THE APPLICATION OF STARCH GRAIN ANALYSIS TO
LATE PREHISTORIC SUBSISTENCE IN NORTHEASTERN CALIFORNIA
Gary J. Scholze
B.S., University of Wisconsin – La Crosse, 2002
THESIS
Submitted in partial satisfaction of the requirements for the degree of
MASTER OF ARTS in
ANTHROPOLOGY at
CALIFORNIA STATE UNIVERSITY, SACRAMENTO
SUMMER
2011

THE APPLICATION OF STARCH GRAIN ANALYSIS TO
LATE PREHISTORIC SUBSISTENCE IN NORTHEASTERN CALIFORNIA
A Thesis by
Gary J. Scholze
Approved by:
__________________________________, Committee Chair
Dr. Michael G. Delacorte, Ph.D.
__________________________________, Second Reader
Dr. Mark Basgall, Ph.D.
____________________________
Date ii

Student: Gary J. Scholze
I certify that this student has met the requirements for format contained in the University format manual, and that this thesis is suitable for shelving in the Library and credit is to be awarded for the thesis.
__________________________
Dr. Michael G. Delacorte, Graduate Coordinator
Department of Anthropology
___________________
Date iii

Abstract of
THE APPLICATION OF STARCH GRAIN ANALYSIS TO
LATE PREHISTORIC SUBSISTENCE IN NORTHEASTERN CALIFORNIA by
Gary J. Scholze
Ethnographic, biogeographic, and archaeological data suggest that root crops were an important part of the Late Archaic diet in northeastern California. Definitive evidence for the prehistoric existence of this pattern remains equivocal, however, requiring the need for new approaches to the problem. One of the more promising methods to accomplish this is starch-grain analysis, which has the potential to improve our understanding of root crop use and its significance with respect to broader issues of resource intensification and environmental change. This thesis examines curated collections from three different regions of northeastern California (Pit River Uplands,
Madeline Plains, Secret Valley), as they relate to the prehistoric exploitation of root crops. This required the development to a standardized morpho-metric starch grain methodology and reference collection to analyze starch grains recovered from a sample of flaked and ground stone artifacts from sites in each of the three study regions. Many plant taxa starch grains are morphologically and metrically distinct, allowing for their identification. Four plant taxa ( Perideridia sp., Lomatium sp., Triteleia sp., and Quercus sp.) and one more inclusive plant family (Apiaceae) were identified. This provides direct iv

evidence of prehistoric root and nut use that has until now formerly been largely unconfined based on inferential archaeological evidence. This demonstrates that northeastern California Native American groups made use of many types of wild root crops that were prepared with a variety of tools, depending on the resource involved. It further demonstrates that starch grain analysis has tremendous potential to improve our understanding of prehistoric diets, as they relate to often-critical food staples that leave few physical traces in the archaeological record.
_______________________, Committee Chair
Dr. Michael G. Delacorte, Ph.D.
_______________________
Date v

ACKNOWLEDGMENTS
Through this long and arduous undertaking, many people have contributed emotionally, structurally, and financially to the achievement of this thesis. This research began on a wing and a prayer and only through the support of the many people that played a role, did this idea take shape and become of substance. As such, I would like to take this opportunity to thank those individuals and organizations for their benevolence and support.
I gratefully recognize the support of the members of my thesis committee, Dr.
Michael Delacorte and Dr. Mark Basgall throughout this research. I am very thankful and indebted to Michael for all of his help not only in reviewing and commenting on the drafts of this thesis, but for years of providing instruction and priceless insight into the archaeology, ideas and changes as well as providing the encouragement to push on.
Mark has also played a crucial role in the development of this thesis by providing invaluable knowledge and guidance in building my research and helped me to develop the confidence in this new methodology and pioneering exploration. He has also aided financially through the purchasing of lab equipment, expensive heavy liquid, and access to the use of a microscope to make this research possible on a grim graduate school budget. For this, I am very appreciative and eternally grateful. Many thanks also go to
Dr. David Zeanah who encouraged me and helped me work through some of the stumbling blocks along the way. I cannot thank them enough for all their help and direction on this unknown path both in academia and in challenges of life. vi

I would also like to thank my colleague and mentor, Rob Cuthrell from UC-
Berkeley. Rob was gracious enough to take time out of his busy schedule to help refine my makeshift starch lab and help guide me through the extraction process. His advice has been priceless and invaluable throughout this long endeavor and for that, I am very appreciative. I also need to thank Eric Wohlgemuth for introducing me to Rob, his aid in plant knowledge, and the comparative sample that was donated for the study. I am also thankful for the help and assistance of Dr. Susan Gleason, who loaned me her comparative starch slides for use in the study.
Many thanks go to the BLM and the Nevada State Museum in Carson City for allowing me to gain access to the archived museum collections and borrow some of the artifacts for the study. Gene Hattori and Rachel Malloy were very gracious in their help to guide me through the loan process. Rachel also spent a few days helping me pull the artifacts for analysis and returning them to the collection, and for this, I am grateful. I would also like to thank Liz Honeysett and Tammara Norton at Far Western
Anthropological Research Group (and Bill Norton for transport) for graciously giving me copies of both the Tuscarora and Intertie project catalogs.
My appreciation goes to my many colleagues at the Archaeological Research
Center at CSU-Sacramento that have showed interest with this project, listened to me ramble on, and helped me work through many problems. Thanks so much to Bridget
Wall for helping me with the graphics and photos; Anthony Pohl for re-inking the artifacts after analysis; Michelle Noble for formatting help; and all others who have vii

listened to me discuss, go off on a tangent, complain, etc. Thank you so much for helping me labor over the many hurdles in order to finish this thesis.
I am thankful to the many that have helped me on my journey to reach this goal. I am grateful to John Hitchcock and Jeanne Goetz for igniting my interest in northern
California archaeology and encouraging me to attend graduate school. I am thankful to
Trudy Vaughn, Patrick Brunmeier, and Chris Dalu for supporting and confirming confidence in my abilities. I am grateful to Jeff, Donna, and the Burcher family for supporting me over the years through my transition to California; I would not be in
California if it were not for them. Finally, I am indebted to my parents, Gerald and
Marcia Scholze, for their continuing interest, encouragement, and curiosity in all of my academic and sometimes wild undertakings in life. I am done, Mom!
Lastly and most importantly, I need to thank my wife, Dr. Danielle Scholze, and my son, Kaleb. She has been my love and support throughout the last six years of this long and rough road to the finish line. I cannot express how grateful I am to her for moving out west, allowing me to pursue and continue my education and start our life together in California. We have struggled through many hardships together and now have conquered another. Kaleb has been my love and inspiration through this entire process, entering my life just as I began this thesis research and has been a main contributor in finishing ever since. For this, I am eternally grateful to you both, thank you, love you, and without you guys I would never have finished.
I also want to thank anyone who I missed and others who I have met along the way. To everyone I can proudly say that I am officially done. Thanks again. viii

TABLE OF CONTENTS
Page
Acknowledgments.................................................................................................................... vi
List of Tables ........................................................................................................................ xiii
List of Figures ......................................................................................................................... xv
Chapter
1. INTRODUCTION .......................................................................................................... 1
Elaboration of Introduction .......................................................................................... 5
Organization of the Study ............................................................................................. 7
2. THE BIOLOGICAL AND ARCHAEOLOGICAL BACKGROUND OF
STARCH GRAIN RESEARCH ....................................................................................... 8
The Biology of the Starch Grain .................................................................................. 8
Seasonal Variability within Plant Taxa ....................................................................... 11
Starch Research Background ..................................................................................... 11
Morphological Identification ..................................................................................... 15
Starch Grain Staining ................................................................................................. 18
Multivariate Approach ............................................................................................... 19
Conclusions ................................................................................................................ 20
3. NORTHWESTERN GREAT BASIN: A PILOT STUDY ......................................... 21
Pit River Uplands/Madeline Plains ............................................................................. 23
Secret Valley .............................................................................................................. 24
Vegetation and Plant Resources ................................................................................ 26
Animal Resources ...................................................................................................... 29 ix

Ethnographic Background ......................................................................................... 30
Achomawi ..................................................................................................... 32
Northern Paiute ............................................................................................. 35
Geophyte Ethnobotany of Surrounding Areas ........................................................... 37
Modoc and Klamath...................................................................................... 37
Shasta ............................................................................................................ 39
Atsegewi ....................................................................................................... 39
Columbian Plateau ........................................................................................ 40
Discussion of Ethnographic Background ................................................................... 41
Archaeological Background....................................................................................... 42
Northeastern California Chronology .......................................................................... 45
Early Holocene ............................................................................................. 49
Post-Mazama ................................................................................................ 50
Early Archaic ................................................................................................ 50
Middle Archaic ............................................................................................. 51
Late Archaic .................................................................................................. 52
Terminal Prehistoric ..................................................................................... 53
Pilot Study Sites ......................................................................................................... 54
Pit River Uplands .......................................................................................... 54
Madeline Plains............................................................................................. 59
Secret Valley ................................................................................................. 60
Why Focus on Northeastern California ........................................................ 64
4. METHODS AND RESULTS OF MODERN STARCH ANALYSIS ............................ 65 x

Methods ..................................................................................................................... 67
Modern Starch Collection Specimen Recovery ............................................. 67
Preparation of Modern Specimens ................................................................ 68
Characterization of Modern Starches ............................................................ 69
Results.. ...................................................................................................................... 71
Epos ( Perideridia sp.) ................................................................................... 73
Biscuitroot (
Sego Lily (
Lomatium sp.) ........................................................................... 75
Calochortus sp.) .......................................................................... 77
Brodiaea ( Triteleia sp.) ................................................................................. 80
Oak ( Quercus sp.) ......................................................................................... 82
Pinyon ( Pinus monophylla ) .......................................................................... 85
Conclusions ................................................................................................................ 86
5. METHODS AND RESULTS OF ARCHAEOLOGICAL STARCH ANALYSIS ........ 91
Sample Selection ........................................................................................................ 91
Artifact Selection ....................................................................................................... 92
Methods ..................................................................................................................... 94
Heavy Liquid Starch Extraction.................................................................... 95
Ultrasonic cleaning - Machine Method............................................ 95
Ultrasonic cleaning - Manual Method ............................................. 96
Starch Residue Concentration ....................................................................... 96
Heavy Liquid (Sodium Polytungstate) Flotation ............................. 97
Sodium Polytungstate Dilution ........................................................ 98
Microscope Analysis..................................................................................... 98 xi

Slide Mounting ................................................................................ 98
Slide Scanning ................................................................................. 99
Starch Identification ....................................................................... 100
Quantification of Results ......................................................................................... 102
Results………… ...................................................................................................... 102
Perideridia
Lomatium
Triteleia
Quercus
sp. Starch Grain Identifications ............................................... 103
sp. Starch Grain Identifications ................................................. 105
sp. Starch Grain Identifications .................................................... 106
sp. Starch Grain Identifications .................................................... 108
Apiaceae (Carrot Family) Starch Grain Identifications .............................. 109
Unknown Starch Grain Identifications ....................................................... 111
Unknown Ovoid/Triangular Ovoid Starch Grain Identifications .. 111
Unknown Lenticular Starch Grain Identifications ......................... 112
Unknown Globose/Spherical Starch Grain Identifications ............ 112
Other Unknown Starch Grain Identifications ................................ 113
Pit River Uplands Starch Grain Data ....................................................................... 114
Madeline Plains Starch Grain Data .......................................................................... 116
Secret Valley Starch Grain Data .............................................................................. 118
Tool Category Starch Grain Data ............................................................................ 120
Summary of Results ................................................................................................. 121
6. SUMMARY AND CONCLUSIONS ........................................................................... 126
Future Direction of Study ........................................................................................ 130
Appendix A. Comparative Starch Attributes ...................................................................... 133 xii

Appendix B. Archaeological Starch Table ......................................................................... 152
References…………….. ....................................................................................................... 154 xiii

LIST OF TABLES
Page
1. Table 4.1 Starch Grain Measurement Criteria .................................................... 70
2. Table 4.2 Length and Area of Starch Grains by Taxon .................................... 72
3. Table 4.3 Length and Area of Perideridia sp. Starch ........................................ 74
4. Table 4.4 Length and Area of Lomatium bicolor Starch .................................. 76
5. Table 4.5 Length and Area of Calochortus macrocarpus Starch .................... 78
6. Table 4.6 Mean Length, Aspect Ratio, and Hilum Aspect Ratio for
Analyzed Taxon .................................................................................... 79
7. Table 4.7 Length and Area of Triteleia laxa Starch .......................................... 80
8. Table 4.8 Length and Area of Quercus sp. Starch ............................................. 83
9. Table 4.9 Length and Area of Pinus monophylla Starch .................................. 85
10. Table 4.10 Summary of Diagnostic Starch Grain Traits ................................... 88
11. Table 5.1 Summary of Analyzed Tools .............................................................. 94
12. Table 5.2 Summary of Northeastern California Starch Grain Results .......... 103
13. Table 5.3 Average Size, and Aspect Ratio of Perideridia .............................. 104
14. Table 5.4 Average Size, and Aspect Ratio of Lomatium ................................ 105
15. Table 5.5 Average Size, and Aspect Ratio of Triteleia ................................... 107
16. Table 5.6 Average Size, and Aspect Ratio of Quercus ................................... 109
17. Table 5.7 Average Size, and Aspect Ratio of Apiaceae ................................. 110
18. Table 5.8 Unknown Starches .............................................................................. 112
19. Table 5.9 Unknown Starch Forms ..................................................................... 113 xiv

20. Table 5.10 Starch Grain Results by Tool Type ................................................ 120
21. Table 5.11 Number of Starch Grains per Artifact ............................................ 120
22. Table 5.12 Number of Starches by Tool Type in each Study Area ............... 122
23. Table 6.1 Flowering Periods of Storable Geophytes ....................................... 128 xv

LIST OF FIGURES
Page
1. Figure 1.1 Ethnographic Use of Roots by Northern California Groups ............ 2
2. Figure 2.1 Generalized Starch Diagram ................................................................ 9
3. Figure 3.1 Vicinity Map of Project Areas .......................................................... 22
4. Figure 3.2 Ethnographic Map of Northeastern California ................................ 31
5. Figure 3.3 Comparative Chronological Sequences for
Northeastern California ...................................................................... 48
6. Figure 3.4 Site Locations in Pit River Valley ..................................................... 55
7. Figure 3.5 Site Locations in Madeline Plains and Secret Valley ..................... 61
8. Figure 4.1 Top and Profile View of Perideridia bolanderi Starch .................. 73
9. Figure 4.2 Top and Profile View of Lomatium bicolor Starch ........................ 76
10. Figure 4.3 Profile and End View of Calochortus macrocarpus Starch .......... 78
11. Figure 4.4 Top and Profile View of Triteleia laxa Starch ................................ 82
12. Figure 4.5 Profile and End View of Quercus kelloggii Starch ......................... 84
13. Figure 4.6 Profile View of Pinus monophylla Starch ........................................ 86
14. Figure 4.7 Northeastern California Starch Grain Key .................................... 89
15. Figure 5.1 Perideridia Starch #1705-493-17 ................................................... 104
16. Figure 5.2 Lomatium Starch #3153-210-2 ........................................................ 106
17. Figure 5.3 Lomatium Starch #3153-1547-12 .................................................... 106
18. Figure 5.4 Triteleia Starch #1623-400-6 ........................................................... 107 xvi

