kooyman 2006:491 that appear smooth surfaced). The grain edges of Morphotype G always appeared at least somewhat 140 irregular, with indented and not uniformly straight edges being common. These irregular edges match the appearance of slides the author made of commercial corn starch and commercial corn flour starch grains (Figs. 6-38 and 6-39). Hard endosperm polygon corn starch grains have surfaces that are rough, and rough surfaces predominated Morphotype G grains. In Morphotype G grains that were not rough surfaced, edges were always somewhat irregular.

Fig. 6-38. Commercial corn starch, 400x. Note Fig. 6-39. Commercial corn flour, 400x. Note polygons with open hila and grain with Y polygons with irregular bumpy edges and grain fissure in center. with deep Y fissure at top center.

From all accounts, there is no ethnographic or archaeological evidence to support the use of corn on the Canadian Plateau in pre-contact times; any corn starch encountered in the lithic residue samples is of modern origin. Corn starch is the most used starch in the world (Bertolini 2010:15) and there are over 20 cultivars with variable attributes (Gaytan-Martinez et al. 2006:135), selected according to application. Corn starches are used extensively not just in the food industry to change product texture, nutritional value, and extend shelf life, but are ubiquitous in other commercial uses in the paper industry, cosmetics, and biodegradable plastics and adhesives, to name only a few (Smith 2010:4). In the production of corn flour products, varieties with a high endosperm hardness are preferred (Gaytan- Martinez et al. 2006:135). Therefore, hard endosperm corn starch can be expected to be present in an assortment of common products and foods that could have 141 come in contact with archaeological materials. Although disappointing, the presence of corn starch found in residue samples from eight lithics is not surprising. This starch was most likely introduced to the tool surfaces during the washing and handling that took place at University of British Columbia before residue analysis was conducted. No special measures were taken to prevent modern food starches from coming into contact with the artifacts.

The similarity between Morphotype G starch grains and those of bitterroot (Lewisia rediviva), present in the BC/AB reference collection, was also considered. One bitterroot grain forms appeared to match Morphotype G and thus they were compared. The following trait list was compiled by preliminary observations the author made of Bitterroot micrographs taken by Brian Kooyman in 2011 for the BCI AB reference collection.

Bitterroot (Lewisia rediviva) Starch from BC/AB Reference Collection Granule form: simple Extinction cross in polar flat view: no information Basic shape in 3D: bells, spheres and polygons Hilum open/closed: open Hilum colour: none Hilum placement: centric or slightly eccentric Hilum area raised bump/depressed: N/A Fissures: present. Short fissures over hilum nearly always present. Fissures can be stellate, X or Y shaped. Arms of fissures always shallow and usually irregular. Fissures usually do not extend beyond the hilum area. Grooves: absent Lamellae: perhaps slightly visible Facets: sometimes present. When present, corners where facets join are rounded. Edges of grains that have facets are always straight and not indented or irregular in appearance. Surface features: smooth, edges of grain do not appear irregular 142

Max length (ranged information not available Max length (mean averaged: unknown, but Bitterroot grains appear to be larger than Morphotype G grains when micrographs taken at 400x are compared. Other notes: Most common Bitterroot starches are bell and sphere shapes, with polygons more rare. Hilum area appears 'torn up' or 'scratched out' in most grains.

There were some similarities in grain characteristics when Morphotype G and the reference collection Bitterroot grains were compared, as the characteristics lists demonstrate. However, there were some important differences in assemblage of grains present and polygon grain morphology between the two starches. To begin with, the starch assemblage of Bitterroot appears to be mostly composed of bell and sphere shapes, with fewer polygons. This means the sphere and bell shaped Bitterroot grains would be the most commonly represented grain type in an archaeological context, with less polygons. Morphotype G had some grains that showed fissures at the hilum that were quite dark, deep and irregular, unlike the usually short and shallow fissures seen in Bitterroot grains. Deep dark fissures are uncommon in bitterroot starch grains, and when present are mainly in spherical grains, not polygons. The surfaces of Morphotype G grains were predominately rough and Bitterroot grains had smooth surfaces. The facet edges of Morphotype G and Bitterroot polygon grains differed as well. Morphotype G grains were always irregular, bumpy, indented and not uniformly straight in at least one view of the grain, whereas Bitterroot grains had edges that were more regular, smooth and linear in appearance.

According to ethnographic account, bitterroot was a very important geophyte food resource for Canadian Plateau First Nations people (Turner 1997:137). The preparation of bitterroot has been described thus:

After cleaning a root of leaves and buds, the gatherer rolls it between the hands or on a rock to loosen its skin, then pulls the skin off in one piece.... The gatherer usually removed the heart immediately by splitting open the top of the root and pulling it out with the fingers.The peeled, de-hearted roots can be eaten fresh after 143 baking or boiling a short time, or strung on strings or sticks or spread out on mats to dry for several days, (Turner 1997:137-138).

Thus, it appears no tools are necessary in order to remove the inedible skin or the bitter red 'heart' of the root. It seems unlikely that people in the past would have used extra unnecessary steps such as employment of acute edge angle, steep edge angle or cobble stone tools to process bitterroot. Additionally, there are five tool types that retrieved Morphotype G starch grains, and it seems unlikely that all these tool types (flakes with steep-angled marginal unifacial retouch, steep-angled utilized flakes, flake with acute-angled marginal bifacial retouch, cobble with linear striations, and acute-angled biface end fragment and pieces) were used to process bitterroot. One would expect that if the Morphotype G starch presence was due to human activity, it would be concentrated on only a few specific tool types, showing a consistent pattern of use. Instead the pattern seems to be that of generalized contamination.

The micrographs of Morphotypes A and G were checked for internal variation within each group to see if there was any patterns in grain morphologies to suggest that Morphotypes A and G were not cohesive groups. Some minor variation in the starch grains (eg. some grain edges were more sharply defined than others within Morphotype G, some grains show and X or Y fissure pattern, etc.). Although there is some morphological variation within these Morphotypes, this can be expected when assigning any items to groups. Grains were only grouped in a morphotype if they met all of the morphological and size criteria, as outlined in the characteristics list for each morphotype outlined previously in this chapter. In addition, the Morphotypes A and G were examined to see if there were certain grain types within each morphotype group that were specific to tool type (acute edge angle 'cutting' tools, steep edge angle 'scraping' tools, and grinding/pounding tools). Again, no patterning of specific grain morphologies related to tool types could be established. This lack of pattering further supports the interpretation that Morphotypes A and G starch grains are contaminates. 144

Vascular Tissues Microscopic pieces of softwood and what could be herbaceous and/or hardwood tissue were found in the residue extractions, including tracheids, vessel elements, fibers, and ray parenchyma. 22% of the White Rock Springs tools had tracheids present in their residue extracts and 19% of the tools had vessel elements present.

Softwoods The tracheids of conifer softwoods were fairly common in tool residues. (Figs. 6-40 and 6-41). Conifer tracheids have large round bordered pits which often appear as maltese crosses under cross polarized light, and thus resemble birefringent starch grains. Pit diameter of softwoods and hardwoods can be measured as an aid to species identification (Panshin and de Zeeuw 1980:405-666). In this study only the size of starch grains were recorded, all other residues were not measured.

Fig. 6-40. Conifer tracheid wood fragment. Note Fig. 6-41. Tracheid of a conifer. Arrow points to large bordered pits. From tool 3694, 400x. a large bordered pit. From tool 15182, 400x.

A tracheid fragment from Douglas Fir (Pseudotsuga menziesii) was found from the residues of a broken hand maul near CF9. This can be identified as Douglas Fir due to the large bordered pits (softwood) and presence of spiral thickening. Spirals that appear to cross the width of the tracheid can be seen on the inner face of the secondary wall. The spirals are closely spaced and almost perpendicular to the 145 tracheid axis, features of spiral thickening present on Douglas Fir tracheids (Panshin and de Zeeuw 1980:138, 463) (see Figs. 6-42 and 6-43). There are only two local softwood types that display spiral thickening around their tracheids: Douglas Fir and Yew (Taxus spp.) (Hoadley 1990:18). Pacific Yew (Taxus brevifolia) is found in British Columbia, where it is common in coastal areas and south-central and south-eastern British Columbia, preferring moist to mesic slopes and creek side habitats (Douglas et al. 1998:26). However, according to distribution maps (Douglas et al. 2002:378; Farrar 1995:124), Pacific Yew is absent in the vicinity of the White Rock Springs site and in interior BC generally, making this species unlikely to be found in the lithic residue extracts.

Fig. 6-42. Douglas Fir tracheid fragment Fig. 6-43. Close up of Douglas Fir tracheid displaying large bordered pits characteristic of in cross polarized light. Note diagnostic conifers in cross polarized light. From hand spiral thickening that is nearly maul tool 3651, 250x. perpendicular to the tracheid axis, 400x.

There were some examples of softwood tissue that exhibited cross-field pitting between the ray parenchyma and the longitudinal tracheids. Cross-field pitting also occurs in the hardwoods, but the pits are much smaller. Philips (1941:267-268), Panshin and de Zeeuw (1980:150) and Friedman (1978:26) outlined types of bordered pits occurring in the tracheids of coniferous woods. Piceoid type pits are present in Picea, Larix and Pseudotsuga. Taxodioid pits are present in Sequoia, Taxodium, Abies and Thuja. Cupressoid pits are found in Chamaecyparis, Juniperus, Calocedrus and Taxus, and occasionally are present in Tsuga. 146

Cupressoid pits are also present in a number of foreign conifers outside of Canada and the USA.

Ray parenchyma overlaying tracheids representing softwood was found in the residue extracts of a microblade (Figs. 6-44 to 6-45). Unfortunately, the shape of the pits in the cross fields were not discernible. A vessel element was also found in the residue extract from this microblade 13572, suggesting it was used on conifer wood.

Fig. 6-44. Conifer ray parenchyma cells overlapping two broken tracheids. Cross field Fig. 6-45. The other end of the softwood pitting like this occurs in conifers, note fragment, also displaying large bordered pits, bordered pits on top left of fragment. From 400x. microblade tool 13572, 400x.

One cobble with linear striations from the toss zone of CF 18 contained wood residue that appeared to show cupressoid pit pairs in the cross field pits. The cupressoid cross field pitting is diagnostic to Juniper (Juniperus), and Yew (Taxus) of the Cupressaceae and sometimes present in Hemlock (Tsuga) of the Pinaceae (Panshin and de Zeeuw 1980:151) (see Fig. 6-46). The distribution of Yew and Hemlock do not fall in the vicinity of Hat Creek valley (Len Hills, pers. comm.). Thus, it is likely that the wood fragment with cupressoid cross field pitting from cobble tool 3674 is from Juniper, probably Rocky Mountain Juniper (Juniperus scropulorum). Rocky Mountain Juniper was used by all First Nations peoples wherever it was available from the Strait of Georgia to western Alberta to fumigate 147 habitations and clear the air (Turner 1998:69-70). Wood fragments like those seen in the residue extracts could represent woodworking activities such as bark stripping, tree-felling, pounding, whittling or grinding and smoothing.

Fig. 6-46. Softwood fragment likely from Rocky Mountain Juniper (Juniperus scopulorum). Note cupressoid pit pairs in the cross fields of ray parenchyma cells and overlapping tracheids. From cobble tool 3674, 400x.

Hardwoods/Herbaceous Plants Vessel elements from hardwoods and/or herbaceous plants were found, although less frequently than tracheids. The type of intervessel pitting on vessel elements (either alternate pitting, opposite pitting or scalariform pitting) can be of diagnostic importance (Panshin and de Zeew 1980:175). A vessel element with alternate type intervessel pitting was identified (Core et al. 1976:74; Hoadley 1990:37; Evert 2006:260). The intervessel pits occur where vessel elements are in contact and are very distinctive (Figs. 6-47 and 6-48). This and other vessel elements were found on flake tool 9995, suggesting this was a tool used perhaps for cutting/sawing or slicing/shaving a woody or herbaceous eudicot. Slit-opening bordered pits in vessel elements were also observed (Figs. 6-47 to 6-51) (similar to those pictured in Mauseth 2009:106). These slit bordered pits do not show maltese crosses in cross polarized light as other bordered pits do, such as those bordered pits on conifer tracheids. 148

Fig. 6-47. Alternate intervessel pitting on a Fig. 6-48. Same vessel element, with alternate woody or herbaceous eudicot vessel element. intervessel pitting distinct in cross polarized From flake tool 9995, 400x. light, 400x.

Fig. 6-49. Vessel element with alternate intervessel pitting from tool 3830, 400x.

Fig. 6-50. Vessel element with slit opening Fig. 6-51. Same in cross polarized light. Note bordered pits and intervessel pitting from tool 'wavy lines' appearance and lack of maltese 3830, 100x. crosses, 100x. 149

The residues from the hand maul pounding tool (Tool 3651) contained far more tracheids, vessel elements and large fragments of woody and/or herbaceous tissue than any of the other tool types, a total of 27 items. This observation gives sound grounds to say that the wood and/or herbaceous residues are not from contamination, given that they were found in such large concentrations on the maul tool only. It appears this groundstone tool was used for processing herbaceous materials (perhaps pounding roots to separate the skins from the roots) and/or woodworking in some fashion. Handmauls were used with wedges to remove large sections of trees and it is possible that people were collecting wood for firing earth ovens to cook geophytes.

Phytoliths Grass phytoliths were present in all 106 tool residue preparations, with forms including elongate trapeziforms and bilobes, as well as phytoliths with dark carbon inclusions observed. An obstacle of grass phytolith classification is that the costal short cell phytoliths that tend to be the most distinctive only make up a fraction of the total siliceous residues from the plant (Twiss 2001:13). As a result, many grass phytolith residues that are found in archaeological samples may not be readily identifiable. However, there were some phytolith forms found in the lithic residue samples that are diagnostic to grass subfamily or general descriptive type.

Grass phytoliths that I initially designated as 'corn cobs' that could be from the Pooideae subfamily were common in the residue extracts from tools (Fig. 6-52). These phytoliths could belong to the Pooideae grasses because they are elongate and wavy and appear to have a vaguely trapezoidal shape, although the more defined trapezoid that is distinctive of the subfamily (Brown 1984:361,366) could not be confirmed. Phytoliths of the Panicoideae subfamily were much more rare, identifiable as bilobes that have short shafts and straight end lobes, some showing indentations or sculpturing (Fig. 6-53). A Panicoid-type short cell bilobe phytolith similar to the bilobes observed in residue extracts is illustrated in Frelund and Tiezen (1994:326). The grass phytoliths likely were present in the small amounts of 150 dirt adhering to cracks and crevices of the stone tools. It may be that some of the tools were used to prepare grass mats and bags (Matthew 1986b:5), and/or make grass string for hanging up roots to dry (Turner 1998:113).

Fig. 6-52. Elongate trapeziform Pooid grass Fig. 6-53. Roughened Panicoid bilobe phytolith with 'corn cob' look. From tool 6368, phytolith. From tool 3677, 400x. 400x.

Phytoliths of what appear to be edge spine cores from a type of short trichome were also represented in the residue extracts (Brian Kooyman, pers. comm.). These edge spines are roughly triangular shaped, with the base sometimes appearing to be indented in the center (Fig. 6-54).

Fig. 6-54. Triangular edge spine core phytolith from a trichome. From tool 17661, 400x. 151

Pearsall arid Dinan created a hierarchical classification system to describe all phytolith type categories, not just grass phytoliths (1992). Their scheme is not a key for plant taxa, but was helpful for describing the articulated phytoliths of epidermal tissue fragments that were found in lithic residue samples. Unfortunately, only an incomplete version of their phytolith classification scheme has been published for illustrative purposes (1992). According to Pearsall and Dinan's scheme, the phytoliths seen in Figure 6-55, are classified as 10ICa100: epidermal quadrilaterals that are rectilinear long (length 2-4 times width), with slightly undulating edges, smooth surface, no perforations, no projections, and pointed, inter-digitating short edges (1992:53). These epidermal quadrilateral phytoliths originate from eudicot leaves, and appear similar in a variety of species such as sages (Brian Kooyman, pers. comm.). Another epidermal plant fragment from tool 3651 was classified as 10IBa200Aa: slightly undulating quadrilateral with smooth surface, no perforations, no projections, and right angles on the short edges (1992:51-52).

Fig. 6-55. Epidermal quadrilateral phytoliths from a leaf of a eudicot. From hand maul tool 3651, 400x.

An epidermal plant fragment with hexagonal shaped phytoliths from hand maul 3651 (Fig. 6-56) was not represented in the partial classification scheme presented in Pearsall and Dinan (1992), but was found in Pearsall (2000). It is described as 24IC: epidermal nonquadrilateral blocky that is rounded or spheroidal on at least 152 one side, multifaceted, projections on some facet edges, and may have grainy surfaces (Pearsall 2000:461). Not only did the residues from the broken maul tool 3651 contain non-contaminant starches, spruce (P/'cea sp.) and other pollen grains, lichen, fungal hyphae, woody or herbaceous fragments, tracheids and vessel elements, but also at least seven articulated plant fragments of epidermal phytoliths (pollen, lichen, and fungal remains are described below).

Fig. 6-56. Epidermal tissue fragment with hexagonal shaped phytoliths. From hand maul tool 3651, 400x.

Tool 16099 also contained epidermal phytoliths. These were large, interdigitating, and amber-brown coloured (see Fig. 6-57). The phytoliths can be described as 10ICc100: epidermal quadrilaterals that are rectilinear (length 2-4 times width), with two serrated edges, two smooth to undulating edges, and perforations (or perhaps projections). The phytoliths are also highly serrated and serrations are rounded (1992:56). The bulbs of Liliaceae and Apiaceae found on the Plateau have abundant silicified plate-like epidermal fragments, and phytolith production is the highest in the bulb part of plants in these families (Eccleston 1999:125). It is possible that the epidermal phytoliths recovered from the White Rock Spring tools represent processing roots of the Lily and/or Carrot families. Alternatively, the interdigitating epidermal phytoliths could represent a fruit/berry 'skin'. Fig. 6-57. Interdigitating pitted epidermal phytoliths from tool 16099, 250x.

Other interesting phytoliths were found in the residue extracts. However, phytolith identification was not a major focus of this study; the author lacks the experience and skill to classify them further. Currently, Brian Kooyman of the University of Calgary is compiling the necessary phytolith reference collection to conduct intensive study on archaeological phytoliths in Alberta and British Columbia. Phytolith microfossils may prove a fruitful avenue to access ancient plant foods.

It can be difficult to differentiate between phytolithic remains, epidermal tissues and vascular tissues. And, to what extent items identified as vascular cells are technically silicified phytoliths is unknown. From her experimental and archaeological observations of vascular cells on tools, Anderson suggested that"... organic materials almost certainly would not have survived in the archaeological conditions or after the cleaning of experimental tools, these residues are probably all silica-covered, or else are silica cast and replicas of these cells," (1980:188). A general staining technique to identify phytolithic remains securely was suggested to the author by Len Hills for future residue studies. Safranin-0 when applied to residue slides will stain nearly all organic matter red, thus mineral phytoliths will not take up the stain and be confidently identifiable (Len Hills, pers. comm.). However, Traverse (2007:631) notes that Safranin-0 does not stain well with glycerin slide mounting mediums, so if this staining technique was to be undertaken, another 154 type of medium would have to be used. Brian Kooyman also suggested ways to confirm the identification of phytoliths in artifact residues. He has observed that because phytoliths are amorphous they usually disappear under cross polarized light, whereas the cellulose in cell walls has a crystal structure and is thus birefringent. Kooyman has also observed that phytoliths typically display a slightly pink hue and vascular tissues are clear and green (pers. comm.). However, when micrographs of phytoliths from residue extractions were re-examined, pink colouration was not apparent to the author.

Pollen Only the three pounding tools contained pollen in their residue extracts. Residue extracts of hand maul 3651 contained spruce (Picea sp.) pollen and cobble tool 3674 contained pine (Pinus sp.) pollen (Rolf Mathewes, pers. comm.) (Figs. 6-58 and 6-59, respectively). Spruce pollen grains have a furrow, usually well defined, as a shallow narrow slit between two bladders that are large and nonpendulous. The bladders of spruce are are finely granular, small relative to the grain body, and posess a fine weblike reticulum on the inner surface which becomes smaller near the bladder-body junction (Owens and Simpson 1986:957; Wodehouse 1935:262). Pine pollen grains display very rough sculpture on the body, with a long furrow between the air filled bladders (Agashe and Caulton 2009:82). The bladders of pine are large relative to the size of the grain and often pendulous, with the inner surface displaying a distinct reticulum of weblike ridges (Owens and Simpson 1986:959). These conifer pollen grains may simply represent 'pollen rain' from the surrounding landscape. On the other hand, a case might be made for the pollen grains on the pounders being present due to human actions given that no pollen was found in the residues from any other tool type. However, hand maul 3651 and cobble tool 3699 were collected at surface level, and only cobble tool 3674 was excavated from the toss zone of CF 18. Fig. 6-58. Spruce (P/'cea sp.) bisaccate pollen Fig. 6-59. Pine (Pinus sp.) bisaccate pollen grain Tool 3651, 400x. grain Tool 3674, 400x.

Other pollen grains were found in the residue extracts, although they could not readily be identified.

Fungal Micro-remains

Hyphae and Spores Fungal hyphae were encountered in 90% of the tool residue samples. At least three types of hyphae were observed, a dark brown type, a clear (perhaps white) type, and another type that is textured and a light yellow colour.

Some brown hyphae appeared to have textured surfaces. This hyphae texture can appear as a stippled look, present in this example at one end (Fig. 6-60). Clamp connections and cross walls were visible in some brown hyphae (Fig. 6-61), a feature characteristic of the phylum Basidiomycota (Sharma 2005:227-228). Morphological features such as presence of clamp connections and septa (cross walls), the density of hyphae, and substrate associations are useful characteristics for taxonomic identification. 156

Fig. 6-60. Fungal hyphae end (on left) appears Fig. 6-61. Brown fungal hypha with clamp sippled. From tool 9984, 400x. connection, a feature of the Basidomycota, and cross walls from tool 3651, 400x.

There are a number of possible fungi that the hyphae could represent. Ethnographically mushrooms were eaten in interior British Columia, including Cottonwood Mushroom (Tricholoma populinum), Chanterelle (Cantharellus cibarius), Shaggy Mane (Coprinus comatus), Oyster Mushroom (Pleurotus ostreatus) and Pine Mushroom (Tricholoma magnivelare) (Turner 1997:39-43; Turner et al. 1987:921). Black Tree Moss, or Black Tree Lichen (Bryoria fremontii Tuck.) is a dark brown lichen that was eaten by most interior BC First Nations groups (Turner 1977:463; 1997:33). Black Tree Moss was a favourite food of the Interior Salish Okanagan-Colville, who baked the lichen in pits over leaves, sometimes with wild onions, Saskatoon berries, or other flavourings (Brodo et al. 2001:79). Keddie (1988:7) notes Black Tree Moss was gathered in large quantities by the Kootenay for food, beaten with stone paddles, and cooked underground for days. Alectoria and Bryoria species of lichens were also used in clothing or shoes by interior British Columbian First Nations, usually by the poor who could not obtain animal skins (Brodo et al. 2001:79,180). Another possible explanation is the hyphae are remnants of threadlike mycorrhizae that form a network on and inside plant roots in the soil. Mycorrhizae are a mutalistic symbiont of over 90% of plant species, helping the plant obtain nitrogen, phosphorus, potassium and calcium, as well as improving water absorption (Bidlack and Jansky 2011:75; Sharma 2005:293; Spooner and Roberts 2005:75; Stephenson 2010:170; Sumbali 157

2005:199). Pine, spruce, fir and birch species all have ectomycorrhizae and are common in the White Rock Springs site area. Ectomycorrhizae on forest plant roots break down minerals by excretion of organic acids (Landeweert et al. 2001:248), and there is no reason to believe they would not act on buried lithic tools. However, the fungal hyphae in the tool residue extracts were not seen in association with plant roots, and thus they are unlikely to be identified as mycorrhizae. There is also a chance that the fungal hyphae in the residue samples originate from saphrophytic fungi that digested organic material that was adhering to the lithics or in the soil associated with the lithics.