19. Figure 5.5 Triteleia Starch #1705-493-14 ........................................................ 108
20. Figure 5.6 Quercus Starch #206-2514-19 ......................................................... 109
21. Figure 5.7 Apiaceae Starch #1705-1333-6 ....................................................... 111
22. Figure 5.8 Select Pit River Artifacts .................................................................. 115
23. Figure 5.9 Madeline Plains Artifacts ................................................................. 117
24. Figure 5.10 Select Secret Valley Artifacts ........................................................ 119
25. Figure 5.11 Frequency of Identified Starches between Study Localities ..... 124
26. Figure 6.1 Frequency of Starch Grains by Tool Category .............................. 130 xvii

1
Chapter 1
INTRODUCTION
The purpose of the present research is to explore the use of root crops among
Native California/Great Basin peoples. The development and intensification of root exploitation has been at the heart of many cultural historical and evolutionary debates over much of western North America. Archaeological identification of this activity, however, is often problematic. One of the reasons for this is the lack of root preservation and/or identification in most archaeological contexts. Some root crops (e.g., Brodiaea,
Dichlostemma, Calochortus ) required limited processing, leaving few tools or other evidence of their use. When tools were required, many were made of perishable material
(e.g., digging sticks, burden baskets) or their function was so generalized (e.g., mortar or millingstone) that it cannot be directly linked to roots.
Other root crops (e.g., Camas, Allium, Lomatium ) required construction of roasting ovens and storage facilities to exploit efficiently, but even these remains may be difficult to locate or identify archaeologically. Given these constraints, most root use has been identified or inferred through a combination of indirect indicators. These include a heavy reliance on ethnographic documentation, biogeographic and ethnobotanical data, the presence of presumed storage and/or roasting features, and purportedly distinctive flaked and ground stone signatures.
Ethnographic data indicate that geophytes (plants propagated by underground roots) were intensively exploited in ethnohistoric times throughout much of western

2
North America (Garth 1953, 1978; Kniffen 1928; Merriam 1926; Olmsted and Stewart
1978; Powers 1877; Ray 1963; Spier 1930; Voegelin 1942). This includes the interior
Columbia Plateau, Pacific Northwest, and northern Great Basin, where other abundant and storable resources such as seeds and nuts were limited. The abundance, predictability, and resilience of wild root crops have all been proposed as reasons for increased settlement and intensification of root use in these areas (Prouty 1995:9).
Baumhoff (1978:19) also states, “this reliance on root crop as a staple by the people of
Northeast California indicates their relationship with Plateau people to the north.” In
California, ethnographic sources make additional reference to root exploitation, with 85% of 73 northern California groups making some use of roots (Figure 1.1 [Scholze 2007]).
This suggests that root use may have been more important than previously thought.
Figure 1.1 – Ethnographic Use of Roots by Northern California Groups.

3
The biogeography of root-producing genera can be used to assess their potential significance among ethnographic groups. Starchy underground organs characterize members of geophyte producing plant families (e.g., Liliaceae and Apiaceae) that generally flower between May and June, but as late as August at higher elevations. The distribution of various species within these families is extremely widespread, growing in a variety of soil types, elevations, and climatic settings. Ethnobotanical data have been used to examine the association between archaeological sites and productive root procurement areas. Housley (1994) believed that fluctuations in moisture could accelerate or delay geophyte maturation, changing the availability of root crops for prehistoric peoples. When combined with ethnographic data, such biogeographic considerations may be used to predict the relative importance of specific root crops on a regional or even local scale.
Direct archaeological evidence for the importance and use of roots is more difficult to recover, but site location and assemblage content can provide important clues.
Ethnographically, root gathering and processing employed similar technologies regardless of the species involved (Holt 1946; Kelly 1932; Kniffen 1928; Ray 1963;
Spier 1930; others). This included scrapers/peelers for the skinning of roots, milling equipment (hopper mortars, millingstones, handstones, etc.) for grinding, and roasting/storage facilities for root preparation/preservation. The processing of certain plant resources requires the use of particular toolkits that develop specific use-wear patterns (Adams 1996; Kolvet and Eisele 2000; Odell and Odell-Vereecken 1980;
Vaughn 1985). Many of these artifacts have been recovered from geophyte-related sites

4 throughout western North America, and may provide a means for more reliably identifying root exploitation (Delacorte 1997a; Foster-Curley 2006; Gleason 2001;
Matson and Alexander 1987; McGuire 2000a; Prouty 1994, 1995; Thoms 1989;
Waechter and Andolina 2005).
Identification of a more reliable archaeological signature of geophyte use will better assess the significance of roots at various times and places in the past. Delacorte
(2002) provides four lines of evidence for upland root exploitation: 1) scarcity of economically important seeds/nuts in association with ground stone, 2) flaked stone tools with steep, extensively step-fractured edges used as scrapers, 3) large collections of milling equipment that lack “intentionally shaped, pecked/refurbished, or extensively curated implements,” and 4) expedient milling assemblages with numerous scavenged and re-used tools. Each of these markers may be individually important, but assume greater significance in combination.
Much the same is true for geophyte-related features that can include storage structures such as rock rings, talus pits, or underground caches and roasting pits or earth ovens (Aikens and Jenkins 1994; Delacorte 1997a, 2000, 2008; Elston 1979; Leach 1988;
Prouty 1995; Thoms 1989; Waechter and Andolina 2005). The quantity, size, and vicinity of storage and processing features in the absence of other, readily storable foods
(e.g., acorns or pine nuts) may provide at least indirect evidence for root exploitation.
The presence of rock-lined storage features and cache pits in proximity to residential settlements may provide another indicator for the heightened use of roots as resource staples during periods of limited food availability (Prouty 1995).

5
When combined, these lines of evidence provide a persuasive, though less than definitive, case for root exploitation. As such, evidence must be sought to explore the presence and importance of root procurement in the past. One of the more promising methods to accomplish this is the application of starch-grain residue analysis. This should allow for the more definitive identification of root use on the basis of incontrovertible starch-grain data. If successful, starch analysis may profoundly improve our knowledge of geophyte use, and our ability to draw inferences about broader aspects of resource intensification, other human behavior, and environmental change over time.
ELABORATION OF INTRODUCTION
Starch grain analysis has the potential to contribute to our understanding of root use and intensification, on various levels. Starch can be used to build an argument and evaluate cultural behavior at many different levels. This includes building hypotheses around: presence or absence, targeting one or more species, and the identification of most taxa (Torrence 2006:120). Each one of these levels of study has the potential to answer many types of questions. At the simplest of levels, the presence or absence of starch can support basic arguments about plant use. More specifically, starch grain variation in relation to specific tool types allows for differential plant use to be evaluated. The targeting of a few specific species (e.g., Perideridia sp ., Camassia sp ., etc.) intended for intensified use can be identified and is the premise of the current research in northeastern
California. Lastly, research involving the identification of most or all taxa can be considered in the assessment of plants within a culture and/or society. The objective of

6 the following study is to identify various geophyte starches in conjunction with a range of tool types within northeastern California (Figure 1.2).
The intention of this research is to shed light on intensified root use by using a more direct approach in which geophytes and/or plants can be observed more clearly within the archaeological record. These identified starches can be utilized in the assessment of root exploitation, giving a more direct characterization of root use within the region. Other identified/unidentified starch can also be employed in terms of quantity with the possibility of taxa elimination and/or the comparison of starches within assorted context categories (e.g., tool type, house, site). This may clarify the use of geophytes in relation to the function of specific tools, processing procedures, and areas or time periods of more intensified use. More importantly, any temporal correspondence between the root intensification package and the adoption of acorn and pinyon intensification packages in other regions may reflect a more generalized shift in resource use. The last, but most prominent objective of the current research is the building of a more universal methodology, which ultimately could be expanded and used throughout California, the
Great Basin, and far western North America.

7
ORGANIZATION OF THE STUDY
Following this brief introduction are seven chapters organized as follows:
Chapter Two focuses on the biology of starch, including its morphological variability and means for identifying different starch types. Chapter Three provides an attenuated summary of the environment and aboriginal subsistence resources in northeastern
California, as they relate to the use and intensification of geophytes. This is followed by a brief ethnographic and archaeological overview of the region, and description of the six archaeological sites employed in the study. Chapter Four reviews the methodological procedures for the establishment of a starch grain reference collection, and the techniques for identifying starch grains using a morpho-metric approach. Chapter Five discusses the sample and artifact selection for the pilot study and starch grain extraction techniques using heavy-density flotation. Finally, Chapter Six provides a summary of research results.

8
Chapter 2
THE BIOLOGICAL AND ARCHAEOLOGICAL BACKGROUND
OF STARCH GRAIN RESEARCH
Paleoethnobotanical data can be divided into three categories: macro-remains, micro-remains, and residues. Macro-remains include things like charcoal, seeds, and fibers that are generally visible with the naked eye. These can be recovered using traditional field techniques such as screening and flotation. Micro-remains, comprising things like pollen, phytoliths, and starch grains, require higher magnification and more involved techniques to recover. Finally, paleobotanical residue analysis tries to identify trace amounts of botanical material present on processing tools or incorporated directly in bone or other tissues (Gleason 2001). The current research will attempt to use both residue and micro-remain starch grain data to provide direct archaeological evidence for the use of geophytes.
THE BIOLOGY OF THE STARCH GRAIN
Starch is formed and transported throughout plants. This includes their seeds or fruits, underground storage organs or roots, leaves, and stems (Sivak and Preiss 1998:1).
Although the density of starch granules varies between plant components, they are present throughout the plant. Starch is the principal food storage mechanism of plants and is synthesized by specialized organs or plastids, which occur in individual cells.
Starch can enter the archaeological record in many ways -- as food remains, building materials, other plant products, or residues from these preserved on tools.

9
There are two types of plastids: chloroplasts, which predominate in leaves and green stems and produce mostly transient starches, and amyloplasts, which produce the starches commonly found in roots, rhizomes, tubers, and seeds (Loy 1994). The latter provide a source of energy for plants that can be broken down as required for plant growth and other activities.
Starch grains within plastids are composed of two organic polymers: amylose, a linear polymer, and amylopectin, a branched polymer. The amylose-amylopectin ratio is typically 1:4 in starch granules and affects their physical properties, such as gelatinization and reactions to stains (Gott et al. 2006). Granules are formed within plastids by building layers of amylose and amylopectin around the hilum or center point. The pattern of these polymers creates a semi-crystalline arrangement that produces a birefringence optical property and the appearance of an “extinction cross” under polarized light (Loy 1994).
This double refraction occurs through the polarization or orientation of light waves creating a distinctive white-on-black Maltese cross effect on undamaged starch granules
(Figure 2.1).
Figure 2.1 – Generalized Starch Diagram.

10
Gelatinization occurs when a starch is heated beyond a certain threshold, and the granules burst into a homogenous liquid (BeMiller and Whistler 1996; Messner 2008).
The temperature at which this occurs varies between the starches of different plant taxa, but at least some are able to survive the temperatures at which they are traditionally cooked for human consumption (Babot and Apella 2003; Chandler-Ezell et al. 2006;
Messner and Schindler 2010).
There are two types of starch granules: simple and compound. Simple grains have only one component, while compound granules are comprised of a few grains within a larger starchy layer (Torrence and Barton 2006:41). The type, size, and shape of granules vary and can be used to identify different starches. Size can range from 1 to 100 microns and shape can include disc, spherical, oval, elongate, rounded, kidney-shaped, polyhedral, and other forms. Other characteristics of starches that are critical to the methodology described below include the appearance and location of the hilum (point at which the “Maltese” or polarization cross intersects [Figure 2.1]), the appearance of lamellae (centric growth rings encircling the hilum), an open or closed hilum, and the presence and location of fissures.
Most of the starch within plants is stored in underground storage organs, seeds/fruit, or above ground stems. Roots and tubers have exceptionally high densities of starch. This allows for their winter survival and subsequent growth the following spring.
Seeds also have a high starch content to facilitate reproductive growth. As such, seed and root starches are especially useful for exploring past subsistence patterns.