Starch taphonomy experiments with stone tools by Barton found fungal hyphae on almost all tools that were left in the soil, as well as tools left on the surface (2009:134). Barton also noted that fungal hyphae tended to surround starch aggregates on his experimental tools. An experiment in which starch was added to a soil surface showed that fungi are able to quickly form and attack large soil aggregates (macroaggregates of >250 pm) with enzymes, concentrating in the areas of highest starch density. Smaller soil aggregates ('microaggregates' of <53 pm and 53-250 pm) are not as easily accessible to the fungi (Guggenberger et al. 1999 cited in Haslam 2004:1725). No fungal hyphae were found associated with starch grains in residue extraction samples, but there was at least one example of plant tissue with attacking fungi (Fig. 6-62). 158

r* A ,

Fig. 6-62. Fungal attack of plant material from tool 16099, 100x.

Pounding maul tool 3651 had at least three clumps of green round cells surrounded by hyaline (clear) material that is bubbly in appearance. These were identified as fragments of an unknown lichen (see Stephenson 2010:plate 71). The green cells are the photosynthesizing algal photobiont and the hyaline material surrounding green cells is part of the fungal thallus of the mycobiont (Figs. 6-63 and 6-64). These lichen fragments appear to be soredia, small reproductive structures that are packages of the photobiont and the mycobiont partners that colonize new locations (Hawksworth and Hill 1984:40-41; Sharma 2005:399-310; Budel and Scheidegger 2008:55, Fig. 4.24).

Fig. 6-63. Reproductive soredium of a lichen. Note green algal cells and surrounding hyphae Fig- 6-64. Another lichen soredium fragment that are hyaline, bubbly and smooth. From hand fr°m the same tool, 400x. maul 3651, 400x. 159

Dr. Heather Addy, a biologist at University of Calgary, states that the algal cells seen in the soredia could be a Trebouxia sp., as that is the most common green photobiont of lichens (pers. comm.). As for the mycobiont, the author expected the typical long and linear appearance of fungal hyphae to be exhibited in the soredia. However, the hyphae in soredia are arranged so that they are overlapping and crowded together, which protects the algal cells from desiccation. The non- filamentous look of the hyphae is due to the fact that hyphae of soredia closely encircle the algal cells, making it difficult to see individual hyphae with LM (Heather Addy, pers. comm.). The lichen residues collected from the ground stone surface of pounding maul 3651 are likely of recent origin and not present from human actions. The pigmentation of the algal cells of the lichen are still relatively green, suggesting the lichen is not of great antiquity. The lichen could simply have been growing on the tool surface and be of modern origin, since it was collected at the surface of CF 9.

What appear to be fungal spores were also found in some tool residue extracts. These are often spherical or ovate and darkly coloured. What is likely a spore from acute-angled tool 3824 was found. The item is bowl shaped, darkly amber-brown coloured, has one deep depression, and its surface is slightly granular (Fig. 6-65). Figure 6-66 is another example of a fungal spore seen in the residue extract from tool 9987 (Len Hills, pers. comm.). Some spores had a pedicel, or appeared to be germinating, like the one seen in Figure 6-67. Fungal spores likely originate from the soil context. Fig. 6-65. Spore from acute-angled tool 3824, 400x.

Fig. 6-66. Fungal spore from tool 9987, 400x. Fig. 6-67. Germinating fungal spore. Tool 3824, 400x.

Animal Micro-remains

Feather Barbule One acute-angled tool from CS2 had part of a feather among its residues that was identifiable to order. The feather piece identified is the pennulum portion of a plumulaceous (downy) barbule. The barbule is un-pigmented, has triangular shaped nodes and a prong near the tapering distal end (Figs. 6-68 and 6-69). 161

Fig. 6-68. Barbule from a feather of an Anseriform Fig. 6-69. Prong apparent on tapering distal bird. Note triangular shaped nodes and lack of end of the barbule, 400x. pigmentation. From acute-angled tool 13574, 250x.

These feather characters match the group of birds in the order Anseriformes (waterfowl including ducks, geese and swans) (Dove and Koch 2010:34-35). Interestingly, the most productive waterfowl habitats in the Lillooet area today are found in the Hat Creek valley (Mathewes 1978:85). Waterfowl certainly were utilized as food sources by past peoples. Birds also hold widespread significance in mythology and spiritual beliefs; often wings and feathers were prized and sought after (Gilbert et al. 1985). Twenty-one Anseriform species have breeding grounds in the Lillooet area, near the Hat Creek valley and White Rock Springs site:

Branta Canadensis (Canada Goose) Anas platyrhynchos (Mallard) Anas strepera (Gadwall) Anas acuta (Pintail) Anas carolinesis (Green-winged Teal) Anas discors (Blue-winged Teal) Anas cyanoptera (Cinnamon Teal) Anas americana (American Widgeon) Anas clypeata (Northern Shoveler) Aythya americana (Redhead) 162

Anthya collaris (Ringed-neck duck) Anthya valisineria (Canvasback) Anthya affinis (Lesser Scaup) Bucephula clangula (Common Goldeneye) Bucephula islandica (Barrow's Goldeneye) Bucephula albeola (Bufflehead) Hisrtionicus histrionicus (Harlequin Duck) Melariitta deglandi (White-winged Scoter) Oxyura jamaicensis (Ruddy Duck) Lophodytes cucullatus (Hooded Merganser) Mergus merganser (Common Merganser)

The feather fragment from acute-angled tool 13574 can be identified specifically as originating from a bird in the order Anseriformes. A finer level of classification to a family or species level is not possible because the barbule is not complete, lacking its proximal base and distal end tip. Eider duck (Anatidae Family) down is used in some pillow and duvet blankets but can be ruled out as a contaminate source. The barbule from the tool lacks the heavy stippling of pigment that is diagnostic to eider species (Dove and Peurach 2002:20). It is unlikely that feathers of Anseriforms such as Canada Goose or Mallard were touched by workers who washed and handled the tool at UBC. This feather barbule is almost certainly an authentic ancient residue resulting from using the tool to butcher waterfowl or prepare waterfowl feathers.

Variety and Density of Residues on Tools

Large fragments of conifer wood were found in the residues of the three pounding tools (3674 and 3699 unshaped cobbles with striations, and 3651 a broken maul with a flat base). The vascular tissues from the maul conform to ethnographic expectation of its employment as a woodworking tool, although there is no documentation mentioning the use of unshaped natural cobbles as woodworking 163 tools (see previous discussion of pestles and hammerstones under Archaeological Tools and Features). These tools could have been used to help fell trees with wedges, pound plant materials in wooden bowls, or to prepare temporary camps and mat lodge constructions. Starches, pollens (including spruce and pine bisaccates), articulated epidermal phytoliths, and lichen were also among the residues identified from the pounders, which could indicate plant food processing and other tasks in addition to wood working. In any case, it seems clear that the pounding tools were multipurpose implements.

With the exception of starch, the density of residues found on tool types cannot be commented on, since counts of phytoliths, vessel elements, fungal hyphae, etc. per tool slide were not carried out. With respect to amount of all starch (including those Morphotype A and G types considered to be natural or modern contamination, a total of 151 from 35 tools) found in each general lithic type, 20 acute-angled tools provided 45% of the starch, followed by 11 steep-angled tools (42%), and 3 cobble/ groundstone tools (13%). Only 26 of the total of 106 lithic tools processed contained non-contaminate starch grains. With respect to amount of non- contaminate starch (n=85 starch grains from 26 tools), 14 acute-angled tools provided 35% of the starch, followed by 10 steep-angled tools (53%), and 2 cobble/ groundstone tools (12%).

Tool Form and Tasks

Woodworking, plant processing and hunting activities were suggested by the residue types present on the tools. Woodworking functions were inferred by the presence of tracheids and vessel elements on the tools, and plant processing was inferred by the presence of plant tissue fragments and articulated epidermal phytoliths. Starch was not used in the criteria to assign a tool as a plant processing tool. Since Appendix B counts of starch include natural contaminate Wheatgrass starch and modern contaminate maize starch, all residue types excluding starch were used as indicators of tasks, therefore the interpretations presented below for 164 plant processing are conservative. Hunting was inferred by the presence of a waterfowl feather barbule on a acute angled tool. The following lists the number of tools in each tool category and the tasks represented by the residues present in each tool type.

CS2 Acute-analed tools (n=33^ woodworking= 12 plant processing= 5 woodworking and plant processing= 2 bird hunting= 1

CS1 Acute-analed tools (n=12^ woodworking= 5 plant processing= 0 woodworking and plant processing= 0

CF18 Acute-analed tool (n=1) woodworking= 1 plant processing= 0 woodworking and plant processing= 0

CF9 Acute-angled tool (n=1) woodworking= 0 plant processing= 0

CS2 Steep-angled tools (n=29) woodworking= 14 plant processing= 14 woodworking and plant processing= 7 165

CS1 Steep-analed tools (n=8) woodworking= 1 plant processing= 2 woodworking and plant processing= 0

CS2 Microblades (complete and fragmentary) (n=18) woodworking= 7 plant processing= 3 woodworking and plant processing= 1

C$2 Macroblade (n=i) woodworking and plant processing= 1

CF19 Cobbles with stnations (n=2) woodworking= 2 plant processing= 2 woodworking and plant processing= 2

CF9 Broken maul (n=-h woodworking and plant processing= 1

Some classes of tools contained residues that indicated they were multifunctional. Both woodworking and plant processing were found on many tools including: 7 steep edge angled tools, 2 acute angled tools, 1 microblade, 1 macroblade, 2 cobbles with striations, and 1 broken maul. Tasks involving wood and plant processing are clearly a theme across all tool categories. 42% of the stone tools (n=106) showed microscopic evidence of woodworking, while 26% showed evidence of plant processing, and 13% showed evidence of both tasks.

Somewhat surprising is the number of steep-angled or 'scraping', and acute-angled or 'cutting' tools that have evidence of woodworking, since the only 166 ethnographically documented woodworking tasks that took place at root processing sites are felling trees for firing earth ovens, and collecting vegetation (branches, twigs, skunk cabbage leaves, etc.) to protect roots while baking. Tasks associated with root gathering and processing required wood tools such as digging sticks and baskets, and these could have been made or repaired on site with acute-angled and steep-angled stone tools. The high incidence of woodworking (42% in the tool sample) might also correspond to the production and use of wood mauls or pestles, artifacts that seldom preserve. Perhaps as hunting forays took place from root processing base camps, shafts for weapons such as bows and arrows, spears or macroblades were made and maintained. There is also chance that microblades (most of which are too small to comfortably fit in the hand) were hafted into wood handles as well.

There were far fewer pounding stone tools represented in the assemblage from White Rock Springs than one might expect, given that the ethnographic record makes reference to them being used to mash berries, nuts and roots. It was suggested to the author that perhaps the tools being used to pound berry, nut, and root resources were primarily made of wood (Scott Raymond pers. comm.).

Although the specific tasks of plant processing cannot be determined at this time, residue analysis has shown that plant processing certainly did occur with at least 26% of the stone tools from White Rock Springs. These tasks that were carried out with cutting, scraping, and pounding tools may have involved cutting stems from roots, slicing large roots into pieces, peeling the cortical tissue or 'skin' of roots, mashing roots or perhaps wheatgrass seeds, splitting rose hips or cherries, among others. Interestingly, starch granules were found on steep edge angle tools that would normally be seen as scrapers (probably as hide scrapers). This could suggest that people were scraping or peeling root 'skins' with steep edge angled tools. Acute angle tools showed low amounts of starch overall. Perhaps these were primarily meat cutting tools or used in cutting plant stems, where little storage starch would be expected (Brian Kooyman, pers. comm.). 167

Absent Residues

Calcium oxalate (CaOx) is a highly insoluble crystalline salt of oxalic acid and calcium. CaOx crystals are known to occur in most plant families (Franceschi and Nakata 2005:42; Prychid and Rudall 1999:727) and it is therefore reasonable to suspect that they may be found in tool residues. However, raphides (needle shaped), styloid (stylus or pen shaped), druse (rounded cluster), rosette or other crystal conformations of calcium oxalate were not observed in any tool residue extracts. Calcium oxalate crystals are present in cacti and have been recovered from archaeological tool residues (eg. Sobolik 1996:461). Hair was not found in any of the residue extracts from the tools. Items that were possible hair shafts were rejected upon further analysis due to lack of appropriate features such as an identifiable medulla, cuticular scales, pigment structures and colour bands (Hausman 1920:497; Titus 1980:26-31; Tobin 2005:35-40).

Results of Controls

For every set of tools processed for residue extraction and slide mounting (usually 11 tools at a time), one control was also processed. A total often controls were mounted and checked. This allowed the testing of materials and the laboratory environment itself for modern contaminate starch and other possible contaminate materials. Often residue extraction controls, two (Control May 14,2010 9989-9983 and Control May 22, 2010 6344-11371) were found to contain one starch grain each. A second set of controls were used to test the main working surface in the lab for ambient environmental starch by setting out a sterilized, starch-free microscope slide on the back counter for each of the two weeks residue extraction was underway. One of these controls, back counter Control 2 was found to contain two starch grains. One of the two grains were identified as a hard endosperm Zea mays (Figs. 6-70 to 6-72). Although only 4 starch grains were recovered in the residue extraction and back counter controls combined, the results do indicate that 168 a low level of corn starch contamination can be expected in laboratory settings even when rigorous precautions are taken.

Fig. 6-70. Hard endosperm corn starch from back counter Control 2, 400x. Note polygon shape, open hilum and X fissure.

Fig. 6-71. Hard endosperm corn starch from Fig. 6-72. Same in cross polarized light, back counter Control 2, rotated, 400x. 400x.

All controls were also inspected for non starch contaminants that might be found in the residue extracts from the lithic artifacts. Modern threads were identified in four of the ten extraction controls and two of two of the back counter controls (Figs. 6-73 and 6-74). Most of these threads were the bright electric blue colour of the nurse scrub top that was worn at all times in the lab, showing that clothing worn by the worker does enter into residue slide preparations. 169

Fig. 6-73. Blue thread from residue extraction Fig. 6-74. Blue thread from residue extraction Control 9989-9983, 400x. Control 11356-2417, 400x.

A hair was identified in one residue extraction control, probably from the author (Figs. 6-75 and 6-76). My hair was always pulled away from my face and worn up in a 'top-knot' to try to prevent it from entering residue slide preparations. Further measures such as use of a hat, hair net, shower or swimming cap may need to be taken to definitively stop the workers hair from entering samples. No phytoliths, tracheids, vessel elements, calcium oxalate crystals, plant tissue fragments, pollen, fungal hyphae, or animal cells were found in any of the controls.

/ J

Fig. 6-75. Hair from Control 2294-1951, 400x. Fig. 6-76. Hair from Control 2294-1951, 400x. Note cuticle and scales. Note scales. 170

Summary of Results

This preliminary study has shown that a wide range of microscopic residues including starch, fragments of woody and/or herbaceous material (tracheids and vessel elements), phytoliths, pollen and spores, feather fragments, fungal hyphae and lichen are recoverable from stone tool assemblages. Similar residues can be expected to be recovered in lithic residue extracts that are excavated from similar environments at other earth oven/plant resource procurement sites on the Canadian Plateau. Woodworking evidence (tracheids and vessel elements) was present on 42% of the stone tools (n=106), and at least 26% of tools had evidence of plant processing (articulated epidermal phytoliths, plant tissue fragments, pollen). Residue analysis indicates that Late Period inhabitants of White Rock Springs used a varied toolkit for multiple tasks. At least 13% of the stone tools were multifunctional, being used for both woodworking and plant processing. 44% of all starch grains described were considered contaminants. The modern contaminant was corn (Zea mays), and the natural contaminant was Bluestem Wheatgrass (Pascopyrum smithii), or a species closely related to Bluestem Wheatgrass. 85 starch grains that were not identified as corn or Bluestem Wheatgrass were found on 26 lithic tools, including acute and steep-angled utilized flakes, acute and steep-angled retouched flakes, a natural cobble with striations, a broken maul, a formed uniface, and fragmentary bifaces and microblades, showing that these tools were involved in plant processing. The three pounding tools contained large conifer wood fragments (including pieces identified as Douglas Fir (Pseudotsuga menziesii)), starch, pollen (including spruce (Picea sp.) and pine (Pinus sp.)), articulated epidermal phytoliths, evidence that these multipurpose tools made contact with wood and vegetable material. 171

Pounding tools yielded the highest number and diversity of residues. Mauls, pestles, cobbles with use wear, grinding slabs, and stone bowls are artifacts that should be targeted for future residue analysis. Grass phytoliths (including subfamilies Pooideae and Panicoideae) were found in all 106 tool residue extracts. Bird processing is documented by a waterfowl (order Anseriformes) downy feather barbule in the residue from an acute-angled lithic tool. Animal feathers and tissues are extremely rare in archaeological residues, but it must be born in mind that animal cells and tissues are especially difficult to identify morphologically. The presence of animal proteins that represent blood, tissue, excretions or secretions on the tools cannot be ruled out. Residue plaques that would suggest animal tissue processing were not visible macroscopically on the surfaces of any tools. However, the presence/absence of animal proteins cannot be confirmed, since no biochemical methods or immunological techniques to identify proteins from specific orders or subfamilies (techniques such as cross-over immunoelectrophoresis (CIEP) or enzyme- linked-immunoabsorbent assay (ELISA), could be carried out, as recommended by methods recommended by Newman et al. (1997:1025). This study has shown that fungi and lichens can successfully be identified microscopically in residues from stone tools. However, their age, preservation potential, and species determinations are unknown. Fungal hyphae were present on 90% of the tools, with at least three types identified. Currently, most residue workers consider fungal material as a contaminant. However, hyphae identification should be developed since fungi and lichens are ethnographically known food taxa. Traditionally significant fungal and lichen species, such as the earth oven cooked Black Tree Lichen (Bryoria fremontii), and edible mushroom species should be included in a modern fungal reference collection. From this residue analysis it appears that many tool types classed by assumed function should be reconsidered as potential multifunctional tools. 172

CHAPTER 7: CONCLUSIONS

Goals Revisited

It seems salient to revisit the research goals of this study and evaluate their success. This work was preliminary and exploratory, and it is important to review inadequacies so further residue analyses can profit from research design flaws. Each research goal outlined in the introduction will be restated with a discussion to assess the degree to which the goal was achieved, followed by recommendations.

1) Learn what types of organic residues can be removed from ancient Canadian Plateau stone tools and viewed with light microscopy.

This goal was met. Plant, animal, and fungal residues were discovered in the residue extracts from the White Rock Springs stone tools. These residues included: starch, phytoliths, tracheids, vessel elements, fungal hyphae and spores, plant tissue fragments, and pollen.

2) Describe and illustrate the residues found.

Residues were well-documented in this study. Fourteen characteristics were recorded for each starch grain found on residue slides. Data on the presence of residue types on each slide was also recorded. Over two thousand micrographs (n=2,282) of residues were taken in the course of this study and detailed notes were taken for every slide viewed.

3) Find direct evidence of plant foods. Investigate plant starch as a residue that might reveal information about which root foods were being processed at White Rock Springs. 173

Over 1/4 of all tools had residues indicative of plant processing, but whether these residues are from foods cannot be confirmed at the present time. Non-contaminate starch accounted for 56% of all starch grains recorded and corn and wheatgrass (Pascopyrum smithii) starch grains accounted for 44%. Although a good portion of all starch grains retrieved from the tools was identified as originating from contaminate sources, the rest of the starch grains probably can be considered authentic residues from human activity. It is important to note that contaminates do not remove authentic residues from artifacts and their presence did not hamper identification of other residues. The presence of softwood and hardwood/ herbaceous vascular tissues in residue extracts support the idea that some of the starch found was from use, not from contamination. The main source of contamination is suspected to be preventable by foregoing lithic analysis (which involves washing and handling) until residue extractions are completed. If residue analysis of Canadian Plateau artifacts is to be conducted in the future, more effort should be directed into protecting the artifacts from modern contaminates such as starch, air-borne pollen, dust, newspaper and paper towels that may leave woody fragments on tools, etc. Artifact collection and lab procedures need to be controlled to reduce the chance that modern residues come in contact with artifacts; these contaminates can make residue interpretation more time-consuming and difficult to conduct.

Many modified starches can be inferred from the assemblage. Most of the starch grains encountered had extinction crosses that were extremely faint to almost non­ existent. A clear extinction cross is expected in native starch grains that are raw and unaltered. Whether the cause of starch modification is natural taphonomic agents or due to human activities, such as tool placement close the heat of earth ovens, is unknown.

Today, Balsamroot plants are abundant around the White Rock Springs site area. It is reasonable to posit that past peoples would have came to this upland locale to harvest and pit cook the edible underground storage organs of Balsmroot plants. 174

Indeed, for Interior Plateau people Balsamroot was one of the most important plant foods (Chambers 2001:149; Chambers et al. 2002:40). Unfortunately, Balsamroot contains negligible starch (Mullin et al. 1997:771), making it difficult to identify in stone tool residues. Kooyman and Kooyman also found Balsamorhiza sagitatta roots in the BC/AB starch reference collection did not exhibit any starch grains (2010), likely because Balsamroot stores most carbohydrates as inulin, not as starch. However, there remains the possibility that Balsamroot processing might be represented by non-starch residues such as distinctive epidermal tissue fragments or epidermal phytoliths adhering to lithic tools. There may also be specific types of lithic use wear, such as micro-striations from peeling roots, that can indicate processing of this important resource.

Nodding onion is also very common at White Rock Springs and based on ethnographic accounts was eaten by all Canadian Plateau people and used in many dishes (Turner 1997:61-62). Unfortunately, Allium spp. stems ('roots') in the BC/AB starch reference collection did not exhibit any starch grains (Kooyman and Kooyman 2010). Scott Cummings also reported that Allium prepared for starch reference material lacked starch (1994:180).

4) Discuss where residues are likely to have originated.

This could have been substantially improved and more confidently established. The pre-extraction handling and washing of artifacts that occurred is deemed inappropriate and undesirable for any future residue analysis. However, the procedures of residue extraction, slide preparation, microscopic viewing, starch grain identification, and data compilation are believed to be suitable for collecting and examining a wide range of residues with light microscopy.

The presence of 85 starch grains that were not identified as corn or wheatgrass found on 26 stone tools (56% of all starch) is encouraging. This means that 26 of the 106 tools processed show evidence of starch that indicates they were used on 175 plants, likely on plant portions that were richer in starch. A specific taxon -the Rosaceae Family- was indicated for Morphotype B starch grains, a family which includes Cherries (Prunus spp.), Raspberries (Rubus spp.), Strawberries (Fragaria spp.), Rose hips (Rosa spp.), and Saskatoon Berry (Amelanchier alnifolia).

Ancient starch survival in soils is expected to be low due to rapid starch digestion by microorganisms. Even though there is a low chance of finding starch (especially ancient starch) in soils, the author believes soils should still be tested to confirm that starch is present only on artifacts. The assumption of starch-free soils surrounding the stone tools tested in this study should be confirmed. Any starch found in soils is likely relatively young, although more taphonomic studies in different soil types are required. Soil samples were taken in each area of the White Rock Springs site, however these were not incorporated into analysis here due to the fact that even if no starch was found in the soil samples, this would not increase the confidence in the results because starch being absent in the soil fits the expectation of zero preservation of starch. At least some soil samples need to be taken close to where artifacts are recovered to compare representation and frequencies of residue types. These soil samples are important to establish that the residues on the tools are not from the natural constituents of the sediments in which the tools are buried. Taking and examining the microfossils that can be found in soil samples at surface level of the site, near the location where tools are excavated and also at other off-site locations should be part of any residue analysis study.

5) Suggest how specific residues can be related to food or technological human use in the past.

The results of this study provide information about food and technological activities at a Canadian Plateau earth oven site. At least 42% of the stone tools showed evidence of woodworking, and 26% of the stone tools showed evidence of plant processing at White Rock Springs. Although residue analysis is just emerging as a 176 method of archaeological inquiry on the Canadian Plateau, these results should encourage further residue studies, here and elsewhere on the Canadian Plateau. Paleoethnobotany and residue analysis remain valuable research tools to address ancient subsistence and technology and more of this type of work should be done. I have offered various suggestions as to how the procedures used in this study can be improved the interpretative power of residue analysis.

Unidentified starch comprising 56% of all starch recovered was found on acute and steep-angled utilized flakes, acute and steep-angled retouched flakes, a natural cobble with striations, a broken maul, a formed uniface, and fragmentary bifaces and microblades. The identity and origin of these starch grains is unknown, as they did not provide a clear match to starch in the BC/AB reference collection. The broken maul tool had fragments of softwood in its residue extracts that could suggest the tools were used for woodworking activities. Bird hunting did occur at White Rock Springs, as supported by the presence of the waterfowl feather barbule from a cutting tool. Eddy also found faunal evidence of bird hunting- at least six Spruce Grouse (Falcipennis canadensis) were recovered at the Campsite 1 and Campsite 2 activity areas (2011:50). This evidence corresponds to an ethnography that states hunting activities by men took place in addition to collecting plant foods at seasonal exploitation locales (Smith 1998:10).