11
Stone tools and other artifacts can operate as “starch traps,” protecting starch from decay by microorganisms so that it may be preserved for long periods of time (Fullagar
2006:177). This facilitates starch analysis in prehistoric contexts in an effort to interpret artifact function and reconstruct prehistoric diet.
SEASONAL VARIABILITY WITHIN PLANT TAXA
Environmental conditions can affect the amount and morphology of starch granules. When stressed, plants develop fewer and smaller starch grains (Gott et al.
2006:42; Messner 2008:82). This includes seasonal variations in carbohydrate production and storage, with various studies showing that some starches accrue primarily during the summer and fall months (Brocklebank and Hendry 1989; Landhausser and
Lieffers 2003; Messner 2008:82). This is attributed to the physiological storage of energy prior to winter dormancy for the subsequent initiation of spring flowering and seed production.
STARCH RESEARCH BACKGROUND
Starch grain analysis has become increasingly important over the last two decades. Data from this research have been used to evaluate plant domestication, prehistoric diet, mobility patterns, artifact function, and more (Loy et al. 1992; Piperno et al. 2000; Piperno and Holst 1998; Van Peer et al. 2003; Williamson 2004). This includes studies from both open-air and protected rockshelter sites that date from ancient to modern times (Matthews and Torrence 2006:19). When combined with other forms of

12 analysis (e.g., lithic, phytolith, ground stone, and pollen studies), starch grain studies have been instrumental in addressing a variety of topics surrounding plant/human interactions and evolutionary trajectories in numerous cultural contexts.
Starch grain analysis on stone tools is a comparatively recent development, with most of the previous research conducted on artifacts from sites with exceptionally good preservation (Fullagar 2006; Loy 1994; Shafer and Holloway 1979) or in conjunction with microscopic use-wear studies (Anderson 1980; Fullagar 1986; Kamminga 1979;
Keeley 1980). Donald Ugent (Ugent et al. 1981, 1982, 1984, 1986) brought starch grain analysis to the attention of archaeology by employing it to assist in the identification of archaeobotanical macro-remains. His research was also instrumental in demonstrating the longevity of starch preservation and the identification of various attributes. From this, a methodological approach to starch-grain analysis has been developed and improved by various archaeobotanists (Cortella and Pochettino 1994; Loy 1994; Piperno and Holst 1998; Torrence and Barton 2006). Briuer (1976) was among the earliest researchers to describe plant and animal residues from rockshelters in Arizona using iodine-potassium-iodide to stain starch granules. Shortly thereafter, Shafer and Holloway
(1979) analyzed stone artifacts with little edge damage from Hinds Cave, Texas. Still later, Loy (1993, 1994) and Fullagar (1986) reported starch granules on various stone artifacts from Australia and Papua New Guinea (see also Beck and Fullagar 1985;
Fullagar 1986, 1988, 1989, 1992). In recent years, starch analysis has been employed in conjunction with use-wear studies on numerous types of material in Australia and the far
Pacific (Fullagar 2006; see Torrence and Barton 2006). Each of these approaches has

13 explored different types of issues surrounding the interaction of prehistoric peoples and plants. The contexts in which starch residues are recovered allow for often-direct association between certain plant taxa and the processing and other cultural activities responsible for their entering the archaeological record.
As work has progressed, starch analysis has become increasingly refined and employed to identify specific plant taxa and types of plant processing. In fact, Loy
(1994) argued that specific tuber taxa could be identified and/or eliminated on the basis of granule morphology and evaluation of raphides. Other researchers have expanded on this, examining ground stone and other artifacts for starch residues (Field and Fullagar
1998; Fullagar 2006; Pearsall et al. 2004; Piperno and Holst 1998; Piperno et al. 2000,
2004). Starch analysis has also provided insights on many issues of plant domestication
(Chandler-Ezell et al. 2006; Dickau 2005; Horrocks and Nunn 2007; Perry et al. 2007;
Piperno 2006) and the domestication process (Piperno and Holst 2004).
More recently, starch-grain research has expanded its focus to include past plant use. Use of various tools for plant processing can drive starch residues into cracks and fissures within the rock (Atchison and Fullagar 1998; Barton et al. 1998; Field and
Fullagar 1998; Piperno et al. 2004). Loy (1994; Loy et al. 1992) pioneered the study of different taro root taxa residues on stone tools, revealing the early transport and management of plants that may have contributed to domestication. Similarly, Crowther
(2005) extracted starch from pottery sherds to detect the use of cultivated taro in the
Pacific. Horrocks and Bedford (2005) expanded on this idea by scraping the sediment and dust from pottery and performing heavy liquid density separation. Their results

14 indicate evidence of gelatinization and the likelihood that starchy food was prepared in pots. They also revealed the presence of an imported starch, dispelling the controversy over whether Pacific Island colonizers brought agricultural practices with them or were originally foragers (Fullagar 2006). Elsewhere, studies in Panama (Piperno and Holst
1998) and Israel (Piperno et al. 2004) have questioned the nature and timing of plant domestication (Fullagar 2006). In colder climates of the Canadian Great Plains, Zarillo and Kooyman (2006), were able to connect certain ethnographic tool types to the recovery of maize and berry starch grains on ground stone tools.
Other avenues of research include expanded analysis of museum curated specimens in an effort to connect ethnographical accounts to archaeological processing tools based on both use-wear and the extracted starch residue (Barton 2007; Fullagar et al. 1999; Scholze 2010). Lastly, starch-grain residue has expanded to the analysis of skeletal remains and preservation of starch within the dental calculus of teeth (Boyadjian et al. 2007; Henry and Piperno 2008). These later studies can be used to assess whether groups or individuals had starch-rich diets, with the condition of the starch providing information on food processing (Messner 2008:77).
More recently, in an effort to better understand prehistoric use of specific plant taxa, Rumold (2010) used starch-grain analysis to identify various species of potato and chili pepper in association with domestic varieties in Peru. Messner (2008) analyzed seed crops at sites in the eastern United States, while Messner and Schindler (2010) evaluated heat treatment and the effects of processing on starch granules.

15
In California, Gleason (2001) experimented with starch research by starting a comparative collection of species within the Klamath River Valley, where she determined there were differences in starch grains between various plant species. Rob Cuthrell
(personal communication) is also conducting ongoing research along the central
California coast through the University of California, Berkeley. Apart from these pioneering studies, however, starch research has played only a minor role in California archaeology.
MORPHOLOGICAL IDENTIFICATION
The taxonomic utility of starch grains in identifying plants has long been known to researchers (Reichert 1913; Ugent et al. 1981, 1982). Many starch grain characteristics are useful for determining taxonomy. These morphological parts include the hilum, lamellae, fissures, and bifringent properties (Figure 2.1). These traits are illustrated in greater detail in several publications (Loy 1994; Messner 2008; Piperno 2006; Piperno and Holst 1998; Torrence and Barton 2006).
The central point or core on which the granule grows is referred to as the hilum
(Piperno and Holst 1998). Its position and attributes are some of the key identification characteristics. The position of the hilum can range from a centric or centrally located point, to an asymmetrical position known as eccentric . Starches formed in storage organs typically have an eccentrically positioned hilum (Messner 2008), which may help to identify the use of certain plant organs.

16
Lamellae generally appear as concentric, or growth-rings, around the hilum of the starch granule. They are not visible in all species and, depending on their appearance and frequency can be distinctive taxonomic features (Messner 2008; Piperno and Holst 1998).
Fissures are lines or cracks that emanate from the hilum outward to the edge of the starch grain (Cortella and Pochettino 1994). These lines can also be morphologically distinctive with their shape and patterning described as stellate , longitudinal , initial , transverse , and tripartite (Messner 2008).
Starch can be identified on the basis of the previous features, but the most distinctive trait is the appearance of bifringence or the “Maltese” cross pattern that occurs when the starch granule is observed through cross polarizing light. Each arm of the extinction cross traverses the hilum, allowing for locational determination when it is not visible. The arm shape replicates granule topography, allowing for surface texture that is sometimes difficult to see to be better detected.
Starch can form in different ways. A simple grain is the creation of a single starch grain within a cell. These are usually cone-shaped and are very common in tuber crops including white potato ( Solanum tuberosum ). A compound grain is made up of several individual granules, sometimes hundreds or thousands, each of which have separate hilums and bifringence (Cortella and Pochettino 1994:172, Messner 2008:64).
Examples of taxa with compound granules are cassava, sweet potato, Chenopodiaceae, rice, and other important grasses. When compound granules are milled, they can break up into sub-granules that sometimes create a simple looking grain (Gott et al. 2006:41).

17
In strict morphological terms, starch grains are recorded using a combination of size and physical characteristics, and compared against a taxonomic key developed from a starch reference collection. Maximum length, width, and three-dimensional attributes are recorded at corresponding magnification for comparison. To accurately record the shape of the starch granule, it needs to be rotated, by lightly pressing on the coverslip, causing the grain to slowly rotate within the mounting medium (Torrence and Therin
2006). Besides size and/or shape, other morphological characteristics can be added for improved identification. These include surface features, color, hilum position, hilum features, and the number and nature of granule facets. A more precise level of taxonomic identification can be made with a larger number of characteristics (Barton 2005; Loy et al. 1992; Piperno et al. 2000, 2004). Piperno and Holst (2004) and Barton (2005) found size to be a very important criterion, followed by shape. Overall, a combination of characteristics and their occurrences is used to generate a diagnostic signature for particular genera (Torrence 2006).
Other methodological approaches to starch identification involve both chemical and physical property tests. These include gelatinization and iodine staining. The former involves the application of heat to the insoluble starch granules. At a certain threshold, the grain will burst with a corresponding loss of bifringence, change in granule shape, and the formation of a “jelly-like” substance (Banks and Greenwood 1975:260; Galliard and Bowler 1987; Messner 2008:68). These transformations occur at distinctive temperatures and times with different starches, but sometimes vary between researchers for the same taxa (Gott at al. 2006:44). Some starches are more resilient to heat, such

18 that physio-chemical alteration may not occur until a higher temperature is reached
(Babot and Apella 2003; Chandler-Ezell et al. 2006; Messner 2008). Given the many variables involved in the gelatinizational process, this identification tool needs to be further explored before it can be relied upon.
STARCH GRAIN STAINING
Iodine staining is sometimes employed to help identify some of the morphological features of starch. Starch turns a deep blue to purple color when stained with an iodinepotassium iodide. This form of identification can efficiently help to determine where starch is preserved among the artifacts analyzed. Other identification methods include gelatinization and x-ray diffraction, both of which are beyond the scope of the present study. These methods work best on artifacts that have not been heated such as scrapers/peelers and millingstones used prior to cooking, where more undamaged starch granules are expected (Loy 1994). In short, physical and chemical tests of starch properties can be useful, but in archaeological contexts of the kind explored here they have less utility.
As mentioned, when starch is heated the bifringence cross is partially or completely lost (Loy 1994; Hall et al. 1989). With moisture and/or an elevated temperature (>30

C), the hydrogen bonds linking the amylose chains are slowly broken allowing the granule to absorb water and begin to swell, a process known as gelatinization. Cooked and processed starch is typically identified by differences in granule morphology and weakening cross development. Cooking also disturbs the starch

19 structure, which loses the extinction cross entirely, and makes the granule microscopically unidentifiable (Lamb and Loy 2005:14-34).
Where gelatinization may have damaged starch granules, Congo red dye staining may be used. Congo red has been used to identify starch grains damaged by cooking or milling activities in the past (Lamb and Loy 2005). Unaltered starch grains are hydrophobic and do not take up the stain, but with the loss of the regular and compact arrangement of starch layers and absorption of water caused by cooking, Congo red dye reacts with the amylose content of such grains (Lamb and Loy 2005). Processing (e.g., shredding, milling, pounding) of starchy plants can also damage the structure of the component grains, allowing them to absorb water and hence stain (Lamb 2003). Thus, staining may help to identify the presence of culturally modified or damaged starches that are unrecognizable using conventional techniques.
MULTIVARIATE APPROACH
As useful as traditional morphological identification can be, a multivariate statistical approach to starch identification using a range of morpho-metric variables can be more productive. This type of analysis uses multiple attributes, including both discrete, and continuous variables, to identify starches (see Torrence et al. 2004, Wilson et al. 2010). Loy et al. (1992) were the first to use a multivariate approach to discriminate various plant species in Southwest Asia and the Pacific region (Torrence
2006). This approach can determine whether an assortment of taxa can be discriminated on the basis of morphological characteristics, and which are the most distinctive

20 characteristics (Torrence et al. 2004).
With the use of appropriate statistics, the probability of failure in distinguishing between two plant types is consistently low and even more so when evaluating plant families, not species (Torrence and Conway 2006).
These types of morpho-metric data can also be used to perform cluster, principal component, and discriminate analysis to assess the importance of different variables in identifying the starches in question. This classification is also of wider interest to researchers who may be unfamiliar with various morphological taxonomic keys.
CONCLUSIONS
In sum, starch grain analysis offers one of the best approaches to improve our understanding of the relationship between the archaeological and ethnographic record of prehistoric plant use. By combining the study of tools and starch residues it is also possible to improve our understanding of various plant preparation or cultural processing activities. As a result of various studies and techniques employed in different regions, starch-grain research has grown in importance and applicability to an ever increasing range of archaeological questions. Among these advancements have been many important contributions to the study of tool function and plant use. The same should be possible for California and the Great Basin, where little or no starch-grain research has occurred thus far. With this in mind, a pilot starch grain study of flaked and ground tools from the Pit River Uplands in the northwestern Great Basin was conducted, as described below.

21
Chapter 3
NORTHWESTERN GREAT BASIN: A PILOT STUDY
The use of geophytes in the northwestern Great Basin and northeastern California was of generally greater importance than surrounding areas, given the dearth of acorns, pine nuts, and other food staples. Evidence for this can be seen in the ethnographic importance of three root crops that were exploited by anywhere from half to threequarters of the local groups: Camassia – 71%, Perideridia – 57%, and Brodiaea – 42%
(Scholze 2007). This implies that roots were the single most important staple that could be gathered and stored in sufficient quantities for winter use.
Research into the intensification and use of roots or geophytes has become a major focus of research in the northwestern Great Basin and surrounding areas, where the distribution and availability of roots have been used to explain changes in settlement, land-use, and other adaptive behavior (Foster-Curley 2006; Gleason 2001; Leach 1988;
Prouty 1995). In the Upper Klamath River region, Gleason (2001) employed starch grain analysis to assess the aboriginal use and management of geophytes. She concluded that starch grains “appear to hold the highest potential of all methods so far examined for uncovering evidence of past geophyte use” (Gleason 2001:524).
The following research focuses on the extraction of starch-grains from various tools that are thought to have been for root/tuber processing in the past. These include flake tools, formed flake tools, reworked projectile points, bifaces, pestles, mortars, millingstones, and handstones.

The artifacts analyzed derive from two linear archaeological projects in northeastern California, and more specifically from six prehistoric sites in three environmentally distinct sub-regions: the Pit-River Uplands, the Madeline Plains, and
Secret Valley (Figure 3.1). Each of these areas is discussed in greater detail below.
Figure 3.1 – Vicinity Map of Project Areas.
22

23
PIT RIVER UPLANDS/MADELINE PLAINS
The Pit River uplands lie adjacent to the South Fork Valley of the Pit River, between the towns of Alturas and Likely, California. This area is part of the Modoc
Plateau, a broad volcanic tableland that extends from Medicine Lake in the north to the
Madeline Plains in the south, and from the Warner Mountains on the east to the Cascade
Range on the west. The geology and geomorphology of this area consists primarily of a broad volcanic tableland divided by separate volcanic cones, rim rock benches, and deep canyons. These isolated elements of the landscape typically run along numerous north to northwest trending faults, seen in the many steep-sided escarpments and ravines throughout the area.
The predominate rock throughout this portion of the Modoc Plateau is comprised of Miocene age basalt and basalt andesite derived from various lava flows. Atop this lies a Pliocene age pyroclastic and rhyolitic rock comprised of lacustrine, ash, and mudflow deposits (Harden 1998; Macdonald 1965). The South Fork Valley bottom has a relatively level to gently rolling terrain with an average elevation of 4,300 ft (1,310 m) (Delacorte
1997a). The South Fork of the Pit River courses north, joining the North Fork just west of Alturas from whence their combined flow drains west and south into the upper
Sacramento River.
Immediately south of South Fork Valley lies a featureless Pleistocene lake bottom known as the Madeline Plains. It is climatically similar to the Modoc Uplands, with an average elevation of 5,300 ft (1,615 m). Pluvial Lake Madeline drained south into Secret
Valley and thence Honey Lake and larger Lake Lahontan system, with the water in Lake

24
Madeline drying sometime around 23,400 B.P. (Delacorte 1997b; Mifflin and Wheat
1979).
Climatic conditions throughout the Modoc Plateau are characterized by cold, wet winters and warm, dry summers with most of the annual precipitation (25-35 cm) falling as snow in winter months. Most of this precipitation is absorbed along the plateau margins, with much of the run-off that remains draining into the Pit River and Madeline
Plains. This creates seasonally wet conditions in many low-lying areas. This has undoubtedly influenced the location of prehistoric settlements since the middle Holocene
(Delacorte 1997b), when essentially modern climatic and biotic conditions were established (Anderson 1990; Foster-Curley 2006).
SECRET VALLEY
Secret Valley lies along the northwestern edge of the Lahontan Basin, which was part of the larger Lake Lohontan system during the Pleistocene. Remnants of this one vast lake system survive as present day Honey and Pyramid lakes. Although not technically part of Pleistocene Lake Lahontan, Secret Valley formed part of an earlier
Miocene/Pliocene lake basin that drained into Honey Lake.
Geologically, Secret Valley is bordered on the north and south by Quaternary and
Tertiary-age basalt uplands. The valley floor consists of Plio/Pleistocene gravels and lake sediments, capped by more recent Quaternary alluvium and playa sediments
(Jennings 1977; Young 1997).

25
The northern end of the valley consists of dissected delta/lacustrine deposit referred to as the French Ranch Alluvium (Young 1997) that predates human occupation of the region. The French Ranch Alluvium was deposited in a Pleistocene deltaic or slow-water environment that was subsequently eroded with the Holocene lowering of the
Lahontan Basin. As a result, a layer of basalt cobbles was deposited on top of the French
Ranch Alluvium that contains most of the local archaeological remains.
Secret Creek flows along the western side of Secret Valley atop the French Ranch
Alluvium, dissecting two sediments that are more recent. These consist of a lower
Mazama ash deposit dating to ca. 7,600 B.P. (Zdanowicz et al. 1999) and an upper
Holocene alluvium, comprising a dark marsh and alluvial sediment. These deposits indicate that Secret Creek has been in its present location throughout the Holocene, creating a relatively stable marsh environment (Young 1997).
Climatic conditions in Secret Valley differ from those at higher elevations to the north (i.e., Pit River Uplands/Madeline Plains). The valley ranges in elevation from
5,250 ft (1600 m) in the north to 4,400 ft (1341 m) in the south at Mud Lake.
Winter temperatures are correspondingly warmer than the Madeline Plains, but hotter summer months little different.
Pollen studies of Salt Creek Marsh at the northern end of Secret Valley indicate the presence of a marsh over the last 9500 years (West 1997). Although aquatic taxa shift somewhat, the surrounding sagebrush scrub community persisted throughout the
Holocene, with juniper becoming more prevalent after the 7627 ± 150 cal yr B.P. deposition of the Mazama tephra.

26
VEGETATION AND PLANT RESOURCES
Secret Valley supports a modified association of the sagebrush-steppe vegetation zone (Young et al. 1988:765). It is drier than its northern variant, supporting a sagebrush/needlegrass plant community. Dominant species in this habitat include sagebrush ( Artemisia tridentata ), rabbitbrush ( Chrysothamnus viscidiflorus ), cottonthorn
( Tetradyma canescens ), and mormon tea ( Ephedra viridus ). Annual grasses include needlegrass ( Acnatherum sp.), melic ( Melica bulbora ), fescue ( Festuca idahoensis ), ricegrass ( Acnatherum hymenoides ), bluegrass ( Poa sp .
), and many others (Wohlgemuth
1997; Young et al. 1988). In addition to these dominant taxa are juniper ( Juniperus occidentalis ), willow ( Salix sp .
), cottonwood ( Populus balsamifera ssp . Trichocarpa) , cattail ( Typha sp .
), rush ( Juncus sp .
), wild rye ( Leymus cinereus ), sunflower ( Helianthus sp .
), and camas ( Camassia quamash ) that grow in wetter areas of the basin.
Plant food resources used by local Honey Lake Paiute and Maidu peoples (Evans
1978; Fowler 1986; Riddell 1960a) can be categorized into seeds and nuts, berries, root crops, and greens (Wohlgemuth 1997).
Important seed crops included blazing star ( Mentzelia sp .
), tansy mustard
( Descurainia sp .
), sea blite ( Sueda depressa ), and sunflower. Berries that were collected and consumed consisted of currant and gooseberry ( Ribes sp .
), raspberry ( Rubus sp .
), elderberry ( Sambucus sp .
), chokecherry ( Prunus virginiana ), desert peach ( Prunus andersonii ), serviceberry ( Amelanchier sp .
), and manzanita ( Arctostaphylos sp .
). None of these berries occurs in Secret Valley proper except for desert peach and currants, but most are found in the nearby Sierra Nevada.

27
Root crops available in Secret Valley included cattail, brodiaea ( Dichelostemma multiflora ), epos ( Perideridia gardineri ), sego lily ( Calochortus sp .
), and camas. Leafy greens available within the valley consisted of miner’s lettuce (
Claytonia sp.) and, no doubt, other taxa that went unreported in the ethnographic literature (Wohlgemuth 1997).
Overall, there are relatively few plant resources in Secret Valley. Seed crops were mainly restricted to the open valley bottom and riparian corridors, and root crops limited to the riparian zone along the north end of the valley. Plants growing outside of Secret
Valley were also used, but most likely on a seasonal or logistical basis.
The Madeline Plains and Pit River Uplands support sagebrush-steppe vegetation, or more specifically a sagebrush/wheatgrass community (Sawyer et al. 2009; Young et al.
1988). Riverine/riparian communities in better-watered areas and historic expansion of the juniper woodland community at higher elevations break this otherwise monotonous flora, but may have been less prevalent in the past. The sagebrush-steppe floristic province is characterized by big and low sagebrush ( Artemisia tridentata and A. arbuscula ), bitterbrush ( Purshia tridentata ), rabbitbrush ( Chrysothamnus viscidiflorus ), green ephedra ( Ephedra viridus ), serviceberry ( Amelanchier sp .
), and currant ( Ribes sp .
).
In the local or northern variant of this community, wheatgrass ( Pseudoroegneria spicata ) replaces the needlegrass ( Acnatherum sp .
) found farther to the south in Secret Valley.
Other annuals and/or grasses of significance include Idaho fescue ( Festuca idahoensis ), bluegrass ( Poa sp .
), squirreltail ( Sitanion hystrix ), ricegrass ( Acnatherum hymenoides ), buttercup ( Ranunculus sp .
), biscuitroot ( Lomatium sp .
), balsamroot ( Balsamorhiza sp .
), mules ears ( Wyethia mollis ), and aster ( Aster sp .
). As mentioned, many of the native

28 grasses and other indigenous flora have been historically replaced by cheatgrass ( Bromus tectorum ), and recent expansion of western juniper ( Juniperus occidentalis ) (Sawyer et al. 2009).
Riverine/riparian communities along the Pit River and wetter areas of the
Madeline Plains comprise an array of plants. These include bulrush ( Scirpus sp.), cattail
( Typha sp.), willow ( Salix sp.), and many types of rushes ( Juncus sp.) and sedges ( Carex sp.). In drier areas of these communities are also wild rose ( Rosa sp.), thistle ( Cirsium sp.), mullein ( Verbascum sp.), and mustard ( Brassica sp.). Other grasses in this community consist of bentgrass ( Agrostis sp.), peppergrass ( Lepidium sp.), hairgrass
( Deschampsia caespitosa ), and wild rye ( Leymus cinereus ) (Delacorte 1997a, Delacorte and Young 2000).
Many plants of the Pit River Uplands and Madeline Plains were of special economic importance, both as food and materials for basketry, shelter, and other purposes. Although seeds were important food, root or geophyte crops may have been more significant, given their abundance and storability. Important root crops included yampa or epos ( Perideridia sp.), biscuit root ( Lomatium sp.), camas ( Camassia sp.), sego or Mariposa lily ( Calochortus sp.), brodiaea ( Brodiaea sp.), wild onion ( Allium sp.), and wild hyacinth or blue dicks ( Dichlostemma sp.). Although all of these may be locally found, only epos, biscuitroot, and camas were of sufficient abundance to be important subsistence resources that were easily collected, dried, and stored for winter use.

29
ANIMAL RESOURCES
Given the relatively uniform sagebrush/grassland community, wildlife is reasonably consistent throughout the area. Large game includes antelope ( Antilocapra americana ) of the open valley bottoms and mule deer ( Odocoileus hemionus ) found in brushy and wooded riverine and foothill habitats. Other large game and carnivores are black bear ( Ursus americanus ), mountain lion ( Felis concolor ), coyote ( Canis latrans ), badger ( Taxidaea taxus ), and bobcat ( Lynx rufus ). Wolves ( Canis lupus ) and grizzly bear
( Ursus arctoshorribilis ) may also have been present in the region, along with bison
( Bison bison
) that are thought to have entered the area from time to time (O’Connell
1975; O’Connell and Hayward 1972). Smaller mammals of significance consist of jackrabbits ( Lepus californicus ), cottontails ( Sylvilagus nuttalli ), marmots ( Marmota flaviventris ), and various squirrels (Sciuriadae), mice (Heteromyidae), and rats
(Cricetidae).
Aquatic animals exploited by local Native Americans included resident and migratory waterfowl and fish found in the Pit River and seasonally flooded areas of the
Madeline Plains. These included various ducks ( Anas sp.), Canada geese ( Branta
Canadensis ), sandhill cranes ( Grus canadensis ), and an array of other birds and fish (e.g., sucker [ Catostomus sp.], squawfish [ Ptchocheilus grandis ], and trout [ Salmo sp.]).
In sum, the study area supported a diverse variety of plant and animal resources.
Some of these, such as roots, could be gathered and stored in bulk and might be expected to influence broader adaptive patterns. Secret Valley has typically shorter, less harsh winters and may have required less food storage, but people of the Pit River Uplands and

30
Madeline Plains would have certainly depended heavily on stored roots to survive the lengthier winters there.
ETHNOGRAPHIC BACKGROUND
As previously discussed, the study area can be divided into three regions: Pit
River Uplands, Madeline Plains, and Secret Valley. Ethnographically, the Astariwawi,
Kosale’ktawi, and Hammawi bands of the Pit River tribe inhabited the Pit River Uplands and portions of the Madeline Plains. The Wadatkuht and possibly Kamodokado groups of the Northern Paiute occupied Secret Valley, the Honey Lake Basin to the south, and shared use of the Madeline Plains with the Achomawi Pit River bands. Mountain Maidu of the Sierra Nevada may have made use of the Honey Lake Basin and Secret Valley
(Simmons et al. 1997; Waechter 2000), and will be touched upon briefly, but not considered in depth (Figure 3.2).
Ethnographic work in the Pit River Uplands/Madeline Plains has been limited and extrapolated mainly from the primary work of Dixon (1905, 1908), Merriam (1926),
Kniffen (1928), Voegelin (1942, 1974), and Garth (1953) in adjacent areas. More recent summaries of ethnographic data are also available in the work of Fowler and Liljeblad
(1986), McCarthy (1997a, 1997b), Olmsted and Stewart (1978), Waechter (2000b), and
Wheeler-Voegelin (1974). Pit River territory extended from Mount Shasta and Lassen
Peak on the west to the Warner Mountains and Eagle Lake on the east, harboring eleven different bands. Of the 11 Pit River bands, five western groups had access to both acorns and salmon and the six eastern groups, including those of the study area, did not

(McCarthy 1997a, 1997b). As such, other storable food staples (e.g., geophytes) were required by eastern Pit River peoples in place of acorns and salmon.
Figure 3.2 – Ethnographic Map of Northeastern California
(adapted from Delacorte 1997a).
31

32
Achomawi
Occupation of the Pit River and adjacent uplands has been attributed to the
Astariwawi, Kosale’ktawi, and Hammawi bands of the Achomawi (Kniffen 1928). The
Astariwawi are the northernmost band of the Achomawi and immediate southern neighbors of the Modoc, but their territory does not extend into the research area.
Kosale’ktawi territory centered around the town of Alturas, where the north and south forks of the Pit River converge, creating important marshland habitats (Kniffen 1928:306;
Merriam 1926). Merriam and Talbot (1974) describe the Kosale’ktawi boundaries as:
The northern boundary extends easterly from Big Sage Reservoir to Cedar
Mountain …… the southern boundary is a straight line from Warren Peak to Signal Butte on South Fork Pit River and continues westerly for 10 to
12 miles; the western boundary, apparently, is a north-south line from Big
Sage Reservoir southward to Signal Butte (Merriam and Talbot 1974:6).
The main Kosale’ktawi village was Kosale’kta or “sidehill where the junipers grow” (Kniffen 1928:307). Two other villages were also identified, one approximately three miles southwest of Alturas along the Pit River and another about 10 miles west of
Alturas near Hot Creek (Merriam 1926; Foster-Curley 2006:20). A third village is described by Merriam at Pine Creek, but was never mentioned by Kniffen. The
Kosale’ktawi was the smallest of the Achomawi bands, with a population estimated at
125 people distributed among the three villages prior to European arrival (Kniffen 1928;
Foster-Curley 2006).
Hammawi territory is located south of both the Astariwawi and Kosale’ktawi, their settlements centering around the present town of Likely at the southern end of South
Fork Valley (Kniffen 1928:304; Merriam 1926). Merriam and Talbot (1974) describe the

33 northern boundary running from Schaeffer Mountain to Warren Peak, the western boundary running from Schaeffer Mountain south to the east side of Grasshopper Valley, and the eastern boundary following the Warner Mountains from Warren Peak south to
McDonald Peak and down to the end of the Fredonyer Mountains at the southern end of the Madeline Plains (Merriam and Talbot 1974:4-5). The southern boundary of
Hammawi territory is identified for each of the ethnographers, but some of the boundaries are perceived as overlapping and may be part of a shared use area based on fishing
(Kniffen 1928; McCarthy 1997b). The main village was Tulu’kapi or “end of the sack,” named for the enclosure of mountains around the end of South Fork Valley (Kniffen
1928). The Hammawi were the largest of the three Achomawi bands, with a total of 250 inhabitants in nine different villages (Kniffen 1928; McCarthy 1997b). When combined with other Achomawi bands, the regional population density is approximately one person per three square miles (Cook 1943:176; McCarthy 1997b).
Winter villages were comprised of large, semi-subterranean circular houses; the floors excavated three to five feet below the ground surface. The superstructure was supported by a center pole and earth covered, and entered and exited through the roof.
Houses of this type sheltered more than one family (Voegelin 1942:64-66). Winter settlements were abandoned during summer months for gathering camps, where shelter was supplied by temporary brush windbreak (Kroeber 1925:311).
As both Kosale’ktawi and Hammawi bands lived east of the distribution of oaks and salmon runs, subsistence centered on other aquatic and terrestrial foods. Aquatic resources focused on local stream fish including “pike” (pike minnow or squawfish),

34 suckers, trout, lamprey, and other indigenous species (Kniffen 1928; McCarthy 1997b;
Olmsted and Stewart 1978). Nearby marshes contained numerous species of oftenabundant resident and migratory waterfowl (Olmsted and Stewart 1978:226). Marsh areas were also prized for their large quantities of edible roots and tubers, as well as tules that provided raw material woven into various mats, clothing, rope, and other textiles
(Kniffen 1928:301, Olmsted and Stewart 1978:227). Deer were hunted in the Warner
Mountains and regions north of the Pit River (Kniffen 1928:305), with communal hunts in the fall attracting sometimes distant groups, who exchanged items not found in the region (e.g., dried salmon, acorns) (Kniffen 1928:305).
Pit River women gathered many types of plant resources, beginning with the collection of roots in the spring. Hammawi women gathered epos on the mesa above the village of Tulukupi (just southwest of Likely) and on the borders of the Madeline Plains
(Kniffen 1928:305). Epos was both eaten fresh and dried in the sun and stored for the following winter. Camas roots, other bulbs, tule, and “tule potatoes” were gathered from the marshy lowlands, and later in the season salmon berries, bear berries, juniper berries, wild plum, and wild buckwheat were also available in higher areas of the region (Kniffen
1928:305). Many of these resources were also dried and ground for winter storage
(Kniffen 1928:305). Men typically fished in the winter, spring, and summer, reserving the fall for deer hunting, although some hunting occurred throughout the year (Kniffen
1928:305-306).