Implications

This study takes a novel microscopic approach to understand how Canadian Plateau peoples used plant resources for food and technology. Residue analysis has provided the first direct evidence of the act of preparing plants with stone tools on the Canadian Plateau. Specifically, tasks related to woodworking, plant processing, and hunting were documented. This residue analysis supports previous archaeological interpretations that many kinds of plant and animal subsistence activities took place at the White Rock Springs site in the Late Period (4500 to 200 BP). There is also a good match between the residue analysis results 177 and the ethnographic record. Ethnographic information details task groups of women, probably with some men and children as well, setting up temporary camps in upland root gathering grounds and cooking the roots in earth ovens. Cooking roots is strongly associated with women in ethnographic accounts, with some noting it as exclusively a women's activity. Arguably, the microscopic evidence of plant processing (plant tissue fragments, articulated epidermal phytoliths, and especially starch) found on the White Rock Springs tools documents women's subsistence roles.

In addition to root collecting and cooking, the ethnographic record makes clear that other subsistence activities such as hunting and pine nut collecting were carried out at earth oven sites. Visits to earth oven sites centred around digging and cooking root resources while other subsistence multitasking occurred simultaneously, and this is in agreement with the residue analysis results. Eddy (2011) and McGill et al. (2010) both found evidence for hunting game from animal bones recovered from cultural contexts White Rock Springs; this supports the microscopic evidence found here for bird hunting. Furthermore, projectile points have been recovered in different areas of the site; the points were likely used for hunting. No projectile points were included in residue analysis, although this artifact class to could be tested in the future for plant residues such as tracheids, vessel elements, or hafting resins.

Recommendations

There is certainly great opportunity for future residue analysis studies on the Canadian Plateau and elsewhere. Loy studied microscopic residues in situ on stone tools and also conducted protein and aDNA analysis from tools from northern British Columbia at the Toad River Canyon site and discovered that many tools contained residues from bison (1993). This is the only other residue study of an archaeological artifact from British Columbia known to the author. Clearly, there is potential for growth and development of residue analysis. 178

Pounding Tools and Future Research Potential With regard to lithic type, the three stone pounding tools yielded the highest number and diversity of residues. Although it should be kept in mind that residues collected from the pounding tools came from the 'working surface' and represented a large surface area, it seems clear that pounding tools (groundstone mauls, pestles, hand-held grinding stones, shaped or unmodified cobbles with use wear evident), and likely also their lower implement counterparts (eg. mortars, tabular grinding slabs, mill stones, grinding dishes, stone bowls) provide the highest potential for further residue analyses (see Fig. 6-77). In the future, unaltered cobbles that display wear marks such as striations, percussion damage, pitting, roughened surfaces, polish formation, rounding, and small scars should be identified and targeted for residue analysis.

Fig. 6-77 Haridmauls with residues and use wear evident on their working surfaces like this one from BC at the University of Calgary Archaeology Museum have good potential for residue analysis.

Contamination Contamination needs to be considered a major issue in any residue analysis. Corn starch is ubiquitous and likely contaminated the artifacts used in this study during the washing and handling that took place at University of British Columbia before 179 residue analysis was conducted. There are some precautions that can be taken to prevent and reduce contamination during excavation before residues are removed from artifacts. An assessment of on-site contamination sources and a protocol of appropriate steps of artifact contamination prevention should be written prior to excavations. This protocol should be distributed and explained to all crew and any visiting excavators. Artifacts of interest should be bagged individually and immediately in new zip lock bags with adhering dirt. Lunch and smoke breaks need to be held away from open excavations and crew members should wash their hands before returning to digging. Artifacts should be handled in a strictly controlled manner until all residue extractions are complete. Plastic gloves need to be tested and examined for starch and other residues that could resemble archaeological residues. A different set of these clean residue-free gloves should be worn for each tool handled.

It became apparent that there is a need to control the clothing worn to process residue extractions and make slide preparation. This precaution recommended by Dr. Kooyman turned out to be a wise one, since the bright blue thread of the scrub top was commonly present in the lithic residues and 53% of 106 slides contained modern threads. The person conducting residue extraction and slide mounting should consistently wear clothing of an unnatural colour that cannot be mistaken for any organic material. The worker should also not wear make-up, hair products, or starched shirts and should contain their hair.

It must be established that residues from excavated tools are found in far greater numbers than the soils surrounding the artifact. Soil samples of the site surface, near artifacts and off-site locations can be taken to increase the validity of residue analysis results. Soil and sediment samples were collected from the White Rock Springs at on and off-site locations. These will be processed in winter 2011/2012, to form a comparative basis for microfossil finds. 180

Appropriate field and lab handling of artifacts to minimize modern contamination in residue samples are summarized below.

Field Excavation and Handling of Artifacts for Residue Analysis • Hold lunch, coffee, and smoke breaks away from open excavations. • Ideally, crew should wash hands after eating and smoking before returning to excavation. • Do not use paper bags to hold artifacts that may be used in residue analysis. • During excavation, immediately bag any artifact that could be selected for residue analysis with adhering sediment into a new zip lock plastic bag. • Artifact ID paper cards with provenience information, catalog numbers, etc. should not come into contact with the artifact. ID cards should be bagged in their own zip lock bag. • The bagged artifact and bagged ID card should then be placed together in another zip lock bag (i.e. 3 bags per artifact). • Store all bagged artifacts in large zip lock bags labelled with 'DO NOT OPEN- ARTIFACTS SELECTED FOR RESIDUE ANALYSIS'. Store separately and away from soil samples, macrobotanical samples, flotation samples, etc.

Lab Handling of Artifacts for Residue Analysis • Artifact residue analysis must take place before any other analysis. • Do not wash or label artifacts. • Only use lab supplies that have been shown to be starch-free and that do not release any residues that could be mistaken for other residue types (such as phytoliths, vascular tissues, pollen, fungal remains, etc.). Before residue extraction and before any artifact bags are opened, make sure poly gloves, hand soap used by worker(s), and all other lab supplies (centrifuge tubes, glass slides, glycerol, etc.) that will come in contact with the tool or residue extraction are tested for the presence of starch and other residues. • Do not wear make-up, hair products, body powders, or starch ironed shirts when opening artifact bags, processing residue extractions, mounting residues on 181

slides, cleaning the lab and/or supplies. Long hair should be worn up, preferably in a hair net or shower cap. • Wear a scrub top (like those used by nurses) in a distinctly unnatural colour (neons work well) at all times while handling artifacts or making microscope slides of artifact residues. • Upon entering the lab or returning from washroom/lunch breaks, wash hands and dry with cotton cloth. • Wipe down counters with cotton cloths that has been washed with bleach and without detergent. Only use cotton cloths and remove paper towels and facial tissues from the lab completely. Paper towels can introduce wood vascular fragments to the residue samples. • Clean the floors of the lab. Use a Swiffer sweeper or other system of wet disposable cleaning cloths to collect dirt, dust, hair, etc. • Use plastic wrap to cover table area where lithics and poly gloves will be laid out. Do not re-use plastic wrap; lay out new sheets for every set of artifacts to be processed. • Lay out the artifacts (number of artifacts depends on how many centrifuge tubes can be spun at once) to be processed for residue extractions on the plastic wrap covered surface, with about 30 cm of space between each tool. • Never touch artifacts with bare hands. Handle artifacts with a gloved hand. Use a new poly glove for every artifact handled and label it with the artifact number.

Residue Extraction Procedure I recommend that if-and only if- selected artifacts display residue plaques visible to the naked eye, that a targeted approach to residue removal be attempted. Investigative handling to locate residue plaques need not conflict with the goal of maintaining low exposure to residue contamination if it is conducted with gloves proven to be starch and residue free. Five mm2 areas should be sampled in plaque and other non-plaque areas of the artifact and the locations of residue testing should be recorded and mapped. In this way, residues found in samples from specific locations on the tool can be documented and artifact use can be more 182 accurately and validly reconstructed. This targeted approach provides specific information as to how tools were used that is higher resolution (but more time consuming to conduct) than the tool immersion and total removal of residues approach and provides a better basis for corroboration with use wear analysis. For example, the targeted approach could allow resin and wood residues to be documented as occurring at only one area of the tool, and thus indicate hafting (eg. Lombard 2008; Rots and Vermeersch 2004). The targeted approach can also help eliminate contamination as a possibility if the food residues are found to be concentrated only on the used edges of the tool and not at other locations of the tool surface.

Reference Collection Plants continually move sugars in the phloem, and in temperate environments this movement is dependent on seasonal changes. The majority of sugars are transported from source to sink when photosynthesizing organs (such as mature leaves) of the plant are most active, and mainly from sink to source when new organs (such as buds, shoots, roots and reproductive structures) are growing (Campbell and Reece 2002:762; Raven et al. 2005:680). Dr. Brian Kooyman and Patrick Kooyman have compiled an excellent BC/AB starch reference collection at the University of Calgary. This collection was and is designed with individual plant variation in mind. Every attempt has been made to collect species at different times of year to ensure that lack of starch from a plant is not due to seasonal cyclical changes in starch aggregation in storage organs. Anyone compiling a starch reference collection should prepare multiple plants of the same species to help account for variation. Plant individuals from different microhabitats in the same general region may show different densities of starch even when they are collected at the same time. In the field, plant identification at stages where the plant is not flowering or has no leaves is an obstacle for collection of reference material. 183

Limitations

There are some important limitations of residue analysis. As mentioned in the section on residue taphonomy, there is currently no way to tell if a stone tool, ceramic, or wood implement was used once or over longer periods by the organic food residues alone, since multiple food items might have been processed simultaneously. Whether the assemblage of residues extracted represent a short (expedient tool) or long period (curated tool) of use, is unknown. Use wear studies that investigate the microscopic markings left on the stone tools from human use will let us know how intensively or ephemerally tools were used, and this is part of the overall research plan for the White Rock Springs site (Brian Kooyman, pers. comm.). Use wear analysis by Brian Kooyman is currently underway for the White Rock Springs lithic materials.

Starch or other residues found on lithics cannot suggest tool function on their own (Odell 2001:56), and an examination of use wear can support or refute the results presented here. Simply because residues are found adhering to stone tools does not prove that those residues resulted from human actions such as root tuber processing. Use wear study in collaboration with residue analysis provides a means by which the case for the human-deposition of residues on lithics can be validated. If the starches found on the White Rock Springs lithics are the result of human processing and not from starch in surrounding sediments, then we expect to find starch from only those tools with use wear. It is worth noting again that starch does not preserve well in sediments due mostly to the action of microorganisms that digest starch, and although this is a concern, it is not a major one.

The targeted approach to residue removal can be used powerfully with use wear analysis. If residue plaques visible on the tools resulted from people using them, then one can expect the greatest concentration of residues and coinciding use wear to be found in samples from the working edges only. On the other hand, if the 184 residues on the tools represent post depositional contaminants, then an even distribution of residues over the entire tool surface might be expected. Another approach that might prove successful is use of SEM to scan tools for starch and other residues (Len Hills, pers. comm.). The use of SEM allows very detailed observations to be made, however, examining a large artifact sample would not be feasible with SEM.

The use of multiple lines of evidence is commonly heralded by archaeologists as the best strategy to draw conclusive statements about past peoples, and the author wholeheartedly agrees with this corroborative approach. Convergence of multiple lines of evidence is key to obtaining the most interpretive power from residue analysis results. However, the degree of success achieved with multiple lines of evidence depends on the communication, collaboration and comparison of results between all research parties involved in their separate analyses.

An excellent example of multiple lines of evidence at work in residue analysis is Boyd et al. (2008). This study showed that maize was commonly eaten by past people in the eastern Canadian prairies between ~AD 700 and 1600 by use of three types of supporting analyses, despite a complete lack of archaeological evidence. The same piece of ceramic residue was subjected to trace element analysis, plant microfossil (starch and phytoliths) and stable isotopic (C and N) analysis, all of which indicated the presence of Zea mays. Work like this highlights the very detailed and nuanced dietary models that can be constructed over time and geographical space by drawing from many specialties within archaeology. Here it is apparent that residue analysis used in conjunction with other techniques provides a powerful, objective line of direct evidence of prehistoric activities. Use of multiple lines of evidence can also help put to rest doubts that the morphological traits used to recognize residues are subjective and inaccurate.

The project at White Rock Springs is multidisciplinary and holds great comparative, and thus interpretive, power due to a research design that layers different types of 185 analyses. As outlined previously, project analyses being carried out include: macrobotanical material (wood charcoal, needles, and seeds by Sandra Peacock, Monica Nicolaides, Sheena McKenzie, and Crystal Weins), microbotanical remains (phytoliths by Brian Kooyman and starch by Kooyman and Kooyman, in preparation), lithic remains (Pokotylo), faunal remains (McGill et al. 2010, Eddy 2011, and B. Kooyman), paleoecology (palynology by Brintnell and Hebda) and radiocarbon dating. Varied analyses are being carried out in tandem at White Rock Springs, and their results will be compared to yield mutually supporting evidence of past activities at the site. 186

REFERENCES

Agashe, S. N. & E. Caulton, 2009. Pollen and Spores: Application with Special Emphasis on Aerobiology and Allergy, Enfield: Science Publishers.

Alexander, D., 1992a. Environmental Units, in A Complex Culture of the British Columbia Plateau: Traditional Stl'atl'imx Resource Use, ed. B. Hayden, Vancouver: UBC Press, 47-98.

Alexander, D., 1992b. A Reconstruction of Prehistoric Land Use in the Mid-Fraser River Area Based on Ethnographic Data, in A Complex Culture of the British Columbia Plateau: Traditional Stl'atl'imx Resource Use, ed. B. Hayden, Vancouver: UBC Press, 99-176.

Alexander, D. & R. G. Matson, 1987. Report on the Potato Mountain Archaeological Project (1985), Victoria, B.C.: Laboratory of Archaeology, University of British Columbia, 107.

Ames, K. M. & A. G. Marshall, 1980. Villages, Demography and Subsistence Intensification on the Southern Columbia Plateau. North American Archaeologist, 2(1), 25-52.

Anderson, P. C., 1980. A Testimony of Prehistoric Tasks: Diagnostic Residues on Stone Tool Working Edges. World Archaeology, 12(2), 181-94.

Ascher, R., 1961. Analogy in Archaeological Interpretation. Southwestern Journal of Anthropology, 17(4), 317-25.

Babot, M. P., 2003. Starch Grain Damage as an Indicator of Food Processing. Terra Australis, Phytolith and Starch Research in the Australian-Pacific- Asian Regions, 19, 69-81. 187

Babot, M. P. & M. C. Apella, 2003. Maize and Bone: Residues of Grinding in Northwestern Argentina. Archaeometry, 45(1), 121-32.

Badenhuizen, N. P., 1969. The Biogenesis of Starch Granules in Higher Plants, New York: Appleton-Century-Crofts.

Baldwin, P. M., M. C. Davies & C. D. Melia, 1997. Starch Granule Surface Imaging Using Low-voltage Scanning Electron Microscopy and Atomic Force Microscopy. International Journal of Biological Macromolecules, 21(1-2), 103-7.

Balme, J. & W. E. Beck, 2002. Starch and Charcoal: Useful Measures of Activity Areas in Archaeological Rockshelters. Journal of Archaeological Science, 29 (2), 157-66.

Banks, W. E., 2006. Artifacts as Landscapes: A Use-Wear Case Study of Upper Paleolithic Assemblages at the Solutre Kill Site, France, in Confronting scale in archaeology: Issues of Theory and Practice, eds. G. Lock & B. L. Molyneaux, New York: Springer, 89-111.

Barnard, H., S. H. Ambrose, D. E. Beehr, M. D. Forster, R. E. Lanehart, M. E. Malainey, R. E. Parr, M. Rider, C. Solazzo & R. M. Yohe li, 2007. Mixed Results of Seven Methods for Organic Residue Analysis Applied to One Vessel with the Residue of a Known Foodstuff. Journal of Archaeological Science, 34(1), 28-37.

Barton, H., 2007. Starch Residues on Museum Artefacts: Implications for Determining Tool Use. Journal of Archaeological Science, 34(10), 1752-62. 188

Barton, H., 2009. Starch Granule Taphonomy: The Results of a Two Year Field Experiment, in Archaeological Science Under a Microscope: Studies in Residue and Ancient DNA Analysis in Honour of Thomas H. Loy, eds. M. Haslam, G. Robertson, A. Crowther, S. Nugent & L. Kirkwood, Canberra: Australian National University Press, 129-40.

Barton, H., R. Torrence & R. Fullagar, 1998. Clues to Stone Tool Function Re­ examined: Comparing Starch Grain Frequencies on Used and Unused Obsidian Artefacts. Journal of Archaeological Science, 25(12), 1231-8.

Barton, H. & J. P. White, 1993. Use of Stone and Shell Artifact at Balof 2, New Ireland, Papua New Guinea. Asian Perspectives, 32(2), 169-81.

Beck, W., 1989. The Taphonomy of Plants. Tempus: Archaeology and Material Culture Studies in Anthropology. Plants in Australian Archaeology, 1, 31-53.

Beirne, P. D. & D. L. Pokotylo, 1979. Hat Creek Project: Inventory, Assessment and Evaluation of the Cultural Heritage Resources. Report prepared for the B.C. Hydro and Power Authority, Vancouver: Department of Anthropology and Sociology, University of British Columbia.

Belitz, H.-D., W. Grosch & P. Schieberle, 2004. Carbohydrates, in Food Chemistry Berlin, Germany: Springer, 245-341.

Bello-Perez, L. A., S. L. Rodriguez-Ambriz, M. M. Sanchez-Rivera & E. Agama- Acevedo, 2010. Starch Macromolecular Structure, in Starches: Characterization, Properties, and Applications, ed. A. C. Bertolini, Boca Raton, Florida: CRC Press, 33-58.

BeMiller, J. N. & R. L. Whistler, 1996. Carbohydrates, in Food Chemistry, ed. O. R. Fennema, New York: Marcel Decker Inc., 157-223. 189

Benson, E. M., J. M. Peters, M. A. Edwards & L. A. Hogan, 1973. Wild Edible Plants of the Pacific Northwest. Journal of the American Dietic Association, 62,143-7.

Berkes, F., 1993. Traditional Ecological Knowledge in Perspective, in Traditional Ecological Knowledge: Concepts and Cases, ed. J. T. Inglis, Ottawa: International Program on Traditional Ecological Knowledge and International Development Research Centre, 1 -9.

Berkes, F., 1999. Sacred Ecology: Traditional Ecological Knowledge and Resource Management, Ann Arbor, Ml: Taylor & Francis.

Berkes, F. & N. Turner, 2006. Knowledge, Learning and the Evolution of Conservation Practice for Social-Ecological System Resilience. Human Ecology, 34(4), 479-94.

Bertolini, A. C., 2010. Trends in Starch Applications, in Starches: Characterization, Properties, and Applications, ed. A. C. Bertolini, Boca Raton, FL: CRC Press 1-19.

Bidlack, J. E. & S. H. Jansky, 2011. Stern's Introductory Plant Biology, New York, New York: McGraw-Hill.

Blake, 1974. Ollie Site (EeRk 9) Final Report. Report on file with the British Columbia Archaeology Branch, Victoria, B.C.

Blanshard, J. M. V., 1987. Starch Granule Structure and Function: A Physicochemical Approach, in Starch Properties and Potential. Critical Reports on Applied Chemistry Vol. 13, ed. T. Galliard, Chichester, UK: Society of Chemical Industry by John Wiley & Sons, 16-54. 190

Bowler, P., M. R. Williams & R. E. Angold, 1980. A Hypothesis for the Morphological Changes which Occur on Heating Lenticular Wheat Starch in Water. Starch-Starke, 32(6), 186-9.

Boyd, M. & C. Surette, 2010. Northernmost Precontact Maize in North America. American Antiquity, 75(1), 117-33.

Boyd, M., C. Surette & B. A. Nicholson, 2006. Archaeobotanical Evidence of Prehistoric Maize (Zea Mays) Consumption at the Northern Edge of the Great Plains. Journal of Archaeological Science, 33(8), 1129-40.

Boyd, M., T. Varney, C. Surette & J. Surette, 2008. Reassessing the Northern Limit of Maize Consumption in North America: Stable Isotope, Plant Microfossil, and Trace Element Content of Carbonized Food Residue. Journal of Archaeological Science, 35, 2545-56.

Brennan, L. A., 1975. Artifacts of Prehistoric America, Harrisburg, Pennsylvania: Stackpole Books.

Briuer, F. L., 1976. New Clues to Stone Tool Function: Plant and Animal Residues. American Antiquity, 41(4), 478-84.

Brodo, I. M., S. D. Sharnoff & S. Sharnoff, 2001. Lichens of North America, New Haven: Yale University Press.

Brown, D. A., 1984. Prospects and Limits of a Phytolith Key for Grasses in the Central United States. Journal of Archaeological Science, 11(4), 345-68.

Brown, R., 1868. On the Vegetable Products Used by the Northwest Amercan Indians, as Food and Medicine, in the Arts, and in Superstitious Rites. Transactions of the Botanical Society of Edinburgh (1866-68), 9, 378-96. 191

Budel, B. & C. Scheidegger, 2008. Thallus Morphology and Anatomy, in Lichen Biology, ed. T. H. Nash III, Cambridge: Cambridge University Press, 40-68.

Buleon, A., P. Colonna, V. Planchot & S. Ball, 1998. Starch granules: Structure and Biosynthesis. International Journal of Biological Macromolecules, 23(2), 85-112.

Campbell, N. A. & J. B. Reece, 2002. Transport in Plants, in Biology, San Francisco: Benjamin Cummings, 748-66.

Canti, M. G., 1998. The Micromorphological Identification of Faecal Spherulites from Archaeological and Modern Materials. Journal of Archaeological Science, 25(5), 435-44.

Canti, M. G., 1999. The Production and Preservation of Faecal Spherulites: Animals, Environment and Taphonomy. Journal of Archaeological Science, 26(3), 251-8.

Carlson, C., 1980. Excavation at the Curr Site, in The Archaeology of Kamloops, Burnaby, B.C.: Department of Archaeology, Simon Fraser University.

Chambers, K. J., 2001. Autecology, Ethnobotany and Agronomy of Balsamorhiza sagittata: Nothwestern Plateau, British Columbia, in Interdisciplinary Degree Programme Victoria: University of Victoria, 188.

Chambers, K. J., E. Small, P. M. Catling, N. J. Turner, P. Bowen & C. P. Keller, 2002. Poorly Known Economic Plants of Canada: Arrow-leaved Balsamroot Balsamorhiza sagittata (Pursh.) Nutt. Canadian Botanical Association Bulletin, 35(4), 40-5. 192

Chandrashekar, A., A. Savitri & K. Somashekar, 1987. Optical Interference Studies on Single Starch Granules. Starch - Starke, 39(6), 195-7.

Chatters, J. C. & D. L. Pokotylo, 1998. Prehistory: Introduction, in Handbook of North American Indians: Plateau, ed. D. E. Walker, Washington, D.C.: Smithsonian Institution, 73-80.

Chisholm, B. S., 1986. Reconstruction of Prehistoric Diet in British Columbia Using Stable-carbon Isotope Analysis, in Department of Archaeology Burnaby, B.C.: Simon Fraser University.

Colonna, P., A. Buleon & C. Mercier, 1987. Physically Modified Starches, in Starch Properties and Potential. Critical Reports on Applied Chemistry Vol. 13, ed. T. Galliard, Chichester, UK: Society of Chemical Industry by John Wiley & Sons, 79-114.

Compton, B. D., R. W. Mathewes & G. Guzman, 1995. Puffballs from the Past: Identification of Gasteromycetes from a Lillooet Archaeological Site and Speculation Regarding their Aboriginal Use. Canadian Journal of Archaeology, 19,154-60.

Conn, R. G. & M. D. Schlick, 1998. Basketry, in Handbook of North American Indians, ed. D. E. Walker, Washington: Smithsonian Institution Press, 600-10.

Copeland, L., J. Blazek, H. Salman & M. C. Tang, 2009. Form and Functionality of Starch. Food Hydrocolloids, 23(6), 1527-34.

Core, H. A., W. A. Cote & D. A. C., 1976. Wood Structure and Identification, Syracuse, New York: Syracuse University Press. 193

Cortella, A. & M. Pochettirio, 1994. Starch Grain Analysis as a Microscopic Diagnostic Feature in the Identification of Plant Material. Economic Botany, 48(2), 171-81.