35
Northern Paiute
Unlike their Achomawi neighbors to the north, the Northern Paiute were comprised of smaller, semi-nomadic groups of hunter-gatherers. Groups consisted of typically two to three related families forming a “camp group,” that traveled together to exploit various seasonal resources (Fowler and Liljeblad 1986:447-448). The Honey
Lake Paiute or Wadatkuht (“Wada eaters”) inhabited much of the Honey Lake Basin,
Secret Valley, and the eastern half of the Madeline Plains. A closely related group,
Tasiget (“Middle Place”) (Stewart 1941) occupied the eastern part of the Honey Lake
Basin and perhaps the Fort Sage Mountains to the south (Waechter 2000b), but not the
Secret Valley region of the study area. Just south of Secret Valley, Riddell (1960b) identified five Wadatkuht encampments and two cave sites: Tommy Tucker (CA-LAS-1) and Amadee (CA-LAS-90) caves.
Both the Honey Lake and other Paiute groups tended to exploit local subsistence resources, giving rise to names like wada eater, trout eater, cattail eater, etc. Settlement and subsistence patterns were adjusted, therefore, to annual and seasonal fluctuations in resource distribution, along with those across the region.
As with most of the Great Basin, Northern Paiute groups of the project area were organized on the basis of nuclear households or families (McCarthy 1997a, 1997b).
Winter, however, usually brought groups of often-related families together in a loose aggregation of anywhere from three to ten households near a reliable source of water and fuel. This included low elevation settings in the Honey Lake Basin and Secret Valley, the Madeline Plains being evidently too harsh to support such winter settlements (Riddell

36
1960a). Winter houses were conical in shape, slightly dugout on the inside, and typically constructed of lightweight materials such as tule mats over a pole framework.
Families moved more frequently during the summer, disbanding into smaller camps where shelter was restricted to simple brush structures designed to provide protection from the sun and wind (Stewart 1941). As various resources became available, families traveled to exploit them, carrying only what was needed to complete the task and survive.
According to Riddell (1960b:32), the acorn was the single most important plant food in the Wadatkuht region and was collected from black oaks growing along the western edge of Honey Lake near Milford. Besides acorns, deer were hunted throughout the year, especially in the fall, and antelope communally hunted during the spring
(Riddell 1960b:40). Smaller animals of economic significance included jackrabbits and cottontails, exploited throughout the year, as well as waterfowl that gathered at Honey
Lake during the winter (Stewart 1941:369). The Susan River and Long Valley Creek provided suckers and trout that were caught with nets during the spring spawning season
(Riddell 1960b:34). All of these resources were probably dried and stored to some extent for winter use.
Roots were another important resource for the Northern Paiute, with groups traveling to upland camps as soon as the ground was visible in the spring (Couture 1978;
Foster-Curley 2006; Fowler 1986; Kelly 1932). Roots were harvested by women with digging sticks in the morning, and processed in the afternoon. Most roots were peeled and cleaned by hand, but larger, hardier varieties were peeled with curved stone tools

37
(Couture et al. 1986). Once cleaned, roots were dried and stored or ground into flour with millingstones. Roots such as camas that contain a high inulin content were pit roasted before they were consumed or dried (Fowler 1986). Once processed, geophytes were stored in baskets and/or shallow rock or grass lined pits near villages (Couture
1978; Foster-Curley 2006; Mahar 1953).
GEOPHYTE ETHNOBOTANY OF SURROUNDING AREAS
Use of root crops is ethnographically well documented in the northern Great
Basin and Columbian Plateau (Ames and Marshall 1981; Coutore et al. 1986; Hunn et al.
1998; Kelly 1932; Meilleur et al. 1990; Prouty 1995; Thoms 1989; Turner and Kuhnlein
1983). By contrast, ethnographic information on the gathering, processing and storage of roots in the Pit River region is poorly documented (Garth 1953; Hall 1944; Kniffen 1928;
Neasham 1957; Voegelin 1942). This requires that we examine root exploitation among adjacent groups, including the Klamath, Modoc, Shasta, and Atsugewi. Although minor differences may have existed, root-processing techniques were probably universal among groups who exploit them, such that it can be expanded to include the Columbian Plateau.
The following discussion summarizes ethnographic root gathering and processing techniques among various groups in proximity to the study area.
Modoc and Klamath
Geophytes played a critical role in the Klamath and Modoc economy (Barrett
1910; Gatschet 1890; Ray 1963; Spier 1930). Chief among these were the tubers and

38 bulbs of camas and epos (Barrett 1910). The Modoc gathered roots in the early spring from temporary logistical camps. The gathering season only lasted about a month, during which women worked frantically to harvest and process enough roots for winter storage
(Ray 1963). Ray (1963:198; see also Prouty 1995) notes the collection of biscuitroot:
This parsley, a perennial plant with a turnip like tuber of large size, was found in the sagebrush-scrub country: Tule Lake was a productive area.
The early appearance and relative abundance of the plant gave it a prominent place in the Modoc dietary. The tubers were cooked and eaten fresh, also dried for winter.
The Klamath and Modoc also gathered epos (cf. yampa) during the late spring fish runs. Epos is found at higher elevations and was gathered when the roots were milky, soft, and easily pulled out of the ground (Prouty 1995). They are found in wet meadows, open pine forests, and on dry rocky slopes, with a harvest season that lasted approximately three to four weeks. This required concerted effort over a brief period by large groups of women in order to collect enough roots for winter storage.
As epos was gathered and depleted, camps were moved to new locations until the season ended. Productivity lasted approximately three to four weeks, much of which was occupied with travel to and from collecting areas and the cleaning of harvested tubers every few days. Women collected for only their immediate family, with typically a basketful gathered daily (Barrett 1910:243; Prouty 1995:28; Ray 1963:198).
Women gathered camas at higher elevations beginning in June and July, when men hunted and fished. Camas was preferred over other root crops, but was less abundant (Prouty 1995:29). Camas was collected much like epos, but often cleaned and partially dried before transport (Barrett 1910; Ray 1963). The season typically lasted

39 about a month, with women sometimes traveling miles to exploit a field until it was depleted (Prouty 1995).
Shasta
The Shasta exploited varied resources, but geophytes were a critical component of the diet (Holt 1946). Gleason (2001) notes that as local root crops were depleted, the
Shasta established logistical camps away from permanent villages. As with the Modoc, roots, especially “Ipos,” were gathered with a digging stick and transported back to villages in burden baskets. Bulbs were then “husked by trampling in a shallow place in the creek,” allowing the skins to float away (Holt 1946:308). Cleaned roots were then dried in the sun, winnowed, and stored in baskets. Another bulb in “the shape of an onion” was buried with hot rocks and cooked over night before it was pounded and molded into a large block (Holt 1946:308). These root crops were then stored at villages either in outside pits or baskets within the home (Holt 1946).
Atsugewi
Epos and other roots provided an important staple for the Atsugewi (Garth 1953).
Women collected epos in large quantities in the spring, along with other roots such as camas, brodiaea, and wild onion. All were collected using a crutch-handled digging stick, like those employed by root-dependent cultures of the Plateau and were then transported back to villages in burden baskets (Foster-Curley 2006; Garth 1953). The collected geophytes were then “placed into a shallow basket with sand and worked back

40 and forth with the feet until the skins came off” (Garth 1953:138). Skinned roots were then placed on large rocks for drying and later stored for winter use. Other roots heavily exploited by the Atsugewi included camas ( Camassia quamash ) and biscuitroot
( Lomatium sp.), both of which were processed in earth ovens. Garth (1953), describes this process,
“a pit was lined with rocks and a fire was built upon it. The ashes were raked out and the bulbs were placed on top of a layer of pine needles and covered with pine needles and a layer of dirt. Then a fire was built on top.
After cooking all night they were taken out and mashed and than made into cakes…dried and stored” (1953:138).
As with other groups, roots were stored by the Atsugewi either indoors in baskets or outdoors in burnt tree stumps (Voegelin 1942). Other root crops (e.g., brodiaea, tiger lily, wild onion) were also cooked in earth ovens, but most were eaten immediately (Garth
1953).
Columbian Plateau
Southern Columbian Plateau groups (Sahaptin and Okanagan-Colville) made heavy use of Columbian River as well as upland resources (Prouty 1995). The collection of roots (mainly Lomatium sp.) was a major subsistence pursuit (Kuhnlein et al. 1982;
Meilleur et al. 1990; Turner and Kuhnlein 1982). In the spring, while men were fishing, women gathered and stored roots for the following winter. Prouty (1995:31) describes the harvest occurring first on sunny, south facing slopes by the river, where roots matured early. From there the harvest moved to shady, north facing slopes, and finally up into the sagebrush-steppe. Fourteen species of roots were collected in conjunction with

41 biscuitroot, including brodiaea, sego lily, and balsamroot (Hunn 1990; Turner and
Kuhnlein 1982). Camas, brodiaea, and yampa were also gathered in well-watered meadow areas (Hunn 1990; Prouty 1995). Camas was one of the most important subsistence resources, occurring in patches that ranged from several acres to several square miles in size. This resulted in a highly predictable resource (Foster-Curley 2006).
Camas was available from the late spring to early summer and gathered with a modified digging stick that was used to expose the roots (Turner and Kuhnlein 1983). The bulbs were cleaned, roasted, and eaten whole or dried for storage at another location (Kuhnlein et al. 1982). In east-central Oregon, hundreds of people would gather to dig, process, and bake camas bulbs in earthen ovens during the early summer (Couture 1978; see Prouty
1995). Another geophyte exploited by the Okanagan-Colville of the northwest Plateau was bitterroot (Turner and Kuhnlein 1982). When collected with Lomatium sp. it could provide 3800 calories per hour (Couture et al. 1986; Jones and Madsen 1991). In sum,
Columbian Plateau groups exploited various geophytes, which accounted for 50% of their caloric intake (Hunn 1990).
DISCUSSION OF ETHNOGRAPHIC BACKGROUND
All of these cultural groups made use of various geophytes during different times of the year. Many identical collection and processing techniques were employed between the Plateau, Great Basin, and Californian groups. Most of the groups likewise shared some type of seasonal settlement-subsistence pattern involving the intensive acquisition of roots during the spring and summer. As such, the location of root harvesting grounds

42 influenced settlement location and probably the duration of occupation and cycle of seasonal transhumance as well.
ARCHAEOLOGICAL BACKGROUND
Archaeological investigations within the project area are limited, apart from two large projects, the Tuscarora Gas Transmission (Delacorte 1997a, 1997b) and Alturas
Intertie (McGuire 2000a, 2000b) studies. Various Bureau of Land Management and U.S.
Forest Service surveys within the Pit River Uplands (Manuel 1983) and excavations by
Riddell (1956, 1960b) in the Madeline Plains and Secret Valley comprise some of the earliest work in the area, leading to the development of a regional cultural chronology.
Other archaeological research in the wider region includes that in the Modoc Uplands
(Galm 1985; Hardesty and Fox 1974), Surprise Valley (O’Connell 1971, 1975;
O’Connell and Hayward 1972), and Honey Lake Basin (Riddell 1958, 1960a). Other studies have investigated numerous prehistoric sites of early Holocene throughout protohistoric age (e.g., Delacorte 1997a; Foster-Curley 2006; Gleason 2001; Hughes
1977; Mack 1983; Manuel 1989; McGuire 2000a; Prouty 1994, 1995; Waechter and
Andolina 2005). Many of these contain geophyte-related features that include rock rings, talus pits, underground storage caches, and roasting pits or earth ovens. Much of this previous research subscribes to one of two views regarding prehistoric human adaptation as either spatially and temporally uniform or variable through both space and time.
Julian Steward (1933, 1938) argued that environment conditioned Great Basin settlement and subsistence patterns. He believed that the widely distributed,

43 unpredictable nature of subsistence resources required a highly mobile settlement strategy entailing frequent residential moves. Building on this premise, Jesse Jennings (1957) argued that Holocene climatic conditions had remained comparatively stable, as had the basic human adaptation. Jennings (1964) elaborated on this concept, proposing a unified
“Desert Culture” that allowed for only minor variation in regional adaptive responses.
This model was widely accepted by American archaeologists, but some questioned the uniformity of the environment in relation to human adaptation. Baumhoff and Heizer
(1965), Heizer (1966), and Heizer and Krieger (1956) are among those who called attention to a more regional approach that looked at human adaptation on a local scale.
In the northern Great Basin, Luther Cressman (1956) retained a belief that significant climate change had occurred, leading to the development of a highly specialized lacustrine adaptation that focused on marsh resources. To the south, in the central Great Basin, Heizer (1967; Heizer and Napton 1970) argued for a similarly variable adaptive history that included enclaves of a specialized marsh or
“limnosedentary” lifeway.
Expanding on Heizer’s notion of semi-sedentary settlement, Margaret Weide
(1968) proposed the existence of a pattern that featured both lowland lacustrine and upland hunting elements in the Warner Valley of southeastern Oregon. According to her, this pattern was comparatively stable from approximately 4500 B.P. to European contact, with a shift to greater use of upland resources due to a decrease in lacustrine productivity.
James O’Connell’s (1971, 1975) interest in the northern Great Basin centered on the issue of adaptive change in Surprise Valley, California. By focusing on different

44 biotic communities, O’Connell (1971) identified differences in house structures and artifact styles that suggested a major change in regional population. O’Connell (1971) also believed that early populations made extensive use of lowland habitats, while later groups were more widely dispersed and made greater use of upland plants. This included many starchy roots (e.g., camas, cattails, spikerush, epos) that were collected and sometimes ground in mortars, but often eaten without further preparation. While acknowledging these adaptive shifts, O’Connell ultimately argued for a basically stable pattern of settlement and subsistence over the last 6,000 years in Surprise Valley.
Soon after, David Hurst Thomas (1973) began working in the Reese River Valley of central Nevada. His research was modeled on Steward’s (1938) description of Great
Basin settlement and subsistence adaptations. Much like O’Connell (1971), Thomas
(1973) explored the relationship between lowland and upland habitats and resource exploitation. More specifically, he focused on the seasonal mobility between spring/summer lowland and fall/winter upland occupation originally described by
Steward (1938). Central to this pattern were the highly abundant and storable upland pine nut crops that diverted Steward’s, and later Thomas’, attention from other important resources, such as roots.
In more recent years, research in the northwestern Great Basin and Columbian
Plateau has focused on the processes leading to root intensification (Aikens and Jenkins
1994; Ames and Marshall 1981; Delacorte 2000; Foster-Curley 2006; Gleason 2001;
Jenkins 1994; Prouty 1995; Thoms 1989; Yu 2006).

45
In the Pacific Northwest, use of camas ( Camassia quamash ) has been archaeologically and ethnographically well documented among Plateau peoples (Ames and Marshall 1981; Connolly 1986; Thoms 1989). This is thought by some to be related to the lack of other storable plant resources, such as nuts and seeds (Foster-Curley 2006).
More recent research in the Fort Rock Basin of southeastern Oregon reveals prehistoric settlement and subsistence patterns focusing on lowland resources/habitats until 1500
B.P., when spring-summer use of upland settings began to focus on storable root crops
(Foster-Curley 2006; Fowler 1993; Jenkins 1994; Jenkins and Connolly 1990).
Much the same appears to be true in the Klamath River Canyon of northern
California, where Gleason (2001) investigated the prehistoric importance of root crops.
She suggests that after 2000 B.P., use of plants increased substantially. More significantly, Gleason (2001) showed that acorn use was directly linked to epos
( Perideridia sp.) root availability. Because geophyte collection occurred in the early spring, the need for acorn crops could be monitored, with the dependability of epos equal or greater to that of acorns (Gleason 2001).
NORTHEASTERN CALIFORNIA CHRONOLOGY
Many chronological sequences have been proposed for northeastern California
(Aikens and Jenkins 1994; Bedwell 1973; Clewlow 1968; Cressman 1986; Cressman et al. 1940, 1942; Delacorte 1997a; Elston 1982; Grayson 1973; Hughes 1986; Layton
1970; O’Connell 1971, 1975; Raven 1984; Riddell 1960a). All of these are based on northwestern Great Basin/southern Plateau projectile point forms tied to Great Basin

46 sequences (Bettinger and Taylor 1974; Heizer and Hester 1978a, 1978b; O’Connell 1967;
Thomas 1981), with regional variations proposed by different researchers.
In Secret Valley and the Madeline Plains, Francis Riddell (1960; Theodoratus et al. 1979) offered a post-Early Holocene sequence based on the Karlo site (CA-LAS-7),
Amadee Cave (CA-LAS-90), and other regional excavations. It identified four major periods: Madeline Plains (6500 - 4500 B.P.), Karlo (4500 - 1950 B.P.), Tommy Tucker
(1950 - 450 B.P.), and Amadee (450 B.P. – Ethnographic period). Each of these was defined on the basis of projectile point types and later re-evaluated by Bennyhoff and
Hughes (1987). They proposed a seven period sequence based on burial lots (i.e., bead and projectile point types). These include: Early Karlo (Early period beads; Elko-earred,
Gatecliff series and Large Side-notched points), Transitional Karlo (Gatecliff series and
Large Side-notched points), Terminal Middle Period (Elko Corner-notched points), Phase
I (shallow-surface beads; Rosegate series points), Early Amadee (Desert Side-notched,
General sub-type points), Late Amadee (Desert Side-notched, Sierra sub-type points), and Historic (Glass beads).
The chronology of the Pit River Uplands relied, until recently, on surrounding areas including Surprise Valley (O’Connell 1971, 1975) and the Fort Rock Basin (Aikens and Jenkins 1994; Bedwell 1973; Cressman 1986). These were of some utility, but failed to accurately portray the sequence in places like the Pit River Uplands, Madeline Plains, and Secret Valley. As Delacorte (1997a:65) observed, the most reliable solution to this problem is to use arbitrary temporal units or periods that have no necessary cultural historical or behavioral significance. With this in mind, the temporal sequence employed

47 for the Tuscarora Gas Transmission (Delacorte 1997a) and Alturas Intertie (McGuire
2000a) projects was adopted for the present study (Figure 3.3). It identifies six periods:
Early Holocene (12,000 – 7,000 B.P.), Post-Mazama (7,000 – 5,000 B.P.), Early Archaic
(5,000 – 3,500 B.P.), Middle Archaic (3,500 – 1,300 B.P.), Late Archaic (1,300 – 600
B.P.) and, Terminal Prehistoric (600 B.P. to Contact). Each of these periods is defined below:

Figure 3.3 - Comparative Chronological Sequences for
Northeastern California (after Delacorte 1997a).
48

49
Early Holocene (12000 – 7000 B.P.)
The Early Holocene (cf. pre-Archaic) is defined by two phases of occupation.
The first consists of a widespread Paleo-Indian tradition associated with large-game hunting that is marked by fluted projectile points that probably predate 9,000 B.P. The second is the Western Pluvial Lakes Tradition (Bedwell 1973), distinguished by the presence of large, Great Basin Stemmed series points and limited quantities of ground stone. These earliest time periods are mostly represented along pluvial lakeshore and marsh contexts. Adaptations during this interval are believed to have focused on marsh/wetland resources and possibly large game that may have been attracted to these favorable lacustrine habitats.
Early Holocene assemblages are typically characterized by Stemmed series projectile points, crescents, steep-sided formed flake tools, bifacial thinning debitage
(McGuire 2000a:27), and little or no milling equipment that is not highly questionable
(Delacorte 1997b). A hunting-oriented foraging pattern is shown to include a vast amount of terrain, with the bifacial reduction of diverse tool stone materials that include obsidian, basalt, and cryptocrystalline stone (McGuire 1997:222). This type of settlementsubsistence pattern and hunting focus suggests a highly mobile, technologically flexible adaptation by regionally limited forager populations who took advantage of various tool stone as it was encountered or embedded within the foraging pattern (McGuire 2000a:27; see Basgall 1988; Kelly 1988).

50
Post-Mazama (7000-5000 B.P.)
The Post-Mazama period is distinguished by the appearance of Northern Sidenotched projectile points that are distributed primarily north of the Humboldt River in the
Plateau and northern Great Basin. Tool stone diversity remains high during the Post-
Mazama period, but ground stone milling equipment appears in substantial quantities for the first time. The Modoc Uplands demonstrate its earliest use, most likely on a seasonal basis, with no appearance of storage features (Hildebrandt and Mikkelsen 1995). Large semi-subterranean pit houses in Surprise Valley (O’Connell 1975) and elsewhere bespeak a shift in settlement and social organization that included seasonal occupation by extended families or similarly complex social groups. Subsistence remains include a high proportion of large mammals, such as bighorn sheep, elk, and bison, which suggest communal hunting. On balance, the Post-Mazama period differs from adaptations in other parts of the Great Basin in its seasonably mobile/tethered settlement pattern and seemingly complex social organization.
Early Archaic (5000-3500 B.P.)
The Early Archaic corresponds with the end of the harsh environmental conditions associated with the Middle Holocene, and may have witnessed a higher concentration of people in the region (Milliken and Hildebrandt 1997). Both flaked and ground stone increase in abundance and may reflect an expansion in subsistence resources (Hildebrandt 2000). Projectile point styles include both split-stem (cf.
Gatecliff, Silent Snake Spring, etc. [Thomas 1981]) and concave-base forms (cf.

51
Humboldt [Heizer and Baumhoff 1961]), which seem to have originated in the northern
Great Basin and diffused south. Flaked stone material in Secret Valley reveals the introduction of basalt core/flake technology (McGuire 1997) that is believed to correspond with the use of resources such as roots and other plants. Elsewhere, ground stone mortars change from “v” to “u”-shaped basins that may indicate a shift in plant use, but apart from this, technologies remained much the same. The semi-subterranean house features in Surprise Valley (O’Connell 1975) were replaced, however, by the domed brush wickiup that persisted into historic times. Faunal assemblages show an increase in small animals and greater use of waterfowl and fish, implying a broadening economy. In other respects, large villages appear throughout the area, connoting seasonally reduced mobility and perhaps summer disbandment of groups reminiscent of ethnohistoric populations (e.g., Klamath, Pit River, etc.).
Middle Archaic (3500-1300 B.P.)
The Middle Archaic period witnessed a major increase in the number of sites and environmental settings exploited. This included increased occupation of sites such as
Karlo (Riddell 1960) and those in Surprise Valley (O’Connell 1975) with well-developed middens, grave goods and domestic structures. Many of the sites from this period were later abandoned, resulting in peak occupation of some areas (Hildebrandt and Mikkelsen
1994). Coinciding with the changes in the Modoc Uplands was an intensive biface reduction technology that characterizes Middle Archaic components throughout the region (McGuire 2000a; Mikkelsen and Bryson 1997). Middle Archaic remains in Secret

52
Valley indicate a major increase in occupation, marked by diverse artifact inventories and evidence of resource intensification (McGuire 1997, 2000a).
Sites of this period are identified by the presence of Elko and Siskiyou Sidenotched projectile points, and an increase in biface manufacture. House structures and other elements of the technology changed little, however, from those of the preceding
Early Archaic interval. Much the same appears to be true for subsistence patterns that show little change from the preceding period. Thus, apart from a greater number of sites, implying a growing intensity of activities, Middle Archaic remains show little difference from those that came immediately before.
Late Archaic (1300-400 B.P.)
The Late Archaic is identified by two different arrow point styles: Gunther and
Rose Spring/Eastgate. The Late Archaic is conventionally defined by the advent of the bow and appearance of these smaller point forms (Delacorte 1997a). Significantly, all of these point styles were originally identified in other places, such that their local dating and significance is far from secure. In addition to the introduction of the bow, the Late
Archaic sees a shift to the use of increasingly local tool stone such as chert, and a potentially greater reliance on large game (Holanda 2000). In Secret Valley, land use changed drastically around 1000 B.P. with a sharp decrease in obsidian reduction and abandonment of at least some sites (McGuire 2000a). Ground stone milling equipment increases in abundance during the Late Archaic, most notably in upland contexts such as the Madeline Plains and Pit River Uplands (Delacorte 1997b). This suggests a major

53 shift in settlement location around 1000 B.P. to intensively exploit root crops
( Perideridia , Camassia , etc.) in areas such as the Fort Rock and Massacre Lake basins
(Aikens and Jenkins 1994; Leach 1988) and Pit River Uplands west of Alturas (Delacorte
2002). Along with the shift in settlement-subsistence, use of more local tool stone materials implies a decline in obsidian biface reduction that may signal a decrease in mobility (Delacorte 1997b) that coincides with the decreased use of Secret Valley.
Terminal Prehistoric (400 B.P. to contact)
The Terminal Prehistoric interval is conventionally placed after 600 B.P., coinciding with the traditional dates assigned to Desert Side-notched and Cottonwood series projectile points (Delacorte 1997a). In the northwestern Great Basin, however, projectile points of these series seem to date somewhat later in time, ca. 400 B.P. In fact,
Desert Side-notched points are believed by some to be cultural markers of Numicspeaking people, who allegedly spread across the region beginning around 1000 B.P.
(Delacorte 2008). Coinciding with these points is a shift in settlement from large seasonal or semi-permanent to smaller household encampments containing fewer specialized tools, a higher ratio of ground to flaked stone artifacts, and a greater use of expedient tools than earlier time periods (Delacorte 1997a; Foster-Curley 2006). This may have also coincided with changes in subsistence resources and how they were obtained.
In Secret Valley these changes include a shift from biface to core/flake technology identified in other parts of the western Great Basin (Bouey and Basgall 1984;

54
Elston 1986) where it is believed to reflect more localized settlement and correspondingly limited tool stone access (McGuire 2000a). On the Modoc Plateau, there is a decrease in archaeological remains (Mikkelsen and Bryson 1997), possibly due to worsening climatic conditions. Conversely, in the Pit River Uplands and Black Canyon both the number and size of sites expand substantially for yet debated reasons (cf. Delacorte 1997b, 2002;
Waechter 2000b).
PILOT STUDY SITES
Artifacts analyzed as part of this study derive from several archaeological sites each having components dating to the Late Archaic and Terminal Prehistoric Periods (ca.
1300 to contact). These sites are situated throughout a range of topographic and physiographic settings all composing an area referred to as the Pit River Uplands,
Madeline Plains, and Secret Valley.
Pit River Uplands
Three prehistoric sites in the Pit River Uplands were selected for the current study: CA-MOD-3150, CA-MOD-3153, and CA-MOD-3448. All are located in close proximity to one another along the Crooks Canyon Creek drainage (Figure 3.4). Each of the sites has Late Archaic/Terminal Prehistoric house structures and diverse artifact assemblages, implying prolonged and/or intensive occupation during the late prehistoric interval. The following site descriptions are adapted from the site reports (McGuire
2000a).