Cousins, S. M., 2000. A Paleoethnobotanical Comparison of Four "Small" Housepits, in The Ancient Past of Keatley Creek, Volume 1: Taphonomy, ed. B. Hayden, Burnaby: SFU Archaeology Press, 87-102.

Croft, S., 2008 A Preliminary Study of Botanical Macrofossil Remains from Selected Sites in Lillooet Area of the Interior Plateau, British Columbia. Manuscript in possession of the author, Burnaby, B.C.: Simon Fraser University.

Croft, S., 2009 Starch Grain Analysis of Three Museum-stored Handmauls and One Earth Oven Stone from British Columbia, Canada. Manuscript in possession of the author, Vancouver, B.C.

Croft, S. & R. W. Mathewes., n.d. Barking up the Right Tree: Understanding the Birch Bark Artifacts of Canadian Plateau Peoples, submitted, Burnaby, B.C.: Simon Fraser University.

Davidson, J., 1915. Botanical Exploration of the Province. Headwaters of Skoonkon, Botanie, Laluwissin, Murray and Twaal Creeks, Between the South Thompson and Fraser Rivers, in Second Annual Report of the Botanical Office of the Province of British Columbia, 1914, Victoria, 44-6.

Davidson, J., 1916. Botanical Exploration of the Bitter-root Grounds and "Three Sisters" Valley, between Spences Bridge and Ashcroft, in Third Annual Report of the Botanical Office of the Province of British Columbia, 1915, Victoria, 134-6. 194

Dawson, G. M., 1891. Notes on the Shuswap People of British Columbia, in Proceedings and Transactions of the Royal Society of Canada, Royal Society of Canada, 3-44.

Deal, M., 1990. Exploratory Analysis of Food Residues from Prehistoric Pottery and Other Artifacts from Eastern Canada. Society for Archaeological Science Bulletin, 13(1), 6-12.

Dering, P., 1999. Earth-Oven Plant Processing in Archaic Period Economies: An Example from a Semi-Arid Savannah in South-Central North America. American Antiquity, 64(4), 659-74.

Dickson, J. H., 1978. Bronze Age Mead. Antiquity, 52(205), 108-13.

Dominguez- Rodrigo, M., J. Serrallonga, J. Juan-Tresserras, L. Alcala & L. Luque, 2001. Woodworking Activities by Early Humans: A Plant Residue Analysis on Acheulian Stone Tools from Peninj (Tanzania). Journal of Human Evolution, 40(4), 289-99.

Donahue, P. F., 1977. 4500 Years of Cultural Continuity on the Central Interior Plateau of British Columbia, in Department of Anthropology Maddison: University of Wisconsin, 475.

Douglas, G. W., D. V. Meidinger & J. Pojar, 2002. Illustrated Flora of British Columbia. Volume 8 General Summary, Maps and Keys, Victoria, British Columbia: Ministry of Environment, Land and Parks. Ministry of Forests.

Douglas, G. W., G. B. Straley, D. V. Meidinger & J. Pojar, 1998. Illustrated Flora of British Columbia. Volume 1: Gymnosperms and Dicotyledons (Aceraceae Through Asteraceae), Victoria: B.C. Ministry of Environment, Lands & Parks and B.C. Ministry of Forests. 195

Dove, C. J. & S. L. Koch, 2010. Microsopy of Feathers: A Practical Guide. Journal of American Society of Trace Evidence Examiners, 1 (1), 15-61.

Dove, C. J. & S. C. Peurach, 2002. Microscopic Analysis of Feather and Hair Fragments Associated with Human Mummified Remains from Kagamil Island, Alaska, in Ethnographical Series Washington, DC: Smithsonian Institution.

Eccleston, K., 1999. The Applicability of Phytolith Analysis to the Columbia Plateau: The Cottonwood-Divide Creek Site, A Test Model, in Anthropology, Moscow: University of Idaho, 281.

Eddy, J., 2011. Accompanying Root Processing: Faunal Exploitation in the Lillooet Region of the Canadian Plateau, Manuscript in possession of the author, University of Calgary, Calgary.

Eerkens, J. W. & H. Barnard, 2007. Introduction, in Theory and Practice of Archaeological Residue Analysis, BAR International Series 1650, eds. H. Barnard & J. W. Eerkens, Oxford: Archeopress, 1-7.

Endo, N., 2009. Bridge River Excavation 2008: Archaeobotanical Analysis, in Report of the 2008 University of Montana Investigations at the Bridge River Site (EeRI4), eds. A. M. Prentiss, E. Carlson, N. Crossland, H. Schremser & L. Reininghaus, Missoula, Montana: University of Montana, 132-8.

Endo, N., 2010. Bridge River Excavation 2009: Archaeobotanical Analysis, in Report of the 2009 University of Montana Investigations at the Bridge River Site (EeRI4), eds. A. M. Prentiss, L. Smith, L. Reininghaus, M. Schirack, M. Wanzenried & O. Ward, Missoula, Montana: University of Montana, 143-8. 196

Evers, A. D., 1969. Scanning Electron Microscopy of Wheat Starch I. Entire Granules. Starch - Starke, 21 (4), 96-9.

Evers, A. D., 1971. Scanning Electron Microscopy of Wheat Starch. III. Granule Development in the Endosperm. Starch - Starke, 23(5), 157-62.

Evershed, R. P., 2008. Organic Residue Analysis in Archaeology: The Archaeological Biomarker Revolution. Archaeometry, 50(6), 895-924.

Evert, R. F., 2006. Xylem: Cell Types and Developmental Aspects, in Esau's Plant Anatomy, Meristems, Cells, and Tissues of the Plant Body: Their Structure, Funcion, and Development Hoboken, New Jersey: John Wiley & Sons, 255-90.

Fankhauser, B. L., 1993. Residue Analysis on Maori Earth Ovens, in Archaeometry: Current Australasian Research, Occasional Papers in Prehistory, No. 22 eds. B. L. Fankhauser & J. R. Bird, Canberra, Australia: Research School of Pacific Studies, Australian National University, 13-37.

Fannon, J. E., R. J. Hauber & J. N. BeMiller, 1992. Surface Pores of Starch Granules. Cereal Chemistry, 69(3), 284-8. Farrar, J. L., 1995. Trees in Canada, Markham, ON: Fitzhenry & Whiteside Ltd. and Canadian Forest Service.

Fladmark, K. R., 1982. An Introduction to the Prehistory of British Columbia. Canadian Journal of Archaeology, 6, 95-156.

Fladmark, K. R., 1986. British Columbia Prehistory. National Museum of Man.

Franceschi, V. R. & P. A. Nakata, 2005. Calcium Oxalate in Plants: Formation and Function. Annual Review of Plant Biology, 56(1), 41 -71. 197

Fredlund, G.G., Tieszen, L.T., 1994. Modern Phytolith Assemblages from the North American Great Plains, Journal of Biogeography 21, 321-335.

Friedman, J., 1978. Wood Identification by Microscopic Examination: A Guide for the Archaeologist on the Northwest Coast of North America, Victoria: British Columbia Provincial Museum.

Frost, F. H., 1929. Histology of the Wood of Angiosperms. I. The Nature of the Pitting Between Tracheary and Parenchymatous Elements. Bulletin of the Torrey Botanical Club, 56(5), 259-64.

Fullagar, R., 1986. Use-wear and Residues on Stone Tools: Functional Analysis and its Application to Two Southeastern Australian Archaeological Assemblages, in Archaeology Melbourne: LaTrobe University.

Fullagar, R., 1988. Recent Developments in Australian Use-wear and Residue Studies, in Industries Lithiques: Traceologie et Technologie, ed. S. Beyries, Oxford: British Archaeological Reports International Series, 41 (ii) 133-45.

Fullagar, R., 1989. The Potential of Lithic Use-wear and Residues Studies for Determining Stone Tool Functions, in A Crack in the Spine: Prehistory and Ecology of the Jimi-Yuat Valley, Papua New Guinea, eds. P. P. Gorecki & D. Gillieson, Queensland: James Cook University, Townsville, 209-23.

Fullagar, R. & B. David, 1997. Investigating Changing Attitudes Towards an Australian Aboriginal Dreaming Mountain Over >37,000 Years of Occupation via Residue and Use Wear Analyses of Stone Artefacts. Cambridge Archaeological Journal, 7(1), 139-44.

Fullagar, R., J. Field & L. Kealhofer, 2008. Grinding Stone and Seeds of Change: 198

Starch and Phytoliths as Evidence of Plant Processing, in New Approaches to Old Stones: Recent Studies of Ground Stone Artifacts, eds. Y. M. Rowan & J. R. Ebeling, London: Equinox Publishing, 159-72.

Gallant, D. J., B. Bouchet & P. M. Baldwin, 1997. Microscopy of Starch: Evidence of a New Level of Granule Organization. Carbohydrate Polymers, 32(3-4), 177-91.

Galliard, T., 1987. Starch Availability and Utilization, in Starch Properties and Potential. Critical Reports on Applied Chemistry Vol. 13, ed. T. Galliard, Chichester, UK: Society of Chemical Industry by John Wiley & Sons, 1-15.

Galliard, T. & P. Bowler, 1987. Morphology and Composition of Starch, in Starch Properties and Potential, Critical Reports on Applied Chemistry Vol. 73, ed. T. Galliard, Chichester, UK: Society of Chemical Industry by John Wiley & Sons, 55-78. Gaytan-Martlnez, M., J. D. Figueroa-Cardenas, M. L. Reyes-Vega, F. Rincon- Sanchez & E. Morales-Sanchez, 2006. Microstructure of Starch Granule Related to Kernel Hardness in Corn. Revista Fitotecnia Mexicana, 29 (numero especial 2), 135-9.

Gerards, T., F. Damblon, B. Wauthoz & P. Gerrienne, 2007. Comparison of Cross- field Pitting in Fresh, Dried and Charcoalified Softwoods. IAWA Journal, 28 (1), 49-60.

Gibson, G. R. & N. Delzenne, 2008. Inulin and Oiigofructose: New Scientific Developments. Nutrition Today, 43(2), 54-9.

Gilbert, B. M., L. D. Martin & H. G. Savage, 1985. Avian Osteology, University of California: Bone Books. 199

Gott, B., H. Barton, D. Samuel & R. Torrence, 2006. Biology of Starch, in Ancient Starch Research, eds. R. Torrence & H. Barton, Walnut Creek, California: Left Coast Press, 35-46.

Greenland, D. J. & J. M. Oades, 1975. Saccharides, in Soil Components, ed. J. E. Gieseking, New York: Springer-Verlag, 213-61.

Guggenberger, G., E. T. Elliott, S. D. Frey, J. Six & K. Paustian, 1999. Microbial Contributions to the Aggregation of a Cultivated Grassland Soil Amended with Starch. Soil Biology and Biochemistry, 31(3), 407-19.

Haile, J., R. Holdaway, K. Oliver, M. Bunce, M. T. P. Gilbert, R. Nielsen, K. Munch, S. Y. W. Ho, B. Shapiro & E. Willerslev, 2007. Ancient DNA Chronology within Sediment Deposits: Are Paleobiological Reconstructions Possible and Is DNA Leaching a Factor? Molecular Biology and Evolution, 24(4), 982-9.

Hall, D. M. & J. G. Sayre, 1973. A Comparison of Starch Granules as Seen by Both Scanning Electron and Ordinary Light Microscopy. Starch - Starke, 25(4), 119-23.

Hall, J., S. Higgins & R. Fullagar, 1989. Plant Residues on Stone Tools, in Plants in Australian Archaeology, eds. W. Beck, A. Clarke & L. Head, Queensland: University of Queensland, 136-60.

Hallett, D. J., D. S. Lepofsky, R. W. Mathewes & K. P. Lertzman, 2003.11 000 Years of Fire History and Climate in the Mountain Hemlock Rain Forests of Southwestern British Columbia Based on Sedimentary Charcoal. Canadian Journal of Forest Research, 33(2), 292.

Ham, L. C., 1979. Hat Creek Faunal Analysis. Manuscript on file in Laboratory of Archaeology, Vancouver: Department of Anthropology and Sociology, 200

University of British Columbia.

Hardy, B. L., 2004. Neanderthal Behavior and Stone Tool Function at the Middle Palaeolithic site of La Quina, France. Antiquity, 78(301), 547-65.

Hardy, B. L., M. Bolus & N. J. Conard, 2008. Hammer or Crescent Wrench? Stone- tool Form and Function in the Aurignacian of Southwest Germany. Journal of Human Evolution, 54(5), 648-62.

Hardy, B. L. & G. T. Garufi, 1998. Identification of Woodworking on Stone Tools Through Residue and Use-Wear Analyses: Experimental Results. Journal of Archaeological Science, 25(2), 177-84.

Haslam, M., 2004. The Decomposition of Starch Grains in Soils: Implications for Archaeological Residue Analyses. Journal of Archaeological Science, 31 (12), 1715-34. Haslam, M., 2006a. Archaeology of the Instant? Action and Narrative in Microscopic Archaeological Residue Analysis. Journal of Social Archaeology, 6(3), 402-24.

Haslam, M., 2006b. Potential Misidentification of in situ Archaeological Tool- residues: Starch and Conidia. Journal of Archaeological Science, 33(1), 114-21.

Haslam, M., 2008. Initial Tests on the Three-dimensional Movement of Starch in Sediments, in New Directions in Archaeological Science, eds. A. Fairbairn, S. O'Connor & B. Marwick, Canberra: Australian National University Press, 93-103.

Haslam, M., 2009. Mountains and Molehills: Sample Size in Archaeological Microscopic Stone-tool Residue Analysis, in Archaeological Science Under 201

a Microscope: Studies in Residue and Ancient DNA Analysis in Honour of Thomas H. Loy, eds. M. Haslam, G. Robertson, A. Crowther, S. Nugent & L. Kirkwood, Canberra: Australian National University Press, 47-79.

Hather, J. G., 1991. The Identification of Charred Archaeological Remains of Vegetative Parenchymous Tissue. Journal of Archaeological Science, 18(6), 661-75.

Hather, J. G. & S. L. R. Mason, 2002. Introduction: Some Issues in the Archaeobotany of Hunter-gatherers, in Hunter-Gatherer Archaeobotany: Perspectives from the Northern Temperate Zone, eds. S. L. R. Mason & J. G. Hather, London: University College London, 1-14.

Hausman, L. A., 1920. Structural Characteristics of the Hair of Mammals. The American Naturalist, 54(635), 496-523.

Hawksworth, D. L. & D. J. Hill, 1984. Reproduction, in The Lichen-Forming Fungi Glasgow Blackie, 29-44.

Hayden, B., 1992a. Introduction: Ecology and Culture, in A Complex Culture of the British Columbia Plateau: Traditional Stl'atl'imx Resource Use, ed. B. Hayden, Vancouver: UBC Press, 3-46.

Hayden, B., 1992b. Conclusions: Ecology and Complex Hunter/Gatherers, in A Complex Culture of the British Columbia Plateau: Traditional Stl'atl'imx Resource Use, ed. B. Hayden, Vancouver: UBC Press, 525-64.

Hayden, B., 1996. Feasting in Prehistoric and Traditional Societies, in Food and the Status Quest: An Interdisciplinary Perspective, eds. P. Wiessner & W. Schiefenhovel, Providence: Berghahn Books, 127-48. 202

Hayden, B., 1997. Observations on the Prehistoric Social and Economic Structure of the North American Plateau. World Archaeology, 29(2), 242-61.

Hayden, B., 1998. Practical and Prestige Technologies: The Evolution of Material Systems. Journal of Archaeological Method and Theory, 5(1), 1-55.

Hayden, B., 2000a. The Opening of Keatley Creek: Research Problems and Background, in The Ancient Past of Keatley Creek, Volume 1: Taphonomy, ed. B. Hayden, Burnaby: SFU Archaeology Press, 1-34.

Hayden, B., 2000b. Dating Deposits at Keatley Creek, in The Ancient Past of Keatley Creek, Volume I: Taphonomy, ed. B. Hayden, Burnaby: SFU Archaeology Press, 35-40.

Hayden, B., 2000c. Prestige Artifacts at Keatley Creek, in The Ancient Past of Keatley Creek, Volume II: Socioeconomy, ed. B. Hayden, Burnaby: SFU Archaeology Press, 35-40.

Hayden, B., 2005a. The Pithouses of Keatley Creek, Burnaby, B.C.: Archaeology Press, Simon Fraser University.

Hayden, B., 2005b. The Emergence of Large Villages and Large Residential Corporate Group Structures among Complex Hunter-Gatherers at Keatley Creek. American Antiquity, 70(1), 169-74.

Hayden, B., 2009. The Proof Is in the Pudding: Feasting and the Origins of Domestication. Current Anthropology, 50(5), 597-601.

Hayden, B., 2011. Feasting. Annual Review of Anthropology, 40(1), forthcoming 203

October 2011.

Hayden, B. & S. M. Cousins, 2004. The Social Dimensions of Roasting Pits at a Winter Village Site, in Complex Hunter-Gatherers, eds. W. C. Prentiss & I. Kuijt, Salt Lake City: University of Utah Press, 140-54.

Hayden, B. & R. W. Mathewes, 2009. The Rise and Fall of Complex Large Villages on the British Columbian Plateau: A Geoarchaeological Controversy. Canadian Journal of Archaeology, 33, 281-96.

Hayden, B. & R. J. Schulting, 1997. The Plateau Interaction Sphere and Late Prehistoric Cultural Complexity. American Antiquity, 62(1), 51-85.

Hebda, R. J., 1995. British Columbia Vegetation and Climate History with Focus on 6 ka BP. Geographie Physique et Quaternaire, 49(1), 55-79.

Henry, A. G., A. S. Brooks & D. R. Piperno, 2010. Microfossils in Calculus Demonstrate Consumption of Plants and Cooked Foods in Neanderthal Diets (Shanidar III, Iraq; Spy I and II, Belgium). Proceedings of the National Academy of Sciences, 108(2), 486-91.

Henry, A. G., H. F. Hudson & D. R. Piperno, 2009. Changes in Starch Grain Morphologies from Cooking. Journal of Archaeological Science, 36, 915-22. Henry, A. G. & D. R. Piperno, 2008. Using Plant Microfossils from Dental Calculus to Recover Human Diet: A Case Study from Tell al-Raqa'i, Syria. Journal of Archaeological Science, 35(7), 1943-50.

Henry, D. O., H. J. Hietala, A. M. Rosen, Y. E. Demidenko, V. I. Usik & T. L. Armagan, 2004. Human Behavioral Organization in the Middle Paleolithic: Were Neanderthals Different? American Anthropologist, 106(1), 17-31. 204

Hill, H. E. & J. Evans, 1987. The identification of Plants used in Prehistory from Organic Residues, in Archaeometry: Further Australasian Studies, eds. W. R. Ambrose & J. M. J. Mummery, Canberra: Department of Prehistory, Research School of Pacific Studies, Australian National University, 90-6.

Hills, L.V., 1971. A Talus Burial Site (EqQw 1), Skwaam Bay, Adams Lake, British Columbia, in: Stryd, A.H., Smith, R.A. (Eds.), Aboriginal Man and Environments on the Plateau of Northwest America, Students' Press, University of Calgary, Calgary, 27-35.

Hill-Tout, C., 1978a. The Salish People Volume I: The Thompson and the Okanagan, Vancouver: Talonbooks.

Hill-Tout, C., 1978b. The Salish People Volume II: The Squamish and the Lillooet Vancouver: Talonbooks.

Hoadley, R. B., 1990. Identifying Wood: Accurate Results with Simple Tools, Newtown, CT: Taunton Press.

Hoover, R., 2000. Acid-Treated Starches Food Reviews International, 16(3), 369-92.

Hoover, R., 2001. Composition, Molecular Structure, and Physicochemical Properties of Tuber and Root Starches: A Review. Carbohydrate Polymers, 45(3), 253-67.

Horrocks, M., 2005. A Combined Procedure for Recovering Phytoliths and Starch Residues from Soils, Sedimentary Deposits and Similar Materials. Journal of Archaeological Science, 32(8), 1169-75.

Horrocks, M. & S. Bedford, 2010. Introduced Dioscorea spp. Starch in Lapita and 205

Later Deposits, Vao Island, Vanuatu. New Zealand Journal of Botany, 48 (3-4), 179-83.

Horrocks, M., G. Irwin, M. Jones & D. Sutton, 2004a. Starch Grains and Xylem Cells of Sweet Potato (Ipomoea batatas) and Bracken (Pteridium esculentum) in Archaeological Deposits from Northern New Zealand. Journal of Archaeological Science, 31 (3), 251 -8.

Horrocks, M., P. A. Shane, I. G. Barber, D. M. D'Costa & S. L. Nichol, 2004b. Microbotanical Remains Reveal Polynesian Agriculture and Mixed Cropping in Early New Zealand. Review of Palaeobotany and Palynology, 131(3-4), 147-57.

Hu, Y.-Q., Z.-L. Zhang, S. Bera, D. K. Ferguson, C.-S. Li, W.-B. Shao & Y.-F. Wang, 2007. What Can Pollen Grains from the Terracotta Army Tell Us? Journal of Archaeological Science, 34(7), 1153-7.

Huber, K. C. & J. N. BeMiller, 2010. Modified Starch: Chemistry and Properties, in Starches: Characterization, Properties, and Applications, ed. A. C. Bertolini, Boca Raton, Florida: CRC Press, 145-204.

Hunn, E. S., 1981. On the Relative Contribution of Men and Women to Subistence Among Hunter-gatherers of the Columbia Plateau: A Comparison with Ethnographic Atlas Summaries. Journal of Ethnobiology, 1(1), 124-34.

Hunn, E. S., N. J. Turner & D. H. French, 1998. Ethnobiology and Subsistence, in Handbook of North American Indians, ed. D. E. Walker, Washington: Smithsonian Institution, 525-45.

Huntington, H. P., 2000. Using Traditional Ecological Knowledge in Science: 206

Methods and Applications. Ecological Applications, 10(5), 1270-4.

Ignace, M. B., 1998. Shuswap, in Handbook of North American Indians, ed. D. E. Walker, Washington: Smithsonian Institution, 203-11.

Jahren, A. H., N. Toth, K. Schick, J. D. Clark & R. G. Amundson, 1997. Determining Stone Tool Use: Chemical and Morphological Analyses of Residues on Experimentally Manufactured Stone Tools. Journal of Archaeological Science, 24(3), 245-50.

Jane, J.-l., T. Kasemsuwan, S. Leas, H. Zobel & J. F. Robyt, 1994. Anthology of Starch Granule Morphology by Scanning Electron Microscopy. Starch - Starke, 46, 121-9.

Jenkinson, D. S., 1981. The Fate of Plant and Animal Residues in Soil, in The Chemistry of Soil Process, eds. D. J. Greenland & M. H. B. Hayes, Chichester, UK: John Wiley & Sons, 505-61.

Johnson, L. E., 1995. Enhanced Early Detection and Enumeration of Zebra Mussel (Dreissena spp.) Veligers Using Cross-polarized Light Microscopy. Hydrobiologia, 312(2), 139-46.

Kaur, L., N. Singh & N. S. Sodhi, 2002. Some Properties of Potatoes and Their Starches II. Morphological, Thermal and Rheological Properties of Starches. Food Chemistry, 79(2), 183-92.

Kealhofer, L., R. Torrence & R. Fullagar, 1999. Integrating Phytoliths within Use- Wear/Residue Studies of Stone Tools. Journal of Archaeological Science, 26 (5), 527-46.

Keddie, G., 1988. Evolution of the Hand-maul. The Midden, XX(1), 6-10. 207

Ketcheson, M. V., 1979. Floral Analysis of Archaeological Sites in the Hat Creek Valley, Vancouver: University of British Columbia.

Kim, H.-S. & K. C. Huber, 2008. Channels within Soft Wheat Starch A- and B-Type Granules. Journal of Cereal Science, 48(1), 159-72.

Kiseleva, V. I., A. V. Krivandin, J. Fornal, W. Blaszczak, T. Jelinski & V. P. Yuryev, 2005. Annealing of Normal and Mutant Wheat Starches. LM, SEM, DSC, and SAXS Studies. Carbohydrate Research, 340(1), 75-83.

Kooyman, B., 2000. Understanding Stone Tools and Archaeological Sites, Calgary, Alberta; Albuquerque, New Mexico: University of Calgary Press; University of New Mexico Press.

Kooyman, B. & P. Kooyman, 2010. A Preliminary Examination of the Archaeological Potential of Starch Grain Analysis in Alberta and British Columbia, Calgary: Department of Archaeology, University of Calgary.