Figure 3.4 – Site Locations in Pit River Valley.
55

56
The archaeological site of MOD-3150 is located on a hillside adjacent to the northeastern shore of Alkali Lake in the volcanic uplands west of the South Fork of the
Pit River. It is described as a “large seasonal base camp” (Rosenthal 2000), with a midden deposit, cairn burial, and seven rock-ring features (four houses and three hunting blinds). Five loci were identified at the site and range in age from the Middle Archaic to the Historic period. The Middle Archaic component is believed to represent a short-term hunting camp or retooling station, based on the number of bifaces, biface debitage, flake tools and elevated location. Logistical hunting parties are thought to have used the uplands as task-specific procurement areas, traveling from more permanent residential bases in the Pit River Valley (Rosenthal 2000).
The Late Archaic and Terminal Prehistoric components could not be separated and comprise the largest part of the site assemblage. Three of the four house features are associated with this late component, indicating a shift from a logistical to a more residential settlement pattern. The site location also shifted during this period, with the house structures located farther to the south among some trees, not the open area used during the Middle Archaic period. Tool diversity is likewise greater during the late prehistoric occupation, including milling equipment, formed flake tools, and projectile point blanks. The presence of milling gear and recovery of burnt seeds and roots indicate that plant resources were of considerable importance during the Late Archaic/Terminal
Prehistoric period.
Site MOD-3153 is a large, complex prehistoric site overlooking the South Fork of the Pit River Valley (Waechter 2000a). It encompasses three different loci, which

57 include 15 rock-ring house structures and talus pits, small lithic scatters, and petroglyphs.
Most of the site data derive from domestic structures and related features in Locus A, which date to the Terminal Prehistoric and Terminal Prehistoric/Historic periods. The
Middle/Late Archaic component consists of a few flaked stone tools, milling implements, and flakes found on the surface of Locus B along a broad, open stream terrace flanking
Crooks Canyon Creek. Locus C is located on an escarpment above the creek and consists primarily of petroglyphs and rock features thought to be of Middle-Late Archaic age
(Waechter 2000a:1.63).
The Terminal Prehistoric structures were spread across a series of benches overlooking a spring and Locus B below. Two of the structures consist of talus or storage pits and the other 13 are larger rock rings. Most of the structures were clustered in three groups on elevated benches with milling equipment and a possible roasting/processing feature interspersed between (Waechter 2000a). Floral remains from the structures show heavy use of charred roots and bulbs with epos ( Perideridia sp.) found in five of the excavated rock-rings and brodiaea ( Brodiaea sp.) found in two
(Wohlgemuth 2000). Few seeds were recovered and consist of primarily sunflower
( Helianthus sp.), blazing star ( Mentzelia sp.), bluegrass ( Poa sp.), and goosefoot
( Chenopodium sp.). All of these genera are more common in Secret Valley and Honey
Lake sites, prompting Wohlgemuth (2000) to suggest differences in seed-processing techniques, site seasonality, or greater use of roots at these upland sites. The latter explanation is the most likely given the widespread recovery of roots in the houses
(Wohlgemuth 2000:1.82). Overall, MOD-3153 epitomizes the shift in Pit River Uplands

58 settlement-subsistence patterns from Middle Archaic to the Terminal Prehistoric/Historic period times, changing from a logistical hunting to intensive root gathering pattern.
The site of MOD-3448 is situated on the northwestern slope of Graven Ridge, south of Crooks Canyon Creek between the two previous sites (Holanda 2000). It consists of three loci: Locus A contains a relatively deep midden and dense flake scatter dating to the Middle and Late Archaic period; Locus B includes two rock-ring house features and a shallow deposit dating to the Terminal Prehistoric period; and Locus C has three rock-ring house structures and associated concentration of ground and flaked stone artifacts that date to the Terminal Prehistoric period. The site appears to have been abandoned just prior to historic times (Holanda 2000).
Plant remains from MOD-3448 were recovered from two of the Terminal
Prehistoric structures. They include charred seeds of bluegrass, goosefoot, hairgrass, blazing star, phacelia, rush, plantain, bitterbrush, prunus, and grass seed. More interesting was the recovery of burnt root and bulb fragments of both epos and brodiaea, much like the neighboring site of MOD-3153.
The Middle/Late Archaic component in Locus A has an assemblage consistent with the procurement and processing of large-game on seasonal or short-term forays and lacks house structures. The Terminal Prehistoric components in Locus B and C indicate a shift in land-use to a more plant-based economy, evidenced by an abundance of ground stone, plant remains, and appearance of house structures connoting long-term occupation.
In sum, the settlement-subsistence pattern in the Pit River Uplands began with one of logistical hunting during the Middle Archaic period that gave way during the Late

59
Archaic and Terminal Prehistoric interval to one of intensive root procurement. Whether this reflects resource intensification (Delacorte 2002) or a move to the uplands to avoid conflict (Waechter 2000b) is unclear, though the former seems more likely on the face of the evidence.
Madeline Plains
The Madeline Plains is a featureless basin created when Pluvial Lake Madeline dried after its last major high stand (Delacorte 1997a). Archaeological samples from this area will be represented by one late prehistoric site, CA-LAS-1623/H (Figure 3.5).
LAS-1623/H sits on the northeastern edge of the Madeline Plains directly atop the
Pleistocene lake sediments forming the Plain (Leach-Palm 2000). It is a shallow cultural deposit (2-3 cm deep), comprising an extensive lithic and enormous ground stone
(3,000+ milling tools) scatter. The dominant point type is the Rose Spring form, which indicates, along with obsidian hydration, that much of the assemblage is of Late Archaic age. The Late Archaic assemblage of ground stone is expedient, unshaped, and frequently recycled and suggests intensive exploitation of local plant, most likely root resources (Delacorte 2000). The flaked stone assemblage is of minimal size but surprising diversity, implying numerous activities that might be expected to accompany intensive plant procurement. That so few flaked and so many ground stone artifacts are found in comparison to the Pit River Upland sites may reflect different steps in the processing of plant crops, or use of perhaps different resources between the two areas. In

60 short, the Madeline Plains show a shift from logistical hunting to more intense occupation and plant processing from Middle to Late Archaic times much like the Pit River Uplands.
Secret Valley
Secret Valley is another Pleistocene lake basin, but includes the rim-rock drainage of Secret Creek and seasonally flooded playa of Mud Lake. Two large archaeological sites were chosen to represent the valley: CA-LAS-206 and CA-LAS-1705 (Figure 3.7).
LAS-206 is located on an alluvial fan on the eastern side of Secret Creek (McGuire
2000a). The site is an extensive tool scatter concentrated in five separate loci. These include two relatively thick habitation middens, as well as numerous subsurface features, including house floors, human and dog burials, and earth ovens/hearths. The most intensive use of the site occurred during the Middle to Late Archaic transition (2350 to
1000 B.P.), with most of the site material dating to the Middle Archaic period. This includes much of Locus A, Structure 1 of Locus E, and Structure 1 of Locus D. Ground stone artifacts are more formalized and seemingly curated than those of either the
Madeline Plains or Pit River Uplands, reflecting a clearly different vegetal processing regime. Mortar/Pestle technology appears during the Middle Archaic period, but becomes less formalized by the Late Archaic interval. Two of the house structures, which date to the transitional period, were occupied for a seemingly long duration and are unusually large. The third house floor dates to the Late Archaic period and was occupied for a shorter period of time or single season (McGuire 1997).

Figure 3.5 – Site Locations in Madeline Plains and Secret Valley.
61

62
A few isolated features and two cairn-capped burials characterize other Late
Archaic and Terminal Prehistoric remains. Dominant flaked stone tools consist of simple flake tools produced with an increasingly core/flake-based technology. Milling tools are less formalized and generally thicker, as expected of short-term, expedient occupations.
Paleobotanical data indicate greater use of dryland ( Phacelia sp., Acnatherum hymenoides, Juniperus sp.) than wetland taxa ( Deschampsia sp., Typha sp.), though the latter were locally less abundant. The site appears, therefore, to have been more heavily used as a residential base during the Middle and Late Archaic than the Terminal
Prehistoric interval, when use of LAS-206 was restricted to shorter seasonal or logistical occupations.
The site of LAS-1705 is situated along the northeastern edge of Secret Valley on an ancient delta/lacustrine deposit that predates human occupation of the region
(Hildebrandt 2000). It is a complex site with both early and late components that include two house structures, three midden deposits, and two human burials of Late Archaic age.
Changes in Middle to Late Archaic assemblages include the addition of mortars/pestles and baked clay, coinciding with increased occupational intensity and the appearance of formal house structures and burials. Locus A has two midden areas, one of Middle to
Late Archaic age with two structures and another of Late Archaic to Terminal Prehistoric age. Both produced diverse tool assemblages including mortars/pestles, abundant charred seeds, and house remains. The Locus B midden had a lower Post-Mazama to
Early/Middle Archaic component with a diverse assemblage and abundant animal bone.
The upper Middle/Late Archaic component produced a larger, more intensively used tool

63 assemblage, a higher proportion of lagomorph to artiodactyl bone, and the introduction of baked clay. Paleobotanical remains, consisting of goosefoot and sunflower, were restricted to the upper Middle/Late Archaic component. Locus C had both a Post-
Mazama and Early Archaic component containing few artifacts, no paleobotanical remains, and a high proportion of artiodactyl remains. Both of these early occupations appear to represent short-term use of the site, characterized by initially Post-Mazama hunting and later other activities. In sum, LAS-1705 witnessed longer occupation as first a short-term encampment and later a residential base. This was followed after 1000 B.P., by a significant decrease in occupation at this and other Secret Valley sites, coinciding with perhaps the late prehistoric expansion of Numic peoples across the Great Basin.
In sum, evidence from Secret Valley indicates a period of Middle Archaic subsistence intensification that may have resulted from either population pressure or some type of environmental change. Paleobotanical data point to a shift from dryland to wetland resources between the Middle Archaic and subsequent Late Archaic/Terminal
Prehistoric periods. These data suggest that Secret Valley witnessed a gradual increase and subsequent decline in prehistoric occupation with the appearance of first village-like settlements and later logistical use of the valley after 1000 B.P.

64
Why Focus on Northeastern California
Archaeological trends in northeastern California are similar to those in other parts of California and the Great Basin. Earlier remains point to highly mobile foragers who relied on large game and other high-return resources, while later materials connote less mobile, more centralized settlement and greater reliance on low-return plant and animal resources. The general trend is one of increasing resource diversity and intensification over time, with intensive use of root crops occurring only later in time.
As previously discussed, use of roots in the northwestern Great Basin and northeastern California was of generally greater importance than surrounding areas, as they were the only easily intensifiable resources. The selected sites from the Tuscarora and Alturas projects contain some of the largest Late Archaic and Terminal Prehistoric assemblages from the region. The collections are also diverse and have exceptional potential for the application of starch grain analysis. Similarities between the sites also allow for comparisons between tool categories and environmental locations (e.g., Pit
River Uplands and Secret Valley) in relation to the exploitation of roots. More to the point, roots appear to have been more important in the Pit River Uplands than Secret
Valley, where settlement patterns also differed. This provides a unique opportunity to explore the veracity of starch grain analysis between settings that share some, but not other cultural variables, where the results can resolve still outstanding issues in prehistory.

65
Chapter 4
METHODS AND RESULTS OF MODERN STARCH ANALYSIS
This chapter describes the methods used to construct a modern starch reference collection that served as the key for identifying archaeological starches recovered from prehistoric tools. Comparative studies indicate that starch granules from modern epos, biscuitroot, and other geophytes are morpho-metrically distinct from those of other plant resources, such as oak and pinyon, and provide a characterization of the grains from these taxa.
To understand the basis of starch-grain analysis within the study area, certain parameters must be understood. First, ancient starches must be preserved on either ground or flaked stoned tools, in storage, roasting, or other feature contexts, or in cultural sediments associated with geophyte/plant use. Second, starch taxa of dietary significance must be identifiable on the basis of diagnostic traits that distinguish them from other archaeological starches. The latter requires a modern reference collection of potentially exploited plants, given regional ethnographic and biogeographic information.
Use of starch-grain analysis in California and the Great Basin has been limited at best. In fact, most of the work in California has been restricted to the upper Klamath
River Basin (Gleason 2001) and ongoing research along the central California coast (Rob
Cuthrell, personal communication), with more sporadic efforts undertaken in the realm of cultural resource management (Eric Wohlgemuth, personal communication).
More recently, Scholze (2010) examined tools in curated collections to ascertain the presence/absence, condition, and longevity of starch granules. This revealed that

66 starch grains could survive on curated artifacts for extended periods, providing the basis for further analysis and research.
To date, no starch grain reference collection has been developed for the study area. As such, a preliminary reference collection of six wild geophyte crops was compiled: Allium sp., Triteleia sp., Calochortus sp., Camassia sp., Lomatium sp., and
Perideridia sp. Two of these taxa ( Allium and Camassia ) were found to contain no starchy residue and could not be identified with the present research methodology. As described below, the comparative collection includes multiple digital photographs of each starch taxon, and reserved starch solution retained for future analysis. The first goal of this work was to establish the degree to which starch grains from different geophyte crops can be distinguished. Given ethnohistoric accounts, Perideridia was assumed to be one of the economically more important Late Archaic subsistence resources. This is especially true given the lack of both pine nuts and acorn in the region.
The reference data generated for this study represents the first step in the construction of a northeastern California comparative sample, although additional taxa will need to be added in the future. These include other geophyte crops and grass and other seed plants, only two of which ( Pinus monophylla and Quercus sp.) were examined as part of the present effort.
Results of the comparative work are extremely promising. Most of the geophytes have distinctive starch grain morphologies that can be usefully applied in archaeological contexts. Starches from geophyte crops appear likewise distinct from those of certain

67 large-seeded food resources, although further research will be necessary to test and refine the identification process.
METHODS
The following section describes the methods used to obtain and analyze the reference specimens. Subsequent sections describe the procedure for analyzing each taxon, as well as the analytical results. Initial efforts to develop a morpho-metric starch identification key were eventually abandoned due to time constraints. Variations in especially the size of individual starch grains and overlap in the size and general morphology of different taxa made it impossible to metrically key particular species exclusively. With this in mind, a more traditional morphological analysis and identification key were developed, with all of the images and data archived for future starch research.
Modern Starch Collection Specimen Recovery
Modern starch samples were obtained from two different sources. The first were plant specimens collected from the study area or elsewhere. These were collected at various times during the year, when different plants were more easily recognized, located, and their starch producing parts in various stages of development. The second source of comparative material was a collection of previously prepared starch grain slides generously provided by Susan Gleason. These were prepared by Gleason (2001) as part of her dissertation research along the Upper Klamath River Canyon approximately 100

68 miles east of the current study area. Morphological variability within plant species was not considered during the present study, given the assumption that starch morphology is reasonably consistent within plant species.
Preparation of Modern Specimens
Key elements for the establishment of a modern reference collection include digital photos of various grains, the preparation of microscope slides, and the preservation of processed plant material in distilled water for long-term storage and future comparative work. The preparation of reference slides with fresh geophyte and seed starch was modeled on two procedures, a semi-permanent collection and a temporary mount for immediate use. Different types of mounting media were used for permanent slides including Permount, karo syrup, and glycerol. Temporary mounts were prepared with distilled water. Preparation procedures generally follow those of Linda
Scott Cummings for a permanent mount of underground storage organs (Field 2006:111) and Beth Gott’s method for the permanent mount of seed starch (Field 2006:109). Both of these methods were selected based on their simplicity as summarized below:
Preparation of Underground Storage Organ (Linda Scott Cummings) (after Field 2006:111)
1) Slice root with a razor blade or scalpel to expose the interior.
2) Take toothpick and scrape starches from the internal surfaces.
3) Rub toothpick on center of microscope slide.
4) Mount slide by adding a drop of karo syrup on top of the starch film and cover with a cover slip. (Push the cover slip down to spread the karo syrup, but not too much that it seeps from the sides of the cover slip.)
5) Seal the edges of the cover slip with clear nail polish.