Kooyman, B., M. E. Newman & H. Ceri, 1992. Verifying the Reliability of Blood Residue Analysis on Archaeological Tools. Journal of Archaeological Science, 19(3), 265-9.

Krok, F., J. Szymonska, P. Tomasik & M. Szymonski, 2000. Non-contact AFM Investigation of Influence of Freezing Process on the Surface Structure of Potato Starch Granule. Applied Surface Science, 157(4), 382-6. Kuhnlein, H. V. & N. J. Turner, 1991. Traditional Plant Foods of Canadian Indigenous Peoples: Nutrition, Botany and Use, Philadelphia: Gordon and Breach Science Publishers.

Kuijt, I., 1989. Subsistence Resource Variability and Culture Change During the 208

Middle-Late Prehistoric Cultural Transition on the Canadian Plateau. Canadian Journal of Archaeology, 13, 97-113.

Kuijt, I., 2001. Reconsidering the Cause of Cultural Collapse in the Lillooet Area of British Columbia, Canada: A Geoarchaeological Perspective. American Antiquity, 66(4), 692-703.

Kuijt, I. & W. C. Prentiss, 2004. Villages on the Edge: Pithouses, Cultural Change, and the Abandonment of Aggregate Villages, in Complex Hunter-Gatherers, eds. W. C. Prentiss & I. Kuijt, Salt Lake City: University of Utah Press, 155-70.

Lamb, J. & T. Loy, 2005. Seeing Red: The Use of Congo Red Dye to Identify Cooked and Damaged Starch Grains in Archaeological Residues. Journal of Archaeological Science, 32(10), 1433-40.

Landeweert, R., E. Hoffland, R. D. Finlay, T. W. Kuyper & N. van Breemen, 2001. Linking Plants to Rocks: Ectomycorrhizal Fungi Mobilize Nutrients from Minerals. Trends in Ecology & Evolution, 16(5), 248-54.

Lantz, T. C. & N. J. Turner, 2003. Traditional Phenological Knowledge of Aboriginal Peoples in British Columbia. Journal of Ethnobiology, 23(2), 263-86.

Leeuwenhoek, A., 1719. Epistolae Physiologicae super Compluribus Naturae Arcanus, apud Adrianum Beman, Delft, Netherlands.

Lepofsky, D., 2000a. Site Formations Processes at Keatley Creek: The Paleoethnobotanical Evidence, in The Ancient Past of Keatley Creek, Volume 1: Taphonomy, ed. B. Hayden, Burnaby: SFU Archaeology Press, 105-34. 209

Lepofsky, D., 2000b. Socioeconomy at Keatley Creek: The Botanical Evidence, in The Ancient Past of Keatley Creek, Volume 2: Socioeconomy, ed. B. Hayden, Burnaby: SFU Archaeology Press, 75-86.

Lepofsky, D., 2004. Paleoethnobotany in the Northwest, in People and Plants in Ancient Western North America, ed. P. E. Minnis, Washington: Smithsonian Books, 367-464.

Lepofsky, D. & S. L. Peacock, 2004. A Question of Intensity: Exploring the Role of Plant Foods in Northern Plateau Prehistory, in Compex Hunter-Gatherers, eds. W. C. Prentiss & I. Kuijt, Salt Lake City: University of Utah Press, 115-39.

Li, C.-Y., W.-H. Li, B. Lee, A. Laroche, L.-P. Cao & Z.-X. Lu, 2011. Morphological Characterization of Triticale Starch Granules During Endosperm Development and Seed Germination. Canadian Journal of Plant Science, 91 (1), 57-67.

Li, J. H., T. Vasanthan, B. Rossnagel & R. Hoover, 2001. Starch from Hull-less Barley: I. Granule Morphology, Composition and Amylopectin Structure. Food Chemistry, 74(4), 395-405.

Lints, A. & C. Surette, 2010. Archaeological Residues: A Practical Guide to Recovery and Handling Artifacts. Poster presented at the Canadian Archaeological Association 43rd Annual Meeting Calgary, Alberta.

Livarda, A. & G. Kotzamani, 2006. Nemea Valley Archaeological Project - Barnavos Excavation, Athens, Greece: American School of Classical Studies.

Lombard, M., 2008. Finding Resolution for the Howiesons Poort through the Microscope: Micro-residue Analysis of Segments from Sibudu Cave, South 210

Africa. Journal of Archaeological Science, 35(1), 26-41.

Lombard, M. & L. Wadley, 2007a. The Morphological Identification of Micro- residues on Stone Tools Using Light Microscopy: Progress and Difficulties Based on Blind Tests. Journal of Archaeological Science, 34(1), 155-65.

Lombard, M. & L. Wadley, 2007b. Micro-residues on Stone Tools: The Bigger Picture fro a South African Middle Stone Age Perspective, in Theory and Practice of Archaeological Residue Analysis, BAR International Series 1650, eds. H. Barnard & J. W. Eerkens, Oxford, England: Archaeopress 18-28.

Loy, T. H., 1983. Prehistoric Blood Residues: Detection on Tool Surfaces and Identification of Species of Origin. Science, 220(4603), 1269-71.

Loy, T. H., 1993. The Artifact as Site: An Example of the Biomolecular Analysis of Organic Residues on Prehistoric Tools. World Archaeology, 25(1), 44-63.

Loy, T. H., 1994. Methods in the Analysis of Starch Residues on Prehistoric Stone Tools, in Tropical Archaeobotany: Applications and New Developments, ed. J. G. Hather, London: Routledge, 86-114.

Loy, T. H., 2006. Optical Properties of Potential Look-Alikes, in Ancient Starch Research, eds. R. Torrence & H. Barton, Walnut Creek, California: Left Coast Press, 123-4.

Loy, T. H. & D. E. Nelson, 1986. Potential Applications of the Organic Residues on Ancient Tools, in Proceedings of the 24th International Archaeometry Symposium, eds. J. Olin & M. Blakcman, Washington: Smithsonian Institution Press, 179-86. 211

Loy, T. H., M. Spriggs & S. Wickler, 1992. Direct Evidence for Human Use of Plants 28,000 Years Ago: Starch Residues on Stone Artefacts from the Northern Solomon Islands. Antiquity, 66(253), 898-912.

Lyons, N., 2000. Analysis of Archaeobotanical Tissues from Roasting Pit Features at Keatley Creek. Manuscript in possession of the author and at the Paleoethnobotanical Laboratory, Department of Archaeology, Burnaby: Simon Fraser University.

Marousis, S. N. & G. D. Saravacos, 1990. Density and Porosity in Drying Starch Materials. Journal of Food Science, 55(5), 1367-72.

Martin, C. & A. M. Smith, 1995. Starch Biosynthesis. The Plant Cell, 7, 971-85.

Martins, F. R., 1976. New Archaeological Techniques for the Study of Ancient Root Crops in Peru, Birmingham: University of Birmingham, 159.

Mathewes, R. W., 1978. The Environment and Biotic Resources of the Lillooet Area, in Reports of the Lillooet Archaeological Project Ottawa: National Museum of Man, 68-99.

Matthew, M., 1986a. Foods of the Shuswap, Kamloops: Secwepemc Cultural Education Society.

Matthew, M., 1986b. Technology of the Shuswap, Kamloops: Secwepemc Cultural Education Society. Mauseth, J. D., 2009. Botany: An Introduction to Plant Biology, Sudbury, Massachusetts: Jones and Bartlett Publishers.

McGill, M., S. L. Peacock & B. Kooyman, 2010. Fish, Fowl, and Four-Legged Friends: Faunal Finds from a Roasting Pit at White Rock Springs. Poster 212

presented at the Canadian Archaeological Association 43rd Annual Meeting, Calgary, Alberta.

McKenzie, S., C. Weins & S. L. Peacock, 2010. The Ins and Outs of Richness and Reuse: A Paleoethnobotanical Case Study. Poster presented at the Canadian Archaeological Association 43rd Annual Meeting, Calgary, Alberta.

Mercader, J., 2009. Mozambican Grass Seed Consumption During the Middle Stone Age. Science, 326(5960), 1680-3.

Messner, T. C., 2008. Woodland Period People and Plant Interactions, New Insights from Starch Grain Analysis, in Department of Anthropology Philadelphia, PA: Temple University, 420.

Messner, T. C., R. Dickau & J. Harbison, 2008. Starch Grain Analysis Methodology and Applications in the Northeast, in Current Northeast Paleoethnobotany II, ed. J. P. Hart, New York: New York State Museum, 111-28.

Messner, T. C. & B. Schindler, 2010. Plant Processing Strategies and their Affect Upon Starch Grain Survival when Rendering Peltandra virginica (L.) Kunth, Araceae Edible. Journal of Archaeological Science, 37(2), 328-36.

Meyer, A., 1895. Untersuchungen Uber die Starkekdrner. Wesen und Lebensgeschichte der Starkekdrner der Hdheren Pflanzen, Jena: Verlag von Gustav Fischer.

Moerman, D. E., 1998. Native American Ethnobotany, Portland, Oregon: Timber Press.

Moore, P. D., J. A. Webb & M. E. Collinson, 1991. Pollen Analysis, Oxford: Blackwell Scientific Publications. 213

Morin, J., R. Dickie, T. Sakaguchi & J. Hoskins, 2008. Late Prehistoric Settlement Patterns and Population Dynamics Along the Mid-Fraser. BC Studies, 160, 9-34.

Mullin, W. J., S. Peacock, D. C. Loewen & N. J. Turner, 1997. Macronutrients Content of Yellow Glacier Lily and Balsamroot; Root Vegetables Used by Indigenous Peoples of Northwestern North America. Food Research International, 30(10), 769-75.

Nageli, K.W., 1858. Die Starkekorner: Morphologische, physiologische, chemisch-physicalische und systematisch-botanische Monographie, Friedrich Schulthess, Zurich.

Newman, M., Julig, P., 1989. The Identification of Protein Residues on Lithic Artifacts from a Stratified Boreal Forest Site. Canadian Journal of Archaeology, 13, 119-130.

Newman, M. E., R. M. Yohe li, B. Kooyman & H. Ceri, 1997. "Blood" From Stones? Probably: A Response to Fiedel. Journal of Archaeological Science, 24(11), 1023-7.

Newman, M. E., R. M. Yohe, H. Ceri & M. Q. Sutton, 1993. Immunological Protein Residue Analysis of Non-lithic Archaeological Materials. Journal of Archaeological Science, 20(1), 93-100.

Nicholas, G. P. & L. Westfall, in press. Long-Term Secwepemc Plant Use: Initial 214

Results of Archaeological and archaeobotanical Investigations in the Interior Plateau, British Columbia., in Secwepemc People & Plants: Research Papers in Shuswap Ethnobotany, eds. M. B. Ignace, N. J. Turner & S. L. Peacock.

Nicolaides, M., 2010. The Proof is in the Pits: A Paleoethnobotanical Analysis of Earth Ovens from the White Rock Springs Site (EeRj 226), an Ancient Root Processing Locale on the Canadian Plateau, in Department of Archaeology Calgary: University of Calgary, 145.

Niemann, C. & R. L. Whistler, 1992. Effect of Acid Hydrolysis and Ball Milling on Porous Corn Starch. Starch - Starke, 44(11), 409-14.

Odell, G. H., 2001. Stone Tool Research at the End of the Millennium: Classification, Function, and Behavior. Journal of Archaeological Research, 9, 45-100.

Odell, G. H., 2004. Tool Function, in Lithic Analysis, ed. G. H. Odell, New York: Kluwer Academic/Plenum Publishers, 135-73. Ommer, R. E. & N. J. Turner, 2004. Informal Rural Economies in History. Labour- Le Travail, 53,127-57.

Owens, J. N. & S. Simpson, 1986. Pollen from Conifers Native to British Columbia. Canadian Journal of Forest Research, 16(5), 955-67.

Palmer, G., 1975. Shuswap Indian Ethnobotany. Syesis, 8, 29-81.

Panshin, A. J. & C. de Zeeuw, 1980. Textbook of Wood Technology: Structure, Identification, Properties, and Uses of the Commercial Woods of the United States and Canada, New York: McGraw-Hill. Parker, M. L., 1985. The Relationship Between A-type and B-type Starch Granules 215

in the Developing Endosperm of Wheat. Journal of Cereal Science, 3(4), 271-8.

Parr, J. F. & M. Carter, 2003. Phytolith and Starch Analysis of Sediment Samples from Two Archaeological Sites on Dauar Island, Torres Strait, Northeastern Australia. Vegetation History and Archaeobotany, 12(2), 131-41.

Peacock, S. L., 1998. Putting Down Roots: The Emergence of Wild Plant Food Production on the Canadian Plateau, in School of Environmental Studies Victoria: University of Victoria, 400.

Peacock, S. L., 2002. Perusing the Pits: The Evidence for Prehistoric Root Resource Processing on the Canadian Plateau, in Hunter-Gatherer Archaeobotany: Perspectives from the Northern Temperate Zone, eds. S. L. R. Mason & J. G. Hather, London: University College of London.

Peacock, S. L., 2007a. Summary of Proposed Research, in SSHRC Application Kamloops: University of British Columbia Okanagan, 10-27.

Peacock, S. L., 2007b. The Archaeology of Ancient Plant Use at the White Rock Springs and Houth Meadows locales in the Hat Creek Valley, British Columbia, in Application for Heritage Permit Kamloops: University of British Columbia Okanagan.

Peacock, S. L., 2008. From Complex to Simple: Balsamroot, Inulin, and the Chemistry of Traditional Interior Salish Pit-cooking Technology. Botany, 86, 116-28.

Peacock, S. L. & K. Chambers, 2003. A Cultural Heritage Overview of the Hat Creek Watershed. Report prepared for the Bonaparte Indian Band, Cache Creek, B.C.. 216

Peacock, S. L. & N. J. Turner, 2000. "Just Like a Garden": Traditional Resource Management and Biodiversity Conservation on the Interior Plateau of British Columbia, in Biodiversity and Native America, eds. P. Minnis & W. Elisens, Norman: University of Oklahoma Press.

Pearsall, D. M. & E. H. Dinan, 1992. Developing a Phytolith Classification System, in Phytolith Systematics: Emerging Issues, eds. G. J. Rapp & S. C. Mulholland, New York: Plenum Press, 37-64.

Pearsall, D. M., 2000. Phytolith Analysis, in Paleoethnobotany: A Handbook of Proceedures. San Diego: Academic Press, 355-496.

Pearsall, D. M., K. Chandler-Ezell & J. A. Zeidler, 2004. Maize in Ancient Ecuador: Results of Residue Analysis of Stone Tools from the Real Alto Site. Journal of Archaeological Science, 31(4), 423-42.

Peng, M., M. Gao, E. S. M. Abdel-Aal, P. Hucl & R. N. Chibbar, 1999. Separation and Characterization of A- and B-Type Starch Granules in Wheat Endosperm. Cereal Chemistry, 76(3), 375-9.

Perry, L., 2002. Starch Granule Size and the Domestication of Manioc (Manihot esculenta) and Sweet Potato (Lpomoea batatas). Economic Botany, 56(4), 335-49.

Perry, L., 2004. Starch Analyses Reveal the Relationship between Tool Type and Function: An Example from the Orinoco Valley of Venezuela. Journal of Archaeological Science, 31(8), 1069-81.

Phillips, E. W. J., 1941. The Identification of Coniferous Woods by their Microscopic 217

Structure. Journal of the Linnean Society of London, Botany, 52(343), 259-314.

Piperno, D. R., 2006a. Identifying Manioc {Manihot esculenta Crantz) and Other Crops in Pre-Columbian Tropical America through Starch Grain Analysis: A Case Study from Central Panama, in Documenting Domestication: New Genetic and Archaeological Paradigms, eds. M. A. Zeder, D. G. Bradley, E. Emshwiller & B. D. Smith, Berkeley: University of California Press, 46-67.

Piperno, D. R., 2006b. Phytoliths:A Comprehensive Guide for Archaeologists and Paleoecologists, Lanham, MD: AltaMira Press.

Piperno, D. R. & T. D. Dillehay, 2008. Starch Grains on Human Teeth Reveal Early Broad Crop Diet in Northern Peru. Proceedings of the National Academy of Sciences, 105(50), 19622-7.

Piperno, D. R. & I. Hoist, 1998. The Presence of Starch Grains on Prehistoric Stone Tools from the Humid Neotropics: Indications of Early Tuber Use and Agriculture in Panama. Journal of Archaeological Science, 25(8), 765-76.

Piperno, D. R., A. J. Ranere, I. Hoist & P. Hansell, 2000. Starch Grains Reveal Early Root Crop Horticulture in the Panamanian Tropical Forest. Nature, 407 (6806), 894-7.

Piperno, D. R., E. Weiss, I. Hoist & D. Nadel, 2004. Processing of Wild Cereal Grains in the Upper Paleolithic Revealed by Starch Grain Analysis. Nature, 430(7000), 670-3.

Pojar, J. & D. Meidinger, 1991. British Columbia: The Environmental Setting, in 218

Ecosystems of British Columbia, eds. D. Meidinger & J. Pojar, Victoria: BC Ministry of Forests, 39-68.

Pokotylo, D. L., 1978. Lithic Technology and Settlement Patterns in Upper Hat Creek Valley, B.C., in Department of Anthropology, Vancouver: University of British Columbia, 358.

Pokotylo, D. L., 1981. Towards an Understanding of Prehistoric Upland Settlement Behaviour in the British Columbia Southern Interior Plateau, in Networks of the Past: Regional Interaction in Archaeology, eds. P. D. Francis, F. J. Kense & P. G. Duke, Calgary: Archaeological Association of the University of Calgary 379-96.

Pokotylo, D. L., M. E. Binkley & A. J. Curtin, 1987. The Cache Creek Burial Site (EeRh 1), British Columbia. Contributions to Human History, 1,1-14, British Columbia Provincial Museum, Victoria.

Pokotylo, D. L. & P. D. Froese, 1983. Archaeological Evidence for Prehistoric Root Gathering on the Southern Interior Plateau of British Columbia: A Case Study from Upper Hat Creek Valley. Canadian Journal of Archaeology, 7(2), 127-57.

Prentiss, A. M., 2009. The Emergence of New Socioeconomic Strategies in the Middle and Late Holocene Pacific Northwest Region of North America, in Macroevolution in Human Prehistory: Evolutionary Theory and Processual Archaeology, eds. A. M. Prentiss, I. Kuijt & J. C. Chatters, New York: Springer, 111-32.

Prentiss, A. M., 2011. Social Histories of Complex Hunter-Gatherers: Pacific 219

Northwest Prehistory in a Macroevolutionary Framework, in Hunter-Gatherer Archaeology as Historical Process, eds. K. E. Sassman & D. H. J. Holly, Tucson: University of Arizona Press, 17-33.

Prentiss, A. M., J. C. Chatters, N. Lyons & L. E. Harris, 2011. Archaeology in the Middle Fraser Canyon, British Columbia: Changing Perspectives on Paleoecology and Emergent Cultural Complexity. Canadian Journal of Archaeology, 35(1), 143-74.

Prentiss, A. M., G. Cross, T. A. Foor, M. Hogan, D. Markle & D. S. Clarke, 2008. Evolution of a Late Prehistoric Winter Village on the Interior Plateau of British Columbia: Geophysical Investigations, Radiocarbon Dating, and Spatial Analysis of the Bridge River Site. American Antiquity, 73(1), 59-81.

Prentiss, A. M., N. Lyons, L. E. Harris, M. R. P. Burns & T. M. Godin, 2007. The Emergence of Status Inequality in Intermediate Scale Societies: A Demographic and Socio-economic History of the Keatley Creek Site, British Columbia. Journal of Anthropological Archaeology, 26(2), 299-327.

Prentiss, W. C., J. C. Chatters, M. Lenert, D. S. Clarke & R. C. O'Boyle, 2005a. The Archaeology of the Plateau of Northwestern North America During the Late Prehistoric Period (3500-200 B.P.): Evolution of Hunting and Gathering Societies. Journal of World Prehistory, 19(1), 47-118.

Prentiss, W. C., D. S. Clarke, D. Markle, J. Bochart, J. Foss & S. Mandelko, 2005b. Report of the 2004 University of Montana Investigations at the Bridge River Site (EeRI4), Report on file, Bridge River Indian Band and Stl'atl'imx First Nation Offices, Lillooet.

Prentiss, W. C. & I. Kuijt, 2004. The Evolution of Collector Systems on the 220

Canadian Plateau, in Complex Hunter-Gatherers: Evolution and Organization of Prehistoric Communities on the Plateau of Northwestern North America, eds. W. C. Prentiss & I. Kuijt, Salt Lake City: University of Utah Press, 49-66.

Prentiss, W. C., M. Lenert, T. A. Foor, N. B. Goodale & T. Schlegel, 2003. Calibrated Radiocarbon Dating at Keatley Creek: The Chronology of Occupation at a Complex Hunter-Gatherer Village. American Antiquity, 68 (4), 719-35.

Prychid, C. J. & P. J. Rudall, 1999. Calcium Oxalate Crystals in Monocotyledons: A Review of their Structure and Systematics. Annals of Botany, 84, 725-39.

Quigg, J. M., M. E. Malainey, R. Przybylski & G. Monks, 2001. No Bones About it: Using Lipid Analysis of Burned Rock and Groundstone Residues to Examine Late Archaic Subsistence Practices in South Texas. Plains Anthropologist, 46(177), 283-303.

Raven, P. H., R. F. Evert & S. E. Eichhorn, 2005. Biology of Plants, New York, New York: W.H. Freeman and Company.

Ray, V. F„ 1939. Cultural Relations in the Plateau of Northwestern America, Los Angeles: Southwest Museum.

Ray, V. F., 1942. Culture Element Distributions XXII: Plateau. Anthropological Records, 8(2), 99-262.

Reichert, E. T., 1913. The Differentiation and Specificity of Starches in Relation to Genera, Species, etc., Washington, D.C.: Carnegie Institute of Washington.

Revedin, A., B. Aranguren, R. Becattini, L. Longo, E. Marconi, M. M. Lippi, N. 221

Skakun, A. Sinitsyn, E. Spiridonova & J. Svoboda, 2010. Thirty Thousand- year-old Evidence of Plant Food Processing. Proceedings of the National Academy of Sciences, 107(44), 18815-9.

Richards, T. H. & M. K. Rousseau, 1987. Late Prehistoric Cultural Horizons on the Canadian Plateau, Burnaby, B.C.: Department of Archaeology, Simon Fraser University.

Rost, T. L., M. G. Barbour, C. R. Stocking & T. M. Murphy, 2006. Plant Biology, Toronto, Ontario: Thompson Brooks/Cole.

Rots, V. & P. M. Vermeersch, 2004. Experimental Characterization of Hafting Traces and Their Recognition in Archaeological Assemblages, in Lithics in Action: Papers from the Conference Lithic Studies in the Year 2000, eds. E. A. Walker, F. Wenban-Smith & F. Healy, Oxford: Oxbow Books, 156-68.

Rousseau, M. K., 1993. Early Prehistoric Occupation of South-Central British Columbia: A Review of the Evidence and Recommendations for Future Research. BC Studies, 99(Autumn), 140-83.

Rousseau, M. K., 2004. A Culture Historic Synthesis and Changes in Human Mobility, Sedentism, Subsistence, Settlement, and Population on the Canadian Plateau, 7,000-200 B.P., in Complex Hunter-Gatherers: Evolution and Organization of Prehistoric Communities on the Plateau of Northwestern North America, eds. W. C. Prentiss & I. Kuijt, Salt Lake City: University of Utah Press, 3-22.

Rousseau, M. K. & D. G. Howe, 1987. Scheidam Flats: A Prehistoric Root Roasting Area near Kamloops. The Midden, 19(2), 7-10.

Sajilata, M. G., R. S. Singhal & P. R. Kulkarni, 2006. Resistant Starch-A Review. 222

Comprehensive Reviews in Food Science and Food Safety, 5(1), 1 -17.

Samuel, D., 1996. Investigation of Ancient Egyptian Baking and Brewing Methods by Correlative Microscopy. Science, 273(5274), 488-90.

Samuel, D., 2006. Modified Starch, in Ancient Starch Research, eds. R. Torrence & H. Barton, Walnut Creek, California: Left Coast Press, 205-16.

Sandstedt, R. M., 1955. Photomicrographic Studies of Wheat Starch. III. Enzymatic Digestion and Granule Structure. Cereal Chemistry, 32,14-47.

Sanger, D., 1968a. The Chase Burial Site EeQw: 1, British Columbia. National Museums of Canada, Bulletin 224 Contributions to Anthropology, 6, 86-186.

Sanger, D., 1968b. The Texas Creek Burail Site Assemblage, British Columbia, in National Museum of Canada. Anthropology papers, no. 17. Ottawa: Queen's Printer, 1-23.

Sanger, D., 1970. The Archaeology of the Lochnore-Nesikep Locality, British Columbia. Syesis, 3(1), 1-129.

Sangster, A. G., M. J. Hodson & H. J. Tubb, 2001. Silicon Deposition in Higher Plants, in Silicon in Agriculture, eds. L. E. Datnoff, G. H. Snyder & G. H. Korndorfer, Amsterdam: Elsevier, 85-113.

Schleiden, M. J., 1849. Principles of Scientific Botany: or, Botany as an Inductive Science, New York: Johnson Reprint corp., 1969.

Scott Cummings, L., 1994. Pollen and Starch Analysis of Features in a Pithouse at the Paquet Gulch Bridge Site (35WS125), in Archaeological Excavations at the Paquet Gulch Bridge Site: A Pithouse Village in the Deschutes River 223

Basin, Southwestern Columbia Plateau, Oregon, eds. T. Connolly, D. Jenkins & J. Benjamin, Eugene, Oregon: Department of Anthropology and Oregon State Museum of Anthropology, University of Oregon, 179-87.

Scott Cummings, L., 2007. Manual for Pollen, Phytolith, Starch, FTIR, AMS Radiocarbon, and Macrofloral Sampling, Golden, CO: Paleo Research Institute.

Seib, P. A., 1994. Wheat Starch: Isolation, Structure and Properties. Oyo Toshitsu Kagaku - Journal of Applied Glycoscience, 41(1), 49-69.

Shafer, H. J. & R. G. Holloway, 1979. Organic Residue Analysis in Determining Stone Tool Function, in Lithic Use-Wear Analysis, ed. B. Hayden, New York: Academic Press, 385-299.

Shanks, O. C., R. Bonnichsen, A. T. Vella & W. Ream, 2001. Recovery of Protein and DNA Trapped in Stone Tool Microcracks. Journal of Archaeological Science, 28(9), 965-72.

Sharma, P. D., 2005. Ecological Groups of Fungi, in The Fungi and Allied Organisms. Oxford, U.K.: Alpha Science International Ltd., 283-301.

Sheppard, J. & B. Muir, 2010. "Oh, I've Seen One Way Bigger Than That!" Current Research on Housepit Settlement Patterns in the Mid-Fraser Region of B.C. The Midden, 42(3), 4-8.

Simao, R. A. & B. R. Cordenunsi, 2010. Characterization of Starch Granules: An Atomic Force Microscopy Approach, in Starches: Characterization, Properties, and Applications, ed. A. C. Bertolini, Boca Raton, Florida: CRC Press, 21-32. 224

Simpson, M. G., 2010. Plant Systematics, 2 ed., Elsevier Academic Press, Burlington, MA.

Singh, N., J. Singh, L. Kaur, N. S. Sodhi & B. S. Gill, 2003. Morphological, Thermal and Rheological Properties of Starches from Different Botanical Sources. Food Chemistry, 81,219-31.

Smith, A. M., 2010. Starch and Starch Granules, in Encyclopedia of Life Sciences, John Wiley & Sons, Chichester, UK, pp. 1 -4.

Smith, A. M., K. Denyer & C. Martin, 1997. The Synthesis of the Starch Granule. Annual Review of Plant Physiology and Plant Molecular Biology, 48(1), 67-87.

Smith, C. S., W. Martin & K. A. Johansen, 2001. Sego Lilies and Prehistoric Foragers: Return Rates, Pit Ovens, and Carbohydrates. Journal of Archaeological Science, 28(2), 169-83.

Smith, H. I., 1899. Archaeology of Lytton, British Columbia, in Publications of the Jesup North Pacific Expedition, ed. F. Boas, New York: AMS Press.

Smith, H. I., 1900. Archaeology of the Thompson River region, British Columbia, in Publications of the Jesup North Pacific Expedition, ed. F. Boas, New York: AMS Press, 401-54.

Smith, H. I., 1913. The Archaeological Collection from the Southern Interior of British Columbia, Ottawa: Government Printing Bureau.

Smith, T., 1998. Our Stories are Written on the Land: A Brief History of the Upper St'at'imc, Lillooet: Upper St'at'imc Language, Culture and Education Society. Sobolik, K. D., 1996. Lithic Organic Residue Analysis: An Example from the Southwestern Archaic. Journal of Field Archaeology, 23(4), 461-9.

Spooner, B. & P. Roberts, 2005. Mycorrhizas- Promoters of Growth, in Fungi, eds. S. A. Corbet, S. M. Walters, R. West, D. Streeter & D. A. Ratcliffe, London: Harper Collins Publishers, 75-93.

Stahl, A. B., 1989. Plant-food Processing: Implications for Dietary Quality, in Foraging and Farming: The Evolution of Plant Exploitation, eds. D. R. Harris & G. C. Hillman, London: Unwin Hyman, 171-94.

Steedman, E. V., 1930. Ethnobotany of the Thompson Indians of British Columbia. Based on field notes by James A. Teit, in Annual Report of the Bureau of American Ethnology, Washington: Smithsonian Institution, 441-522. Stephenson, S. L., 2010. The Role of Fungi in Nature, in The Kingdom Fungi: The Biology of Mushrooms, Molds, and Lichens, Portland, Oregon: Timber Press, 163-85.

Sterling, C., 1968. The Structure of the Starch Grain, in Starch and its Derivatives, ed. J. A. Radley, London: Chapman and Hall Ltd., 139-67.

Stewart, H., 1973. Artifacts of the Northwest Coast Indians, Saanichton, B.C.: Hancock House Publishers.

Stoddard, F. L., 1999. Survey of Starch Particle-Size Distribution in Wheat and Related Species. Cereal Chemistry, 76(1), 145-9.

Stryd, A. H., 1970. Preliminary Report of the 1970 Archaeological Excavations Near Lillooet, British Columbia, Report on file with the British Columbia Archaeology Branch, Victoria, B.C. 226

Stryd, A. H., 1972. Housepit Archaeology at Lillooet, British Columbia: The 1970 Field Season. BC Studies, 14, 17-46.

Stryd, A. H., 1973. The Later Prehistory of the Lillooet Area, British Columbia, in Department of Archaeology Calgary: University of Calgary.

Stryd, A. H., 1974. Lillooet Archaeological Project: 1974 Field Season. Cariboo College Papers in Archaeology, 1,1-47.

Stryd, A. H., 1978. Reports of the Lillooet Archaeological Project: Introduction and Setting, in National Museum of Man Mercury Series, eds. A. H. Stryd & S. Lawhead, Ottawa: Archaeological Survey of Canada.

Stryd, A. H., 1980. A Review of the Recent Activities Undertaken by the Lillooet Archaeological Project. The Midden, 12(2), 5-20. Stryd, A. H., 1985. Lillooet-Pioneer Road Upgrading Project Lillooet, B.C. Detailed Heritage Resource Inventory and Impact Assessment, Report on file with the British Columbia Archaeology Branch, Victoria, B.C.

Stryd, A. H. & L. V. Hills, 1972. An Archaeological Site Survey of the Lillooet—Big Bar Area, British Columbia. Syesis, 5,191-209.

Stryd, A. H. & M. K. Rousseau, 1996. The Early Prehistory of the Mid-Fraser- Thompson River Area, in Early Human Occupation in British Columbia, eds. R. L. Carlson & L. Dalla Bona, Vancouver: UBC Press, 177-204.

Sujka, M. & J. Jamroz, 2007. Starch Granule Porosity and its Changes by Means of Amylolysis. International Agrophysics, 21(1), 107.

Sullivan, A. P., R. A. Cook, M. P. Purtill & P. M. Uphus, 2001. Economic and Land- Use Implications of Prehistoric Fire-Cracked-Rock Piles, Northern Arizona. 227

Journal of Field Archaeology, 28(3-4), 367-82.

Sumbali, G., 2005. Fungi as Mutual Symbionts, in The Fungi, ed. B. M. Johri, Harrow, U.K.: Alpha Science International Ltd., 198-212.

Szczawinski, A. F., Turner, N.J., 1980. Wild Green Vegetables of Canada, National Museum of Natural Sciences, Ottawa.

Szymoriska, J., F. Krok, E. Komorowska-Czepirska & K. Rebilas, 2003. Modification of Granular Potato Starch by Multiple Deep-freezing and Thawing. Carbohydrate Polymers, 52(1), 1-10.

Tang, H., T. Mitsunaga & Y. Kawamura, 2006. Molecular Arrangement in Blocklets and Starch Granule Architecture. Carbohydrate Polymers, 63(4), 555-60.

Teit, J. A., 1900. The Thompson Indians of British Columbia, New York: AMS Press.

Teit, J. A., 1906. The Lillooet Indians, New York: AMS Press.

Teit, J. A., 1909. The Shuswap, New York: AMS Press.

Teit, J. A., 1912. Traditions of the Lillooet Indians of British Columbia. The Journal of American Folklore, 25(98), 287-371.

Teit, J. A., 1917. Folk-Tales of Salishan Tribes, in Folk-Tales ofSalishan and Sahaptin Tribes, Memoir, American Folk-Lore Society, ed. F. Boas, New York: Stechert & Co., 1-134.

Teit, J. A., 1930. The Okanogan, in Annual Report of the Bureau of American Ethnology, ed. F. Boas, Washington: Smithsonian Institution, 198-294. 228

Tester, R. F., 1997. Starch: The Polysaccharide Fractions, in Starch: Structure and Functionality, eds. P. J. Frazier, A. M. Donald & P. Richmond, Cambridge, UK: The Royal Society of Chemistry, 163-71.

Tester, R. F., J. Karkalas & X. Qi, 2004. Starch-composition, Fine Structure and Architecture. Journal of Cereal Science, 39(2), 151-65.

Therin, M., 1998. The Movement of Starch Grains in Sediments, in A Closer Look: Recent Studies of Stone Tools, ed. R. Fullagar, Sydney: Sydney University 61-72.

Thompson, R. G., 2006. Documenting the Presence of Maize in Central and South America through Phytolith Analysis of Food Residues, in Documenting Domestication: New Genetic and Archaeological Paradigms, eds. M. A. Zeder, D. G. Bradley, E. Emshwiller & B. D. Smith, Berkeley: University of California Press, 82-95.

Thorns, A. V., 1989. The Northern Roots of Hunter-gatherer Intensification: Camas and the Pacific Northwest, in Department of Anthropology Washington State University.

Thorns, A. V., 2008a. The Fire Stones Carry: Ethnographic Records and Archaeological Expectations for Hot-rock Cookery in Western North America. Journal of Anthropological Archaeology, 27(4), 443-60.

Thorns, A. V., 2008b. Ancient Savannah Roots of the Carbohydrate Revolution in South-Central North America. Plains Anthropologist, 53(205), 121-36.

Thorns, A. V., 2009. Rocks of Ages: Propagation of Hot-rock Cookery in Western North America. Journal of Archaeological Science, 36(3), 573-91. 229

Thorns, A. V., A. R. Laurence, M. Kamiya & V. M. J. Bryant, 2011. Experimentally Tracking Geophyte Microfossils from Raw and Baked-Residue Status. Poster presented at the 76th Annual Meeting Society for American Archaeology, Sacramento, California.

Titus, L. W., 1980. Hair Identification for the Archaeologist, A Yukon Sample, in Department of Archaeology, Burnaby: Simon Fraser University, 160.

Tobin, D. J., 2005. The Human Hair Fibre, in Hair in Toxicology: An Important Bio- Monitor, ed. D. J. Tobin, Cambridge, UK: Royal Society of Chemistry, 34-56.

Tomasik, P., 2002. Saccharides in Food-composition and Transformations (in Polish), in Food Chemistry, Warsaw, Poland: WNT.

Torrence, R., 2006. Starch in Sediments, in Ancient Starch Research, eds. R. Torrence & H. Barton, Walnut Creek, California: Left Coast Press, 145-76.

Torrence, R. & H. Barton (eds.), 2006. Ancient Starch Research, Walnut Creek, California: Left Coast Press.

Torrence, R., R. Wright & R. Conway, 2004. Identification of Starch Granules Using Image Analysis and Multivariate Techniques. Journal of Archaeological Science, 31(5), 519-32.

Traverse, A., 2007. Paleopalynology, Springer, Dordrecht, The Netherlands.

Turnbull, C., 1977. Archaeology and Ethnohistory in the Arrow Lakes, Southeastern British Columbia, Ottawa: National Museum of Man Mercury Series.

Turner, N., 1977. Economic Importance of Black Tree Lichen (Bryoria fremontii) to 230

the Indians of Western North America. Economic Botany, 31(4), 461-70.

Turner, N., 1988. Ethnobotany of Coniferous Trees in Thompson and Lillooet Interior Salish of British Columbia. Economic Botany, 42(2), 177-94.

Turner, N. J., 1981. Indian Use of Shepherdia canadensis, Soapberry, in Western North America. Davidsonia, 12(1), 1-14.

Turner, N. J., 1992. Plant Resources of the Stl'atrimx (Fraser River Lillooet), in A Complex Culture of the British Columbia Plateau: Traditional Stl'atrimx Resource Use, ed. B. Hayden, Vancouver: UBC Press, 405-69.

Turner, N. J., 1997. Food Plants of Interior First Peoples, Vancouver: University of British Columbia Press.

Turner, N. J., 1998. Plant Technology of First Peoples in British Columbia, Vancouver: University of British Columbia Press.

Turner, N. J., 1999. "Time to Burn" Traditional Use of Fire to Enhance Resource Production by Aboriginal Peoples in British Columbia, in Indians, Fire and the Land in the Pacific Northwest, ed. R. Boyd, Corvallis, Oregon: Oregon State University Press, 185-218.

Turner, N. J., 2003. 'Passing on the News': Women's Work, Traditional Knowledge and Plant Resource Management in Indigenous Societies of North-western North America, in Women & Plants: Gender Relations in Biodiversity Management and Conservation, ed. P. L. Howard, London: Zed Books, 133-49. 231

Turner, N. J., R. Bouchard & D. I. D. Kennedy, 1980. Ethnobotany of the Okanagan-Coville Indians of British Columbia and Washington, in Occasional Papers of the British Columbia Provincial Museum Victoria: Ministry of Provincial Secretary and Government Services.

Turner, N. J. & A. Davis, 1993. "When Everything was Scarce": The Role of Plants as Famine Foods in Northwestern North America. Journal of Ethnobiology, 13, 171-201.

Turner, N. J., M. B. Ignace & R. Ignace, 2000. Traditional Ecological Knowledge and Wisdom of Aboriginal Peoples in British Columbia. Ecological Applications, 10(5), 1275-87.

Turner, N. J., H. V. Kuhnlein & K. N. Egger, 1987. The Cottonwood Mushroom (Tricholoma populinum): A Food Resource of the Interior Salish Indian Peoples of British Columbia. Canadian Journal of Botany, 65, 923-7.

Turner, N. J. & D. C. Loewen, 1998. The Original "Free Trade": Exchange of Botanical Products and Associated Plant Knowledge in Northwestern North America. Anthropologica, 40(1), 49-70. Turner, N. J., L. C. Thompson, M. T. Thompson & A. Z. York, 1990. Thompson Ethnobotany: Knowledge and Usage of Plants by the Thompson Indians of British Columbia, Victoria: Royal British Columbia Museum.

Turner, N. J. & K. L. Turner, 2007. "Rich in food": Traditional Food Systems, Erosion and Renewal in Northwestern North America. Indian Journal of Traditional Knowledge (special issue on Ethnic Foods), 6(1), 57-68.

Turner, N. J. & K. L. Turner, 2008. Where Our Women Used to Get the Food: Cumulative Effects and Loss of Ethnobotanical Knowledge and Practice; Case Study from Coastal British Columbia. Botany, 86,103-15. 232

Twiss, P. C., 2001. A Curmudgeon's View of Grass Phytolithology, in Phytoliths: Applications in Earth Sciences and Human History, eds. J. D. Meunier & F. Colin, Lisse: A.A. Balkema, 7-26.

Ugent, D., 2006. History of Starch Classification, in Ancient Starch Research, eds. R. Torrence & H. Barton, Walnut Creek, California: Left Coast Press, 116-9.

Ugent, D., T. Dillehay & C. Ramirez, 1987. Potato Remains from a Late Pleistocene Settlement in South Central Chile. Economic Botany, 41(1), 17-27.

Ugent, D., S. Pozorski & T. Pozorski, 1982. Archaeological Potato Tuber Remains from the Casma Valley of Peru. Economic Botany, 36(2), 182-92.

Ugent, D., S. Pozorski & T. Pozorski, 1984. New Evidence for Ancient Cultivation of Canna edulis in Peru. Economic Botany, 38(4), 417-32.

Ugent, D., S. Pozorski & T. Pozorski, 1986. Archaeological Manioc (Manihot) from Coastal Peru. Economic Botany, 40(1), 78-102. Usher, P. J., 2000. Traditional Ecological Knowledge in Environmental Assessment and Management. Arctic, 53,183-93.

Van Peer, P., R. Fuliagar, S. Stokes, R. M. Bailey, J. Moeyersons, F. Steenhoudt, A. Geerts, T. Vanderbeken, M. De Dapper & F. Geus, 2003. The Early to Middle Stone Age Transition and the Emergence of Modern Human Behaviour at site 8-B-11, Sai Island, Sudan. Journal of Human Evolution, 45 (2), 187-93.

Wadley, L., M. Lombard & B. Williamson, 2004. The First Residue Analysis Blind Tests: Results and Lessons Learnt. Journal of Archaeological Science, 31 233

(11), 1491-501.

Wandsnider, L., 1997. The Roasted and the Boiled: Food Composition and Heat Treatment with Special Emphasis on Pit-Hearth Cooking. Journal of Anthropological Archaeology, 16(1), 1-48.

Ward, H. M., 1895. Book Review of Untersuchungen uber die Starkekorner. Nature, 52(1357), 640-1.

Weber, A. P. M., 2006. Synthesis, Export and Partitioning of the End Product of Photosynthesis, in The Structure and Function of Plastids, eds. R. R. Wise & J. K. Hoober, Springer, 273-92.

Whistler, R. L. & W. L. Thornburg, 1957. Starch Granule Formation, Development of Starch Granules in Corn Endosperm. Journal of Agricultural and Food Chemistry, 5(3), 203-7.

Wickwire, W., 1993. Women in Ethnography: The Research of James A. Teit. Ethnohistory, 40(4), 539.

Wilson, R. L., 1973. Salvage Excavations in the Vicinity of Kamloops, B.C. Report on file with the British Columbia Archaeology Branch, Victoria, B.C.

Wilson, R. L., 1980. Archaeological Investigation Near Kamloops, in The Archaeology of Kamloops, Burnaby, B.C.: Department of Archaeology, Simon Fraser University.

Wilson, T., 1916. The Use of Wild Plants as Food by Indians. Ottawa Naturalist, 30 (2), 17-21.

Wittke, K. L., B. Hayden & M. A. Lauwerys, 2004. A Coiled Basket Fragment and 234

Other Organic Artifacts from the Keatley Creek site, British Columbia. Canadian Journal of Archaeology, 28(1), 144-50.

Wodehouse, R. P., 1935. Pollen Grains: Their Structure, Identification and Significance in Science and Medicine, New York: Hafner Publishing.

Wollstonecroft, M. M., 2000. The Fruit of Their Labour: A Paleoethnobotanical Study of Site EeRb 140, a Multi-Component Open-Air Site on the British Columiba Plateau, in Department of Archaeology, Burnaby: Simon Fraser University.

Wollstonecroft, M. M., 2002. The Fruit of their Labour: Plants and Plant Processing at EeRb 140 (860+/- 60 Uncal to 160+/- 50 Uncal B.P.), a Late Prehistoric Hunter-Gatherer Site on the Southern Interior Plateau, British Columbia, Canada. Vegetation History and Archaeobotany, 11, 61-70.

Woolfe, J. A., 1987. The Nutritional Value of the Components of the Tuber, in The Potato in the Human Diet, Cambridge: Cambridge University Press, 19-57.

Wright, M. C., 2003. The Woman's Lodge: Constructing Gender on the Nineteenth- Century Pacific Northwest Plateau. Frontiers: A Journal of Women Studies, 24(1), 1-18.

Wyatt, D., 1971. A Preliminary Outline of Nicola Valley Prehistory, in Aboriginal Man and Environments on the Plateau of Northwest America, eds. A. H. Stryd & R. A. Smith, Calgary: The University of Calgary Archaeological Association, The Students' Press, 60-72.

Wyatt, D., 1998. Thompson, in Handbook of North American Indians, ed. D. E. 235

Walker, Washington: Smithsonian Institution, 191-5.

Wylie, A., 1985. The Reaction against Analogy. Advances in Archaeological Method and Theory, 8, 63-111.

Yoo, S.-H. & J.-l. Jane, 2002. Structural and Physical Characteristics of Waxy and Other Wheat Starches. Carbohydrate Polymers, 49(3), 297-305.

Zarrillo, S., 2010. Residue Extraction and Slide Residue Mounting Procedure, Calgary: Latin American Studies Laboratory, Department of Archaeology, University of Calgary.

Zarrillo, S. & B. Kooyman, 2006. Evidence for Berry and Maize Processing on the Canadian Plains from Starch Grain Analysis American Antiquity, 71(3), 473-99.

Zarrillo, S., D. M. Pearsall, J. S. Raymond, M. A. Tisdale & D. J. Quon, 2008. Directly Dated Starch Residues Document Early Formative Maize (Zea mays L.) in Tropical Ecuador. Proceedings of the National Academy of Sciences, 105(13), 5006-11.

Zobel, H. E, 1988. Molecules to Granules: A Comprehensive Starch Review. Starch - Starke, 40(2), 44-50. 236

APPENDIX A: Residue Extraction and Slide Mounting Procedure

Preparation 1. Wash hands upon entering 749 Latin American Research Lab and wear no make-up, hair products or body powders. Put on brightly coloured scrub top. 2. Fill ultrasonic bath with dH20, up to 1-2 inches from the top.

Vinegar-boiling to Clean Lab Supplies 1. Prepare pressure cooker by checking that the holes in the lid are clear, and placing the biggest pressure control knob (15 PSI) on the vent tube. If needed, lubricate the gasket with a small amount of Vaseline before placing the gasket around the lid rim. 2. Place metal tongs, plastic containers, glass microscope slides, plastic containers, plastic weigh boats, plastic pipettes and pipette holders in the pressure cooker (number of plastic containers, slides and pipettes=number of tools to be processed + one for control). 3. For plastic pipettes, suck up vinegar before boiling begins. 4. Fill pot of pressure cooker with vinegar 2/3 to 1/2 full, depending on how much plastic and glassware is being cleaned. 5. Close lid clockwise, making sure the top lid handle is directly on top of pot handle when closed. Put on hotplate and set to max heat. When you hear the pressure control knob jiggle and release steam, lower heat so that the pressure control rocks gently. 6. Let boil for at least 30min to ensure total gelatinization of any modern starch, turn off heat and let cool. Do not attempt to open a warm pressure cooker as the contents are still under pressure and steam burns can occur. You can speed cooling of a sealed pressure cooker by dousing the pot in cool water in the sink. 7. Once cool, remove all plastic and glassware from the pressure cooker with poly gloved hands and place in mesh basket over sink. 237

8. Use jugs of deionized water (dhbO) to rinse off the vinegar from the cleaned plastic and glassware. Be sure to suck up and rinse out the inside of the plastic pipettes with dhteO. 9. Set cleaned materials on a clean cotton cloth to air dry. Any clean materials not being used immediately must be placed and sealed into new zip lock plastic bags. 10. With a funnel, pour vinegar back into original containers for re-use 11. Clean pressure cooker with AJAX bleach powder and steel wool and set out to air dry.

Prepare for Residue Extraction 1. Lay out a long strip of plastic wrap as a working surface on the counter. Set out artifacts in their bags, evenly spaced on the plastic wrap. 2. Label a set of centrifuge tubes with a residue-dedicated permanent pen. Label the set of tubes with the lithic artifact number, EX 1 (extraction 1) and date. Label one control tube with EX 1 and the date. Place all labeled centrifuge tubes in a holding rack. 3. Take out vinegar-boiled plastic pipettes for the number of lithics being processed plus 2, re-seal bags. Label each plastic pipette with artifact number, plus one labeled Nhh for use with the 5% ammonia. Place in pipettes holder. 4. Take out appropriate plastic containers for the number and size of lithics, reseal bag. Label each plastic container with artifact number and set out plastic containers by each artifact, still in their bags. 5. Set out the cleaned pipette holder. Cover sides and bottom of holder with aluminum foil.

Jet Bath Wash of Artifacts (N.B. not recommended for future residue studies) 1. Lay out a very long strip of plastic wrap on some cleaned counter space. 2. Label one poly glove per artifact with artifact number and lay out on plastic wrap in order of processing. 238

3. Put on one poly glove (on left hand if right-handed). 4. Remove first lithic from its bag with the gloved hand. 5. Hold the lithic with the gloved hand and wash with a jet bath stream of dhteO over a plastic weigh boat. 6. Set each lithic into a plastic container and place with each artifact bag. Remember to set out one blank/control plastic container. 7. Remove glove without touching the outside of the glove with the other hand and set on the plastic wrap strip. 8. Pour wash water into JB labeled centrifuge tube. 9. Repeat steps 21-25 until all lithics have had a jet bath and are in plastic containers. 10. If jet baths are to be made into slides, centrifuge for 5min at 3,000 RPM and carefully pour off the extra water from the tube.

Residue Extraction 1. Open each artifact bag and place lithic in its labelled plastic container, without touching the artifact with your hands. 2. With the pipette labeled Nhh, suck up 5% ammonia and place in all plastic containers, enough to cover the lithics as entirely as possible. If plastic pipette touches a lithic at any time, discard immediately and use a new one. 3. Put lids on plastic containers to prevent spillage during sonication. 4. Place the lithics/plastic containers carefully in the sonicator, spaced evenly away from each other and the walls of the sonicator. 5. Turn the sonicator on and set the timer for 30 minutes. 6. After 30 minutes, pull plastic containers carefully out of sonicator and place on counter. 7. Put glove on that corresponds to the lithic, pick up lithic with gloved hand. 8. While holding lithic over its plastic container, rinse off the ammonia from the lithic with a water bath to remove any extra sediment. 9. Place lithic beside its bag on the plastic wrap for air drying. 239

10. With labeled pipette, suck up sediment and ammonia from plastic container, dispense liquid into labeled centrifuge tube and place in the rack. 11. Put pipette back into holder and set aside plastic containers for vinegar cleaning.

Centrifuge 1. Balance all centrifuge tubes using dH20 squeeze bottle, fill to 15ml without touching the bottle tip to the centrifuge tube. Place all tubes into the centrifuge evenly spaced and balanced. 2. Press the green start button and centrifuge will run for 5 min at 3,000 RPM. 3. Remove all tubes from centrifuge and place in rack. 4. At the sink, suck up almost all of the liquid supernatant with the appropriate pipette and discard into the sink. Do this for all vials and place back in rack. Do not throw away pipettes. 5. Fill centrifuge tubes with dH20, all tubes up to10ml or all to15ml 6. Use vortex to break up sediment before tubes are placed into centrifuge again. 7. Repeat steps 40-45 three more times, so the tubes are centrifuged a total of 4 times.

Slide Mounting 1. Put on scrub top. 2. Make sure working counter is clean. If dusty, clean counter with tap water and cotton cloth. 3. Lay down a sheet of poster and a sheet of plastic wrap on top of it as a working surface. Set out plastic pipette tips into holder and close lid. Set out 100% pure glycerol (refractive index 1.475), a tiny beaker, microscope slides, slide covers, residue-dedicated pencil, residue-dedicated clear nail polish and toothpicks. 240

4. Lay out two slides per lithic residue extraction and label with artifact number, extract number (eg. EX 1) and slide number (eg. SL 1, SL 2) plus one for the control. 5. Using VWR pipette, attach tip. Aspirate 20|JL of wet residue from the bottom of the centrifuge tube and place on labeled SL 1; repeat for SL 2. Eject the tip by holding its base and placing it beside the slides. Do this for each residue sample in centrifuge tube. 6. Allow residue to air dry on the slides. 7. Pour about 2mL of pure glycerol into the tiny beaker. 8. Take out another pipette tip for use with the glycerol and all slide mounts 9. Aspirate 20pL of glycerol. Place glycerol on each slide. If the dry residue appears dirty and contains lots of sediment, use 30|jL of glycerol on the slide. If pipette tip touches sediment, discard immediately and use a new one. 10. Mix the dry residue extract and glycerol with the ejected pipette tip on slides 1 and 2 (same extract on both slides). 11. Clean each glass cover slip with alcohol wipe and Kimwipe. 12. Cover with cover slips. 13. If there is excess liquid in the mount coming out from the sides of the slide cover, gently soak it up with a small piece of Kimwipe. 14. Open nail polish and place brush standing up by the cap. Dip one toothpick into the nail polish and pull out about a drop of polish and seal one slide cover edge. For each side of the slide cover, use a new toothpick and discard. Do not reuse toothpicks. 15. Cover a box tray with aluminum foil and place completed slides on it 16. Turn on oven to 40-45°C. Place tray and slides into oven and dry for -30 minutes or until dry. Remove tray from oven using tongs or cotton cloth if tray is hot. 17. If any slides appear to be leaking after removal from oven, put a second coat of nail polish on the edges of the slide cover using toothpick technique outlined above. 241

Maintenance 1. Clean counters with dhhO and a clean cotton cloth before beginning each residue extraction or slide mounting. 2. Use Swiffer dry and wet cloths to clean floors when any dust is seen accumulating on floor. 3. Refill jugs with dhhO. 4. Wash cotton cloths on their own in bleach only. 5. Test each new shipment of non-powdered starch-free gloves to ensure they do not contain starches. Also test each new jug of ammonia. 6. Lay out a slide on the countertop and label it 'control' with the date, every week place a new one down, mount the old one and check it for starch.

This procedure of starch residue extraction and slide mounting is a modification of protocol used by Sonia Zarrillo (PhD candidate, University of Calgary) (2010) and Newman and Julig (1989:121). Residue extractions and slide mounting took place in the Latin American Research Laboratory, where Sonia Zarrillo conducts her research at the University of Calgary.

Legend for Appendices B and C: a= absent p= present s= simple grain c= closed hilum o= open hilum cen= centric hilum e= eccentric hilum s.e.= slightly eccentric hilum 242 243 APPENDIX B: All Residues from Lithic Sample tfi (0 w T3 _ ® (D * _ c re re £"d i® a> o © i O) fl) ore .3 re re a> CO o 1. o 4= DVC W JC 3* w£ C Q- TJg re £re w mre re 03 33 to> 5* O re re x co m U.X Si­ Q.HLL W U1 X OO 6355 steep edge a a N/A EX1 SL1 angle CS2

6350 steep edge p brown infilled p a p p a a a a N/A EX1 SL1 angle CS2 elongate 4245 steep edge P a a p a a a a a N/A EX1 SL1 angle CS2 3830 steep edge P a a 6 N/A EX1 SL1 angle CS2 5364 steep edge ;p a a p? 13 N/A EX1 SL1 angle CS2 6114 steep edge iP a a wood cell w/ 4 N/A EX1 SL1 angle CS2 adjoining cell w/ simple pits 4241 steep edge ;p p p a a a N/A EX1 SL1 angle CS2 6389 steep edge p trichomes a a insect or plant N/A EX1 SL1 angle CS2 eggs? est. 87 non starch in a aggregation? 17661 steep edge brown infilled a a spiral thickening on 1 N/A EX1 SL1 angle CS2 elongate wood cell, wood frags present

6355 steep edge P p p a a a a N/A EX1 SL2 angle CS2 j 9993 steep edge P triangulars p p a a p a a N/A EX1 SL2 angle CS2 APPENDIX B: All Residues from Lithic Sample Bath Tool Type Other Starch Slide # Frags Plant Funaal Hyphae Tissue Hair Modern Thread Jet Calcium Oxalate Elements Other Phytoliths Grass Phytoliths Vessel Tracheids ! 9995 steep edge P a a P P P a a a rootlet w/ clumps? 12 N/A EX1 SL1 angle CS2 28.5,94.5 5357 steep edge P a P a P P a a P a a N/A EX1 SL2 angle CS2 6357 steep edge P a a a a p? a a a a a N/A EX1 SL1 angle CS2 18995 steep edge P bell P a P P a a a a a N/A EX1 SL2 angle CS2 6348 steep edge P casket, bell? a a P P a a P a a N/A EX1 SL2 angle CS2 13577 steep edge P golf club a 'a iP P a a P a 2 N/A EX1 SL2 angle CS2 10522 steep edge P brown infilled lobed a a P P a a P unidentifiable wood 1 N/A EX1 SL2 angle CS2 grass phytoliths? frags present 3831 steep edge P brown infilled lobed a a P P a a a a a N/A EX1 SL1 angle CS2 grass phytoliths? 6358 steep edge P jagged a a P P a a a insect integument? a N/A EX1 SL2 angle CS2 unidentifiable wood frags present

18997 steep edge P cones P a P P a a P a 1 N/A EX1 SL1 angle CS2

11355 steep edge P a a a p P a a a a 1 N/A EX1 SL1 angle CS2 16435 steep edge P a P P P P a a P a a N/A EX1 SL1 angle CS2 APPENDIX B: All Residues from Lithic Sample w (0 » in _ Etj a) x: "53 _c an 0) ca i® o re 0) (0 o wo 0)0) °>S .2 to c « o> £ CQ 2 ** £ «E •o® .fc £co o »aj o (0"* (0 X n«2 a> CO si •j-x:s (0 d>— LLX X Q.HLL co OCL OCL >UJ Sh OO -> 9991 steep edge unidentifiable wood N/A EX1 SL2 angle CS2 frag present 6345 flake with club a p? triangular fungal N/A EX1 SL1 steep-angled spore? marginal unifacial retouch CS2 9979 flake with N/A EX1 SL1 steep-angled marginal unifacial retouch CS2 10518 flake with a ia p p a a N/A EX1 SL2 steep-angled marginal bifacial I retouch CS2 i 10519 steep-angled brown infilled elong, a p p a a a a N/A EX1 SL2 formed uniface clubs CS2 16099 steep edge brown infilled elong, a a p a a a a epidermal phytoliths a N/A EX1 SL1 angle formed clubs uniface CS2 9989 acute edge ; P p p a a JB, no EX1 SL2 angle CS2 starch 10516 acute edge !P i rounded ends ! a i a I p p a a JB, 3 EX1 SL2 angle CS2 ; elongate, ruffle i starch ; surface APPENDIX B: All Residues from Lithic Sample <0 CO +->(0 _ 0) £ £ TJ 4) * ** _c ETJ; "53 m 0) to « d> (0 .2i* ra •S3" ® o (0 2* «o k- o .c OiC •o® c#l® JZ m m o »E C o. Or ii (a 2 ~ 55 El1 (0 sl 2i= (0 co x Q.HU. V) ® Oa. > >Lli LLX x OO 6352 acute edge ) a p? triangular fungal 1 JB EX1 SL1 angle CS2 spore? 9866 acute edge a 2 JB, no EX1 SL1 angle CS2 starch 13569 acute edge P a a a a 1 JB, 6 EX1 SL2 angle CS2 starch 4242 acute edge P ; rounded ends a JB EX1 SL2 angle CS2 elongate, ruffle ; surface, square 18996 acute edge |P a JB, 6 EX1 SL2 angle CS2 starch 6346 acute edge ;P a a p? a a JB EX1 SL1 angle CS2 9983 acute edge P a a a JB EX1 SL2 angle CS2 3675 acute edge P {P a a rootlet? a JB EX1 SL1 angle CS2 3677 acute edge P bilobe P- P P a a a JB EX1 SL2 angle CS2 hard j wood 9 9985 acute edge P club a a a a a ; N/A EX1 SL2 angle CS2 16434 acute edge P a a a N/A EX1 SL2 angle CS2 APPENDIX B: All Residues rom Lithic Sample (0 V) CO (0 * £ TJ L_ a) 4jjjJ _C in ra £-a io>c a> ra .2i* re o H » to s< O © si U.I II X CLi—LL IW -» 00. OQ. £ >UI OO 13574 acute edge plumulaceous 1 , N/A EX1 SL1 angle CS2 feather barbule 6354 acute edge 1 N/A EX1 SL2 angle CS2 4246 acute edge P a a 1 N/A EX1 SL2 angle CS2 9990 acute edge P rounded ends a N/A EX1 SL2 angle CS2 elongate, ruffle surface 9987 acute edge fungal spores a N/A EX1 SL2 angle CS2 10520 acute edge club, rounded ends ; a a a 1 N/A EX1 SL1 angle CS2 elongate ruffle surface 9994 acute edge P brown infilled a a a N/A EX1 SL1 angle CS2 elongate 4244 acute edge P brown infilled P a a a N/A EX1 SL1 angle CS2 elongate 16436 acute edge P 3 bump, club P a a a N/A EX1 SL1 angle CS2 3694 acute edge P 6 N/A EX1 SL2 angle CS2 3652 acute edge iP brown infilled P. a N/A EX1 SL2 angle CS2 elongate 3-4 types seen APPENDIX B: All Residues rom Lithic Sample Bath Tool Type Other Starch Slide # Plant Frags Fungal Hypnae Tissue Modern Thread Jet Calcium Oxalate Hair Elements Grass Phytoliths Other Phytoliths Vessel Tracheids

3824 acute edge P club P P P. 2 a a a P diatom? spore? 3 N/A EX1 SL2 angle CS2 types 9981 acute edge P a P P P a a a a woodworking tool? a N/A EX1 SL2 angle CS2 9992 acute edge P a a a P P a a a a a N/A EX1 SL1 angle CS2 9988 acute edge P club a a P a a a a pitted wood frag, a N/A EX1 SL1 angle CS2 present 19001 acute edge P brown infilled a a P P a a a a 1 N/A EX1 SL1 angle CS2 elongates 11181 acute-angled P club a a P a a a a a a N/A EX1 SL1 utilized flake CS2 in feature 18999 acute-angled P a a a P a a a a a a tool fell EX1 SL2 biface end on floor fragment CS2 during EX 1

6344 utilized P P? a a a P a a P a a N/A EX1 SL1 microblade CS2 6383 complete P rondel a P P a a a a a N/A EX1 SL1 microblade CS2 11076 complete P a a a P P a a a a a N/A EX1 SL1 microblade CS2 13570 complete P a a a a P a a P a a N/A EX1 SL1 microblade CS2 APPENDIX B: All Residues from Lithic Sample i j Bath L. Tool Type Other Starch

Slide # '5 Plant Frags Fungal Hypnae Tissue Modern Thread Jet Calcium Oxalate

Elements X Other Phytoliths Grass Phytoliths Vessel Tracheids i i r 13572 complete P a P P P P a a P a a N/A EX1 SL1 microblade CS2 6377 microblade CS2 P a P a a P a a a a 1 N/A EX1 SL1 6384 microblade P a P a a P a a a a a N/A EX1 SL2 fragment, proximal end CS2 11368 microblade CS2 P golf club a P a P a a a a 1 N/A EX1 SL1 6368 microblade P a a a a P a a a a a N/A EX1 SL1 fragment, medial portion CS2 6386 microblade P a a P a P a a a a a N/A EX1 SL1 fragment, medial portion CS2 10527 microblade P a a a a P a a a a a N/A EX1 SL1 fragment, medial portion CS2 11371 microblade P a a a a a a a 'a wood frag, present a N/A EX1 SL1 fragment, medial portion CS2 APPENDIX B: All Residues from Lithic Sample Bath Type Tool Other Starch Slide # Plant Frags Fungal Hypnae Tissue Modern Thread Jet Hair Calcium Oxalate Grass Phytoliths Elements Other Phytoliths Vessel Tracheids 11366 microblade CS2 P squares a P P P a a a a 2 N/A EX1 SL1 11357 microblade- a a a a P P a a a P a N/A EX1 SL2 distal end CS2 4249 microblade- a a a a a P a a a a a N/A EX1 SL1 distal end CS2 10002 microblade- P a a a P a a a a a a N/A EX1 SL2 distal end CS2 4637 microblade- P brown infilled club a a P a a a a a a N/A EX1 SL1 distal end CS2 10526 microblade- P a a a P P a a a a a N/A EX1 SL1 distal end CS2 9984 retouched p a P a P a a a P a a N/A EX1 SL2 macroblade fragment, proximal end CS2 11356 split biface CS2 P a a a P P a a P? a 2 N/A EX1 SL1 13575 split biface CS2 P wedge-shaped, a a P P a a la a :1 N/A EX1 SL1 triangular, blocky shaped 1560 steep edge P wavy bumps on a a P a a a P? unidentifiable wood 18 N/A EX1 SL1 angle CS1 interior of elongate tissue present 2022 steep edge P a a a P P a a a a a N/A EX1 SL1 angle CS1 APPENDIX B: All Residues from Lithic Sample Bath Tool Type Other Starch Plant Frags Fungal Hypnae Tissue Modern Slide # Thread Calcium Oxalate Hair Elements Jet Other Phytoliths Grass Phytoliths Vessel Tracheids ' 2169 steep edge P triangular a a P P a a a a a N/A EX1 SL1 angle CS1 2178 steep edge P a a a P P a P? P a 5 N/A EX1 SL1 angle CS1 1830 steep edge P a a a P P a a P a a N/A EX1 SL1 angle CS1 2133 steep edge P a a a P P a a a a a N/A EX1 SL1 angle CS1 2295 steep edge P a a a P P a a a a a N/A EX1 SL1 angle CS1 1919 steep edge P triangular ; a a P P a a a a 1 N/A EX1 SL1 angle CS1 2433 acute angle P brown infilled a a P. P a p? a a 1 N/A EX1 SL1 utilized flake elongate 3-4 CS1 types seen 2437 acute angle P brown infilled a a P, 2 P a a a a a N/A EX1 SL2 utilized flake triangular types CS1 3570 acute angle P a a a P P a a a a a N/A EX1 SL2 utilized flake CS1 3676 acute angle P textured elongate P ;P P. 2 P ;a a a a a N/A EX1 SL1 flake w/ types marginal unifacial retouch CF18 APPENDIX B: All Residues from Lithic Sample

%

a> Bath

ft* Other Starch Plant Frags Fungal Hypnae

<75 Tissue Modern Thread Jet Calcium Oxalate Hair Elements Phytoliths Other Phytoliths Grass Vessel Tracheids 2294 acute angle j P club a a P a a a a a a N/A EX1 SL2 flake w/ unifacial retouch on alternate faces CS1 2486 acute angle pappaaa a a N/A EX1 SL2 flake w/ marginal unifacial retouch CS1 1790 acute angled a a p p a a a a 38 N/A EX1 SL2 flake w/ bifacial : retouch CS1 1794 acute angle : triangular fungal a N/A EX1 SL2 flake w/ spore? pitted wood marginal bifacial; frag, present retouch CS1 1814 acute angle p? a a N/A EX1 SL1 flake w/ marginal bifacial retouch CS1 1951 acute angle brown infilled p a p, 2 p a p? a a a N/A EX1 SL2 flake w/ elongate types marginal bifacial retouch CS1 APPENDIX B: All Residues from Lithic Sample tn to (0 +•>(0 sz T3 _ c _ 4) Eu (U * •J ® ran ©ra .2 ra o> O ra a> (0 o w"o C. 0)0) o>c TJ 0) C (fl® JZ lm CD 2 2* 0)+? O »E c o. Op JO v> 2 03 yj ssr (0 $« 3 > ra ra x CLi—LL V) © OQ. On. >LU llX Sh x OO ~3 3568 acute angle a N/A EX1 SL2 biface end fragment CS1 3569 acute angle brown infilled | p a N/A EX1 SL1 biface medial elongate fragment CS1 2417 acute biface brown infilled p 1 N/A EX1 SL1 pieces CS1 , elongate phytolith 3654 acute angle P, 2 a a N/A EX1 SL2 flake w/ = types marginal bifacial; retouch CF9 3651 broken maul textured elongate p p p p a p? p spruce pollen 13 N/A EX1 SL2 surface CF9 grains, 7 articulated plant fragments of epidermal phytoliths, lichen soredia 3674 cobble with p a p p a p? p saccate pollen N/A EX1 SL2 linear striations grains, animal toss zone CF18 tissue? 3699 cobble with P P P P a a p pollen grains? 2 N/A EX1 SL2 linear striations surface CS2/ CF17/CF18 151 APPENDIX C: Starch Morphotypes

E c a) re * ®r> czz o o a 2 *•> re 1 s >:.2 LL sz < 2 <5 •a H +£(0 C oc «(0 W II E w V) ° y re = « II 3 o ••§ 2 o I > X O H -1 O LU O CO £ 6114 flake with s fuzzy diffuse ovate none for s.e. N/A EX1 steep- hilum, SL1 angled but marginal bright unifacial pink retouch marking CS2 s around edge of grain fuzzy diffuse could be c? none cen N/A ovate disc

intermediate irregular o none s.e. darkened oval in cross polarized light 4 s intermediate super none darkened irregular in polygon polarized light

1 17661 steep edge s intermediate irregular none N/A EX1 angle oval SL1 formed uniface CS2 1 9995 flake with s fuzzy pear- none N/A EX1 steep- shaped SL1 angled oval marginal unifacial retouch CS2 2 C s fuzzy and slightly none cen light at hardly visible ovate top of grain, dark pink as you focus down APPENDIX C: Starch Morphotypes |jm Other Notes Fissures Grooves Surface Texture Lamellae Facets Max Length Taphonomy 'a a P a furrowed 21.3751 two cuts texture or fissues i at edges of grain

; short deep a P a smooth 35.625 N/A can't straight fissure rotate, in over hilum, 1 fine mountin i scratch near edge 9 medium w/ bubbles i deep arced a P a rough 19.095 j rough ! fissure, stellate, ;edges ; both over hilum I | i I t «transverse fissure a a a rough 19.95 j fissures at j edges of j grain and | nicks over ! surface 2 fissures cross a P a rugulose 19.951 edges of each other in an V from tiny ; grain or X depending fissures ! roughened ; on view, not over i

extinction cross ' i i . 1 arced end to a a a smooth 16.53 iN/A end

;

1 : ia a 1 ring p near j a smooth 7.98 N/A grain edge APPENDIX C: Starch Morphotypes re * * ® c O o a Si £ S o a) gl U. +3 re < 2 re JC H +; c O (0 .c a 5 re C ID w r=II OT E Wo 55 o o e '+> o a 3 o O o CO X > iiX o s r- -I o UJ O extremely ovate disc c none, cen? in polar fuzzy when in but flat view, flat view. Arms grain hilum distinct when grey or area rotated on darkene complete edge and d iy appear to darkened cross each .shadowy other twice in in a wavy pattern polarized light fuzzy oval with c none N/A 2 facets? A fuzzy, arms ovate disc c none cen N/A very diffuse in flat view, arms more distinct when rotated fuzzy irregular o pink cen N/A sphere tinge

G intermediate polygon o none cen N/A

intermediate ovate o in none N/A pear- top shaped down view intermediate irregular brown? cen N/A sphere

11 fuzzy and near none cen depresse hardly visible sphere or d. Hilum slight oval can appear as solid, ring or faint large depresse d pink area APPENDIX C: Starch Morphotypes Mm Other Notes Fissures Grooves Surface Texture Lamellae Facets Max Length Taphonomy a equatorial a a smooth 25.935 one edge groove with of grain around undulating appears edge 'waves' rough

p, 2 j rough 11.4 N/A

equatorial a a ; smooth 17.1 N/A groove ; around i edge

; a ; smooth 7.125 (rough jedges

; p- stellate p 3 w/ j rough 19.951 N/A rounde j i d edges j

: a I smooth 17.671 central j channel ! hole from through enzymatic [centreof |action? grain I smooth 8.55 islightly S rough j edges i 1 ring p near I a 1 smooth 5.985 N/A i grain edge APPENDIX C: Starch Morphotypes Area Starch Grains Slide # Hilum Colour Hilum # Visibility 3D Shape Hilum Placement Location Grain Form Extinction Cross Tool Type, Hilum Morphotype 12 D s intermediate gumdrop c none cen? N/A bell, bowl sides, straight base

1 6377 microblade s fuzzy irregular 0 none e appears EX1 fragment, oval, when dark to SL1 proximal Chinese probe light as end CS2 hat shape d you focus down through the grain

1 111368 microblade s intermediate irregular 0 none cen N/A EX1 fragment, oval SL1 proximal end CS2 1 13577 steep- s intermediate polygon 0 purple e darkened EX1 angled round , prob a SL2 utilized edges depressi flake CS2 on 2 s intermediate sphere 0 none e dark at slightly top view, irregular lightened at lower view 1 G 10522 steep- s distinct polygon 0 none s.e. N/A EX1 angled SL2 utilized flake CS2 1 18997 Flake with s fuzzy sphere o none cen darkened EX1 steep- SL1 angled marginal bifacial retouch CS2 APPENDIX C: Starch Morphotypes JH (A in 0) O) E O C o k.a> COa* 3 o ® CO 3 c r~ *» CO 0) H o ^ O CO S E 0 %3 *0) X £ il o CO 01 (0 F a a O Z LL 5 £

; perhaps 22.8 j cratering ; small cratered, ;from ! starch rough and ! enzymes? 13 is grainy stuck to ! surface, starch I fungal 12. : hypha and ! other ! material j associated : X fissure over ; rugulose 22.8'could in side .hilum, fine i due to ;have view, ; fissures around ' many fine milling pouty : edges ! radial ! damage- lips look ; fissures i tiny where : fissures central | around channel : granule enters I edges into the grain p small stellate a ; rough 16.245 N/A

smooth 19.38! N/A

2 fissures across ; a i smooth 12.825 ragged hilum forming an ; ; roughened! : X (stellate :edge ! i fissures) !

small fissure S cratered 17.1 N/A across hilum

a i smooth 20.52 N/A APPENDIX C: Starch Morphotypes Area Starch Grains Slide # Hilum Hilum Colour # Visibility 3D Shape Location Hilum Placement Grain Form Extinction Cross Tool Type, Hilum Morphotype 1 11355 steep- s fuzzy ovate disc c none cen ragged EX1 angled depresse SL1 utilized d area flake CS2 F 11366 microblade fuzzy very none s.e. darkened EX1 fragment, irregular in SL1 medial CS2 sphere polarized light

intermediate oval none s.e. lightened

D 11356 biface fuzzy gumdrop none darkened EX1 fragment, bell w/ in SL1 longitudinal curved polarized split CS2 bowl light sides, wavy base fuzzy irregular berry cen darkened sphere coloure d dot

C 13575 biface intermediate oval pink cen hilum EX1 fragment, but very faint oval area pink SL1 longitudinal shape and split CS2 depresse d APPENDIX C: Starch Morphotypes pm Other Notes Fissures Grooves Surface Texture Lamellae Facets Max Length Taphonomy longitudinal a a a smooth 25.65; N/A fissure a a a a rough 7.125ihilum area severely damaged, extinction cross not apparent, only edges of grain illuminated . Heat damage? a P a smooth 17.1 N/A starch grain could not be removed j from j adhering i plant material a a a a rough, 14.251 roughend j rough • edges

i

| a a *a a rough 8.55 very rough

• j edges, hilum area

f 'burnt out' ' ia arced 1 ring p near j a smooth 8.835 N/A I groove? .grain edge : APPENDIX C: Starch Morphotypes Area Starch Grains Slide # Hilum Colour # Hilum Visibility 3D Shape Location Hilum Placement Extinction Cross Grain Form Tool Type, Hilum Morphotype 1 G 1560 flake with s intermediate polygon 0 none s.e. N/A EX1 steep- SL1 angled marginal unifacial retouch CS1 2 G s intermediate polygon 0 none s.e. N/A blocky when probe d 3 s intermediate irregular 0 none cen darkened oval in normal light 4 s fuzzy tiny bell 0 pink e? N/A ring 5 A s fuzzy disc c none cen N/A

6 s intermediate squarish o dark e N/A oval 7 G s distinct polygon 0 none s.e. N/A

8 s fuzzy super c none not darkened irregular visib in oval or le polarized polygon light

9 G s intermediate polygon o none s.e. N/A APPENDIX C: Starch Morphotypes lim Other Notes Fissures Grooves Surface Texture Lamellae Facets Max Length Taphonomy X shaped fissures ; a a P rough and 14.25; N/A over the hilum cratered

Y shaped fissure a a P rough 17.67 N/A ; over hilum

; X or Y shaped a a a rough 19.95 N/A ; fissures over the hilum < 1 fissure over a a a smooth 3.135 N/A i hilum • a equatorial a a smooth 22.8 N/A groove around edge . .. , 1 short fissure a a a smooth 8.55 N/A over hilum IY shaped fissure a a P rough and 25.65 N/A {over hilum cratered or i pitted (?) a a P a rough 12.54 a little j hilum tunnel area bored into appears the grain, 'burnt enzymes? out' with only jedges ; showing i extinctio

; ! n cross. | question i able ; starch ; 1 through hilum a a P [slightly 19.95 N/A i rough or | cratered j APPENDIX C: Starch Morphotypes

o> -4-» a c 0) c <0 a> c o o a £ 0) 2 a 8 £ re < o © SiH +; u. JC 5 2 £ ;o re c W ll e i E a 8 8 —'1 r= O 3 O 55 o o — CO 3 * o .o p 2 s,s Q I > I O 2 a UJ o CO X Q. X 10 fuzzy irregular brown cen darkened sphere in polarized light

11 G s intermediate irregular o none cen N/A polygon

12 s fuzzy reniform c none s.e. N/A

13 s intermediate polygon o none cen N/A

14 G s intermediate polygon o none s.e. N/A

15 s intermediate irregular o pink dot cen N/A polygon or ring (depend s on focus) 16 intermediate ovate dark cen N/A irregular pink dot 17 intermediate bell, brown e N/A straight sides, straight base 18 intermediate polygon o none s.e. N/A APPENDIX C: Starch Morphotypes

>> (0 (/> 0) O) E © TO 4) ® i. « £ > It) U C o 4) 0) 3 "55 +* ro 3 § E c r* +- in o

| Y and stellate s a a smooth, 17.1 N/A ; pink i fissures I j slightly ' outline J irregular j around i edges j edge of j grain land fissures ! Y and stellate rough i 19.095 N/A : fissures ia smooth j 14.25 iN/A

| smooth 8.55 iN/A

one arced fissure a i smooth 8.55 N/A over hilum

Y, stellate fissures ; a a rough 19.95! N/A APPENDIX C: Starch Morphotypes a> a © CO a> r- c c a> £ % o o a a) L. I s o 4) u. n> < JC [O S-S O (0 -C 5 g a H_ •£(u c (fl CO l! EI E w6 V. c/5 o o (0 4= O •s — o —3 °(0 3 * O o o k. D X > X o S F -i 0 UJ o CO X Q. X 2178 steep- intermediate irregular o, berry s.e. N/A EX1 angled ovate can coloure SL1 utilized appe d or flake CS1 ar as clear faint white dot or clear white dot intermediate polygon o brown cen N/A

G intermediate polygon o brown s.e. N/A when probe d intermediate bell, o none N/A but faint straight sides, wavy base intermediate disc oval c none , cen darkened elongate in polarized light

A 1919 acute- fuzzy, arms disc oval c none cen darkened EX1 angled very diffuse in SL1 biface polarized medial light fragment CS1 03 03 & 030> CD c <303 Fissures

13574 flake with s intermediate sphere dark cen N/A EX1 acute- irregular berry SL1 angled colour marginal unifacial retouch CS2 APPENDIX C: Starch Morphotypes

> « (A © O) E V ra u o> 2 > 8 £ o o a> 3 O ® £ (Q 3 c (0 « o •cn "K° w 8 E o i3 *m 51X £ O Z il <5 (0 n W F <0 a 2 .ro faint short arc i smooth 14.25! N/A ; over hilum stellate one arced a I smooth 17.1 N/A ; groove over hilum ;fine radial fissures a ! p, especially! a ! smooth 17.385 ishort wide hilum around fissure at area very hilum ;edge of defined ; grain w/ donut encirclin git 1 ring or hint a smooth 9.405 irough of lamellae i ;edges : near grain j ! edge !

transverse fissure equatorial |a a ; rough 22.8 rough : groove j i surface iaround | iedge I

' 1 short arc fissure j a ! smooth 8.551 rough lover hilum ! |edges

;deep Y and faint -a S smooth 11.4 N/A i Receive 1 stellate fissures | ! d Jet iaround hilum j Bath, 6 ! starch i grains j found in JB I rough 9.975; N/A APPENDIX C: Starch Morphotypes CO d> _ c u. n < (0 H je = 1 •u _ (0 c S t (0 ii E tn w o o 'to = w = o _• .O o '*2 Q X > X o — ra a ui O CO X Q. x 6354 acute- intermediate oval o pink e darkened EX1 angled irregular in normal SL2 utilized and flake CS2 polarized light |A 4246 acute- fuzzy disc none cen? N/A EX1 angled SL2 utilized flake CS2

10520 acute- intermediate disc, oval c none cen darkened EX1 angled in SL1 utilized polarized flake CS2 light

A 3694 flake with intermediate disc w/ none cen N/A EX1 acute- uneven SL2 angled sides marginal (rounded bifacial Chinese retouch hat CF18 shape) intermediate disc w/ none cen N/A uneven sides intermediate, disc, oval none cen donut 2 arms are a egg around wide band shape hilum, darkened in polarized light intermediate disc, oval c none cen N/A egg shape A fuzzy disc w/ c none cen N/A uneven sides APPENDIX C: Starch Morphotypes

•C > (0 tn ffl O) © ra E >- V) £ > 8 £ o 0) a) 3 o © ro 3 § £ c (0 a> -J EL o £ o 01 8 E o 13 *a) x £ o z il a ra ra V) F n a u. S .ra ivery rough 19.951 could have ; grinding i damage

equatorial a | rough? lots 25.65 [surface groove ) of items j appears around I obstructing {pitted, X is; edge j view of ; very weak j ! grain deep stellate equatorial a j rough due 19.95 deep groove ' to fissuring i fissure at ; ! edge, damage to | idarkened j hilum area ! small arc over equatorial a i smooth 19.95 N/A hilum groove

j equatorial a smooth 29.07 N/A I groove

: small fissure equatorial a ismooth 28.5 damage to j I across hilum I groove hilum area, j strange X, j i donut

[equatorial a ! smooth 22.8 IN/A ! groove

equatorial a >a ; smooth 25.65 iN/A ; groove APPENDIX C: Starch Morphotypes 4-» a 4> _ uE c a> c re £ * o o a £ CD £ a 8 o a> LL +3 re < £ H£.1 +i j: is {5 a ;g — (u C 2 IS V) ii E E k. o O '5 —*1 (0 = o 3 l° 3 * o 55 .o O 12 a X > I o — re 2 O in o CO X Q. fuzzy disc w/ none cen donut uneven around sides hilum, hilum area cross inside the bigger cross in polarized light 3824 acute- fuzzy reniform, c?, none N/A EX1 angled irregular but SL2 utilized sides not flake CS2 really visibl e 2 not visible disc oval c none cen N/A elongate 3 intermediate, disc oval, c none cen N/A arms diffuse uneven btw groove 2433 acute- intermediate polygon o purple e N/A EX1 angled SL1 utilized flake CS1

G 1790 flake with intermediate polygon o none cen N/A EX1 acute- SL2 angled marginal bifacial retouch CS1 intermediate polygon o none s.e. N/A when probe d intermediate cuneiform o dark e depresse berry d and colour darkened APPENDIX C: Starch Morphotypes

to U> V o> 0) ra © a, E k. w £ > o c o 0> a) 3 o 75 ro 3 c (0 o 4) t S Si o £ o (0 U E O 3 0) X JC o (0 <0 W F (0 a 2 equatorial ismooth 29.07 N/A [All 6 groove starch on this slide appear ! to have similar [ morphol logy

small arc over i smooth 25.935 iX blurry hilum

equatorial p ; smooth ; 25.65 N/A groove j j equatorial a i smooth ! 27.931 N/A : groove !

• stellate ;a smooth 14.25! N/A j can't [ rotate, in : mountin 9 medium ' w/ ibubbles Y and stellate | rough 20.52 N/A fissures

' 1 fissure over : slightly 12.8251 N/A : hilum rough

smooth, w/ : 14.25 N/A < patches of j j shaded areas APPENDIX C: Starch Morphotypes d> -w (0 a a) c Q) C £ * o o a £ 0) £ i t o a) u. +3 n E < 2 jo JZ ;g O (/) •C a) a _ re c C to (0 E E (0 k. ° y "5 o ~ c/> = o 3 o 3 o o c/5 v. a x > I o CO 5 £3 a UJ O CO X Q. intermediate polygon o none N/A

intermediate polygon, o none s.e. N/A but cuneiform in some views intermediate irregular o none s.e. darkened polygon in normal rounded and corners polarized light intermediate oval o pinkish e N/A irregular when probe d 8 G intermediate polygon o none e N/A

G intermediate polygon o none cen perhaps when depresse probe d d 10 intermediate disc c none cen? N/A, light ring around hilum?

11 s intermediate oval o? none e N/A sharpene d point APPENDIX C: Starch Morphotypes [jm Other Notes Fissures Grooves Surface Texture Lamellae Facets Max Length Taphonomy 1 deep fissure a P a smooth, w/ 17.385 N/A over hilum and patches of fine stellate shaded fissures areas deep Y and faint a a P slightly 19.95 N/A stellate fissures rough, w/ around hilum I patches of | shaded 'areas arc over hilum j slightly 11.4 hilum area rough, ; darkened I irregular I edges

i a . straight over ;a ; smooth 12.255 iN/A : hilum

deep Y and faint a iP | slightly 14.25! fissure at stellate fissures [ rough iedge ; around hilum I stellate a j smooth, 14.25 N/A [ irregular j edges

; one long fissure : equatorial > p ! smooth 25.65 j long ; i at edge of grain groove I irregular j fissure [from j !grinding? j major deep X • a j overall 19.38' fissures so j fissures ; smooth j deep must I i surface, 1 j befrom j ledge ipressure j ; appears ! damage, {rough extinction j :cross j \ strange i APPENDIX C: Starch Morphotypes

0) re .c a a> c Q) c o £ c o o a & o a> u. '•ff ra < sl O <0 •C •C 5 !: is V) ^1 ll E i E (f) a +3C Oto = w = o 3 O 3 o w ° s 2 o I 5 X o — re s £3 a UJ O CO x a 12 G intermediate polygon o pink s.e. darkened in polarized light

13 s intermediate super o pink s.e. N/A irregular polygon

14 G intermediate polygon o none cen N/A rounded corners 15 G intermediate polygon o slit pink s.e. darkened rounded in corners polarized light 16 intermediate polygon o brown s.e. darkened in polarized light 17 intermediate polygon o brown e N/A

18 intermediate polygon o brown s.e. darkened in polarized light 19 G intermediate polygon o slit none cen slit darkened area in polarized light APPENDIX C: Starch Morphotypes

> CO w a> O) E a> « a> © k- (0 £ > n 0 c o a> 4) 3 o a> co 3 c U) o © o £ ^ (0 W E o 1 * X •C il o n ra (0 a o z 0 stellate slightly 14.535j extinction ! edge or • rough, j cross ; rim of ; irregular ! appears to grain edges ;have an appears ! extra arm pink from | damage, hilum area i enlarged ; and black Sin polarized light

X shaped fissures t a i smooth, 17.67 j N/A pink over the hilum j irregular : edge in j edges one I microgra i ph but jnot ; others X shaped fissures J a j smooth, 16.53 jX fissures ! over the hilum j irregular deep |edges 11 deep fissure j smooth, 14.25 ident on through hilum, j irregular grain fine stellate !edges surface

: X shaped fissures a rough 20.52; N/A i over the hilum, fine stellate

stellate a smooth 17.11 N/A < possible and ! corn- e shaded : hilum Y shaped fissures a 'rough 17.1! N/A ; over the hilum, j fine stellate

; 1 deep fissure rough 14.25! N/A !over the hilum, fine stellate APPENDIX C: Starch Morphotypes 0) (0 a Q) - c a) c a> £ o o a 0) f 2 o a> go u. '•M ra < 2 <5 •C H -j- £ £I a — ro c s s w si E (0 k. o o ra ^ Is II 3 ° 3 o o w .o o L. t< 2 Q i > X o — (0 S a UJ O o z a. 20 G intermediate polygon o? none cen N/A

21 s intermediate polygon o none cen N/A rounded corners

22 D intermediate gumdrop c none N/A bell, curved bowl sides, straight base 23 G intermediate polygon o none s.e. N/A rounded corners

24 s intermediate irregular o none cen N/A oval

25 intermediate sphere o none e donut slightly around irregular hilum 26 fuzzy irregular o none e darkened oval in polarized light 27 intermediate pear- o none e N/A shaped oval 28 G intermediate polygon o pink cen donut ring around hilum

29 G s intermediate polygon o none cen N/A APPENDIX C: Starch Morphotypes pm Other Notes Fissures Grooves Surface Texture Lamellae Facets Max Length Taphonomy deep X a a P rough 28.51 edges • show fissures, extinction cross irregular Y and stellate a a P smooth 17.1 N/A fissures and shaded, I irregular edges a a a perhap rough 11.4 N/A s1? edges

i Y and stellate a a P slightly 17.1 N/A fissures rough, j irregular edges ; arc over hilum a P a smooth 15.675 rough j and edges shaded ia a P a smooth 15.675 N/A j

a a a a rough 19.95 very rough j edges, hilum area j 'burnt out' j stellate fissures a a a smooth 17.1 N/A over the hilum

; stellate a a P slightly 16.53 N/A ! rough, irregular edges Y and stellate a a P smooth j 16.53 N/A | fissures and shaded, indented edge APPENDIX C: Starch Morphotypes a> a E ra Q) c k. c a> c 0) % o o a a> l. I s o a) Si Li. *s £ro = ! < jS <5 f- *1 C o — ra O O o CO I > siX O 5 I— -I o uj O X CL 30 G intermediate polygon o pink cen N/A

31 intermediate polygon o none e N/A

32 G intermediate polygon o pink cen N/A

33 intermediate oval c a little darkened irregular pink in depresse hilum d in area normal light 34 B intermediate sphere o pinkish cen N/A 35 intermediate polygon o brown e N/A rounded corners

36 intermediate polygon o none s.e. N/A rounded corners 37 intermediate polygon o none s.e. darkened rounded in corners polarized light APPENDIX C: Starch Morphotypes

> v> o> ra 8 £ o u (0 3 "5 £ (Q 3 c r-4) +«4) OT o a) o

a a : a ' smooth 6.84 N/A X over hilum a ia ' a ismooth 14.535 N/A fissures and grain [outline j appear j pink, possible corn- e hilum ; Y and stellate rough ! 21.375 N/A fissures

; X and stellate s smooth, 19.95; N/A ; fissures ; irregular ;edges APPENDIX C: Starch Morphotypes Area Starch Grains Slide # # Hilum Hilum Colour Visibility 3D Shape Location Hilum Placement Grain Form Extinction Cross Tool Type, Hilum Morphotype 38 A s fuzzy disc oval c none cen? N/A elongate

1 C 3674 cobble with s fuzzy and oval o purple cen darkened EX1 linear hardly visible symmetri pink SL2 striations, c depresse toss zone d in CF18 normal light 2 s fuzzy, barely hemisphe o? none e darkened visible rical or and vase- diffuse in shape polarized light 3 A s fuzzy disc oval c none cen? darkened in polarized light 4 s fuzzy disc c none cen darkened in polarized light from diffuse ex cross 1 A 3699 cobble with s fuzzy disc c none cen? darkened EX1 linear in SL2 striations, polarized surface light CS2/CF17/ CF18 2 G s intermediate polygon o none s.e. darkened in polarized light

1 A 3651 broken s fuzzy disc c none cen darkened EX1 maul with in SL2 flat base, polarized surface light CF9 APPENDIX C: Starch Morphotypes

> V) (0 0) 0) E 0) ra 0 m o u tf> £ > CO 0 2 a> a> 3 o © 4-» ra 3 c r" +-• (0 0 o S o (0 8 £ o 1 * 5X ! o z o co CO n 4 u. 5 |2 equatorial ; smooth 17.1 j extinction groove cross ; enlarged very diffuse, ' slightly rough ;edges a 1 ring p near a ' smooth 8.55 IN/A grain edge

p at random, not a rough, surf 39.9 j extreme on hilum appears ! damage pitted and bumpy

equatorial :p ; smooth 19.95 j extinction groove i ! cross very (diffuse

equatorial p i smooth 25.65 extinction groove j cross very diffuse

faint X over hilum equatorial ! p I smooth 29.071 extinction groove j cross very ! diffuse

Y and stellate [rough i 18.525 deep fissures j j fissures ; : |and | j fissures at ;edges ; equatorial :p ! smooth i 24.225 • N/A | groove > APPENDIX C: Starch Morphotypes 4) a uE c a> 0)re £ * a> r- O o o. i. f s o a> UL re < 3 <5 £ ig I-§1 •£ c o w x a _ (0 c w W II« E « o I. w ° 9. «= o .-= 3 o .o Q E Q X > llI O s o m a CO if intermediate polygon c?, none N/A but not really visibl e fuzzy disc none cen darkened in polarized light intermediate polygon c? green s.e. depresse hilum d? area

fuzzy disc oval c none cen N/A elongate

s intermediate polygon o faint darkened rounded brown in normal corners dot light, donut around hilum intermediate polygon o pink darkened in normal and polarized light, depresse d 8 fuzzy disc none cen N/A APPENDIX C: Starch Morphotypes

•C4-« >. (A w 0) O) £ £ >© ra 8 £ o i- w 3 o TO 3 c x:a> +*a) « ffl S| o ° in S E O 1 s X £ a il o (9 n W3 FQ) (0 a o z u. S ; stellate j cratered 17.1 N/A ; can't i and rough dislodge from dirt

equatorial p i smooth 19.95! rough groove and edges ishaded central channel j smooth 26.22 damage to: can't over hilum | and j hilum i rotate, shaded area? hilum j area appears distinctly bounded . Grain i ihas j green j I tinge """ f ' j equatorial p smooth 17.671 pits on , • groove and wavy ! surface surface:e I [smooth i 23.085j N/A [not corn, ; j donut I • j hilum ! - area

;X over hilum, fine ja j rough | 19.95 damage to ; stellate [ hilum i

[equatorial p j smooth 19.95! N/A ! groove and 'shaded APPENDIX C: Starch Morphotypes Area Starch Grains Slide # Hilum Colour # Hilum Visibility 3D Shape Location Hilum Placement Extinction Cross Grain Form Tool Type, Hilum Morphotype 9 A s fuzzy disc c none cen donut Chinese around hat hilum (uneven sides of disc)

10 s fuzzy oval 0 none e N/A elongate

11 s fuzzy oval slight c none e darkened point in polarized light 12 s fuzzy oval c none cen? darkened symmetri in c polarized light

13 A s fuzzy disc, oval 0 none cen N/A egg- shaped

1 B 9991 flake with s intermediate sphere 0 pink cen depresse EX1 acute- d? SL2 angled darkened marginal in normal unifacial and retouch polarized CS2 light 2 s intermediate cuneiform o none s.e. darkened (heart in shaped) polarized light APPENDIX C: Starch Morphotypes

> V) (0 d> 4) 0) 0) CD E L. (0 £ > JS 0 2 o a) a> 3 o 0) ra 3 c 0) O Si o B o M S E U 1 * X £ il o R> to m Q. o z S .2 equatorial ! smooth 34.2 N/A groove

transverse fissure, a ; smooth 19.95 fissures short fissure out j deep ! from hilum X over hilum a i rough 22.8 i damage to | I hilum

one irregular [smooth 14.251damage to; hilum 1 area, • appears | burnt out j I equatorial p ! smooth 25.65 i N/A i | groove j I around ! edge !a smooth 5.71 N/A

deep arced hilum | smooth 22.8(fissure is can't fissure, fissure j and wavy | deep and j rotate, in : continues at grain I surface j irregular j mountin edge 1 ;9 ! medium # Starch Grains Morphotype (/) m -± z 2 § o Slide #

i C1 D) 03 £DX- =(Qd . 3 OC (D N t- J- Tool Type, %O a CLj ®I 05 Location ro > w -o Grain Form "0 m z c g N x Extinction o Cross CO Q> o

w o o —* 3D Shape T32T O «< "O(D Hilum W Visibility 3 DO Hilum CD Colour O CD Hilum Placement ffi w 3 5' M CL = 5" Q. m" £» z sr. 3 c 3 I(3 a—• :r= ]c c^ 3 EI m n 1 (D c - 3 Hilum Area 3 x =3 D 0) I g. Q. Z 3 C 3 3 o 0) Fissures in C/5 c CD 5' CQ CD 3 "g §I Grooves CD O 3Q>' > -o •a m Lamellae z g x o » CO Facets 0)—I 3"o C/5 o 3 o Surface o Texture •

Taphonomy

Other Notes