69
Preparation of Seed (Beth Gott) (after Field 2006:109)
1) Grind about one teaspoon of seeds into a coarse powder with a glass mortar and pestle.
2) Dry filter the ground material through a 0.4mm sieve.
3) Grind again (to break up cells to obtain single granules and to eliminate seed coats). The result is fine powder residue in the glass mortar.
4) Transfer a small amount of the powder to a slide and smear with a small metal spatula.
5) The aim is to remove the larger particles. Holding the slide at an angle and tapping the slide on its side, and/or using a spatula with a gentle smearing movement can achieve this. The smearing action also spreads the particles along the slide.
6) Mount with karo syrup or Permount – one large drop will spread under the cover slip to fill the space. Seal the cover slip with nail polish. The medium stays fluid for a number of years and the granules can still be tumbled for detailed examination.
7) Care is taken to isolate the starch granules and eliminate any extraneous material. The aim is to provide a clean preparation to facilitate image analysis.
Characterization of Modern Starches
At least 50-100 starch granules were examined for each modern taxon, employing a combination of attributes adapted from Piperno and Holst (1998), Torrence et al.
(2004), and Wilson et al. (2010). These are listed in Table 4.1 and discussed further below.
Observations for each starch granule were recorded manually and converted to an
Excel spreadsheet. Two or more digital photos were also taken of each starch granule for all of the analyzed plant taxa (Table 4.2). All measurable starch grain images were systematically analyzed with the seventeen attributes listed in Table 4.1 to assess differences in granule morphology within plant species.

70
Table 4.1 – Starch Grain Measurement Criteria.
Type
Length
Width
Aspect Ratio
Area
2-Dimensional shape
3-Dimensional form
Hilum vacuole
Hilum orientation
Hilum - Aspect Ratio
Lamellae
Fissure
Striae
Facet
Polarization cross style
Degree of Polarization
Polarization cross leg angle
Nominal
Numeric
Numeric
Numeric
Numeric
Nominal
Nominal
Present/Absent
Nominal
Numeric
Present/Absent
Present/Absent
Present/Absent
Present/Absent
Nominal
Nominal
Numeric
Simple, Compound, Semi-Compound
Bell, Ovate, Sub-Ovate, Round, Sub-
Round, Reniform
Bell, Globose, Hemispherical,
Spherical, Lenticular
Centric, Eccentric, Very Eccentric,
Unknown
When Present - Weak or Strong,
Concentric or Linear
When Present - Linear, irregular, transverse, x or y shape
Straight, Wavy, Figure-8
Weak, Medium, Strong
The recording of starches was performed in different stages. At least two separate images were taken for each starch grain in the reference collection and starch grains extracted from analyzed artifacts: one with normal incident light (i.e., white light showing the perimeter of the grain) and a second cross-polarized image (taken in crosspolarized light that shows the distinctive Maltese cross). Observations made under normal incident light include the presence or absence of fissures, facets, striae, lamellae, and hilum vacuoles, as well as the grain type, 2-dimensional shape, and 3-dimensional

71 form. This was followed by a cross polarizing examination of the grains to assess the hilum orientation, style of polarization cross, and degree of polarization. Both normal incident and cross-polarized images were captured using a Sanyo color CCD digital camera. These images were then imported into image analysis software (Image J) and the length, width, aspect ratio, area, hilum aspect ratio, and polarization cross leg angle calculated.
RESULTS
The following section describes the morphological analysis of modern Brodiaea ,
Calochortus , Lomatium , Perideridia , Pinus monophylla , and Quercus starch granules.
The characterization of starches was based on only one or two species per genus, such that species-specific distinctions may have been missed, but differences between genera are clearly distinguishable. As Piperno and Holst (2004) and Rumold (2010) discuss, some taxa may be identified on the basis of a single starch granule, but others require numerous archaeological starch grains sharing a polythetic series of diagnostic attributes or “grain population signature” to be reliably identified (Rumold 2010:235). This latter approach proved especially useful for the current research, where many of the archaeological starch grains were individually ambiguous. Descriptions of the analyzed taxa are provided below, including their distinguishing characteristics. In the case of
Perideridia and Lomatium, grain morphology is sufficiently similar that individual specimens are sometimes indistinguishable and the two genera combined into a single category.

72
Table 4.2 provides the lengths and areas for each of the analyzed starch taxa. As can be seen, none of the starches can be separated on the basis of length and area alone, requiring the use of other morphological and metric attributes to be reliably identified.
When these are included, however, there is sufficient variability between the starches to differentiate taxa, given adequate archaeological samples.
Table 4.2 – Length and Area of Starch Grains by Taxon (μm).
Taxonomic Identification Mean Length Length Range Mean Area Area Range
Perideridia sp.
Lomatium sp.
11.28
13.16
4 - 18
6 - 23
84.18
125.53
40 - 200
50 - 275
Calochortus sp.
17.34 8 - 28 147.47 50 - 300
Triteleia sp.
Quercus sp.
22.16
12.92
10 - 30
6 - 25
278.79
89.58
100 - 450
25 - 200
Pinus monophylla 6.81 4 - 10 29.48 10 - 60
Epos ( Perideridia sp.)
Fifty starch grains from each of two Perideridia specimens from northeastern
California were analyzed as part of the comparative samples. These revealed a combination of traits relating to size, form, surface, texture, and polarization that distinguish larger Perideridia starch granules from those of other geophyte and seed starches examined. Among the most diagnostic traits are grain size, a conical often bell-

73 shaped form, one or more flat facets along the granule edge, elongated centric hilum, and typically longitudinal fissures when present (Fig. 4.1).
Figure 4.1 – Top and Profile View of Perideridia bolanderi Starch.
Starch Granule Size – Perideridia sp. starches range in length from 4-18 microns, with a mean length of 11.28 (Table 4.3). They range in area from 40-200 square microns, with a mean of 84.18 square microns. While size alone cannot be used to identify
Perideridia starch grains, other attributes of their morphology are more distinctive.

74
Table 4.3 – Length and Area of Perideridia sp. Starch (μm).
Taxonomic
Identification
Specimen
Form
Reference/Source
Length
Range
Mean
Length
Std.
Dev.
Area
Range
Mean
Area
Std.
Dev.
Perideridia bolanderi .
Perideridia bolanderi
Perideridia gairdneri
Perideridia howellii
Perideridia oregana
Fresh
Tuber
Fresh
Tuber
Fresh
Tuber
Fresh
Tuber
Fresh
Tuber
G. Scholze (2007)
Alturas, CA
Gleason (2001)
Upper Klamath
Valley, CA
4 - 18 11.28 2.57 40 - 200 84.18 34.84
4.1 - 7.2* 6.1* N/A N/A N/A N/A
Gleason (2001)
Upper Klamath
Valley, CA
2.3 – 8* 5.95* N/A N/A N/A N/A
Gleason (2001)
Upper Klamath
Valley, CA
2.1 - 4.1* 3.02* N/A N/A
Gleason (2001)
Upper Klamath
Valley, CA
2.3 – 8* 5.95* N/A N/A
N/A
N/A
N/A
N/A
* Gleason (2001) slides were analyzed uncalibrated. These slides were reanalyzed under the calibrated microscope producing different starch grain lengths than she reported.
Starch Granule Shape and Form – Observation of numerous Perideridia starch grains, indicate that most are symmetrical, with a conical, bell-shaped, or triangular/ellipsoid form (Fig. 4.1). Some immature Perideridia starch granules are ellipsoid in form, resembling other geophyte starch granules, but are typically smaller in average size. Larger grains with their conical or bell-shaped form and flat or angled facet on the end are, however, unique to this taxon.
Other Characteristics – Other prominent characteristics of Perideridia starch include an elongated centric hilum, a strong degree of polarization, and longitudinal fissures on top of the conical or bell-shaped form. In polarized view, the arms of
Perideridia grains extinction crosses are usually straight with a small bend or curve towards the margin of the granule.

75
In summary, it appears that Perideridia starch granules are characterized by a set of diagnostic traits that would permit their identification if recovered from archaeological contexts. Further analysis is necessary to determine how frequently conical-bell forms, flat faceted ends, and slightly curved extinction cross arms occur, and the size of modern
Perideridia grains where these features tend to occur.
Biscuitroot ( Lomatium sp.)
Fifty starch granules were inspected from fresh tubers of Lomatium bicolor collected from northeastern California. Lomatium bicolor is available in the research area, but numerous species of this genus are found within the region and should eventually be analyzed in order to obtain a better representation of the plant. Lomatium starch exhibits many of the same features as Perideridia sp., having a conical-to-bellshaped form and a flat facet on one end. The presence of a vacuole within the hilum, however, distinguishes Lomatium from Perideridia . Figure 4.2 illustrates the appearance of this vacuole.
Starch Granule Size – Lomatium starches range in length from 6-23 microns, with a mean of 13.16 microns (Table 4.4). They range in area from 50-275 square microns, with a mean of 125.53. The aspect ratio of 1.19 was the lowest of any analyzed starch, but all of the analyzed starches were similar in this regard.

Figure 4.2 – Top and Profile Views of Lomatium bicolor Starch.
76
Table 4.4 – Length and Area of Lomatium bicolor Starch (μm).
Taxonomic
Identification
Specimen
Form
Reference/Source
Lomatium bicolor
Fresh
Tuber
Length
Range
Mean
Length
Std.
Dev.
Area
Range
Mean
Area
Std.
Dev.
G. Scholze (2009)
Alturas, CA
6 - 23 13.16 2.86 50 - 275 125.53 56.82
Starch Granule Shape and Form – As mentioned, Lomatium starches are collectively similar in shape and form to Perideridia , a closely related member of the carrot (Apiaceae) family. This includes the frequent occurrence of conical, bell, or triangular ovoid-shaped forms. Sub-round to oval and spherical to globose-shaped forms can also occur with Lomatium , but are less frequent.

77
Other Characteristics – Other prominent characteristics of Lomatium include a flat facet on one end of the starch grain, much like that of Perideridia , and the presence of a large hilum vacuole. These vacuoles range in size, but are typically a third of the length of the starch grain. As such, the Lomatium vacuoles are proportionately smaller than those found in Pinus monophylla (see below).
In sum, Lomatium starch grains resemble those of Perideridia , another member of the Apiaceae or carrot family. Further analysis of these and other Apiaceae species will be necessary to determine if the conical bell-shaped form with a flat facet is a distinguishing characteristic for all Apiaceae starches or limited to only certain members of this aboriginally important group of plants. Attributes that appear to distinguish
Lomatium from Perideridia and possibly other members of the carrot family include a large hilum vacuole and typically larger starch grain size.
Sego Lily ( Calochortus sp.)
A total of 50 grains of Calochortus macrocarpus starch from a specimen collected in northeastern California was analyzed. Attributes that appear to distinguish this from other starches analyzed include its grain size, an ovoid, often teardrop-shaped form, a tapered end, and a highly eccentric hilum (Figure 4.3).
Starch Granule Size – Calochortus starches range in length from 8-30 microns, with a mean of 17.34 microns (Table 4.5). In area, they range from 50 to 300 square microns with a mean of 147.47. Apart from size, other attributes that may be useful in distinguishing Calochortus relate to the form of individual starch granules.

78
Table 4.5 – Length and Area of Calochortus macrocarpus Starch (μm).
Taxonomic
Identification
Specimen
Form
Reference/Source
Length
Range
Mean
Length
Std.
Dev.
Area
Range
Mean
Area
Calochortus macrocarpus
Fresh
Tuber
Std.
Dev.
G. Scholze (2009)
Alturas, CA
8 - 30 17.34 3.98 50 - 300 147.47 57.98
Starch Granule Shape and Form – Observation of numerous Calochortus grains indicate that most are symmetrical in cross-sections, occurring in ovoid, ellipsoid, and teardrop forms. Some immature granules are sub-round to spherical, resembling other geophyte granules, but are typically smaller in size than other taxa. In the case of larger grains, however, their teardrop form and tapered end are unique to Calochortus .
Figure 4.3 – Profile and End Views of Calochortus macrocarpus Starch.

79
Other Characteristics – Other prominent characteristics of the Calochortus granules include a highly eccentric hilum (Table 4.6), a strong degree of polarization, and a coil-effect linear surface feature running along the edge of the grain from the tapered end part way into the granule. In polarized view, the arms of the Calochortus grain extinction cross are usually straight, with a small bend or curve toward the margin of the granule.
Table 4.6 – Mean Length, Aspect Ratio, and Hilum
Aspect Ratio for Analyzed Taxon (μm).
Taxonomic
Identification
Mean Length Aspect Ratio
Hilum Aspect
Ratio
Perideridia sp. 11.28 1.28 1.26
Lomatium sp.
Calochortus sp.
Triteleia sp.
Quercus sp.
13.16
17.34
22.16
1.19
1.67
1.42
1.26
3.18
2.06
Pinus monophylla
12.92
6.81
1.59
1.32
1.28
1.22
Given their distinctive characteristics it seems likely that Calochortus starch would be easily identified if recovered archaeologically. Further analysis of other species in the lily family will be necessary to determine whether the teardrop form, tapered ends, and highly eccentric extinction cross arms found in Calochortus are unique to this species and restricted, as currently appears, to mature/larger starch granules.

80
Brodiaea ( Triteleia sp.)
Fifty grains of Triteleia laxa collected by Eric Wohlgemuth (2010) from the
Sacramento River Valley were analyzed for the comparative sample. Although more abundant in the Central Valley, brodiaea is present at higher elevations and well-watered locations within the study area. Diagnostic attributes identified among larger brodiaea starch granules comprise both simple and compound grain types, large grain size, a pear to teardrop-shaped form, knobby protuberances on the end of the granule, eccentric hilum, presence of hilum vacuoles and ring-like lamellae, and web-like fissures when present.
Starch Granule Size – Triteleia starch ranges from 10-30 microns in length (mean
= 22.16 ± 3.24) and from 100-450 square microns in area (mean = 278.79 ± 72.44) (Table
4.7). Other distinguishing traits of these granules are their frequently compound structure that can include up to four grains. Compound granules of this sort were measured in their entirety, because individual grains were difficult or impossible to distinguish.
Table 4.7 – Length and Area of Triteleia laxa Starch (μm).
Taxonomic
Identification
Specimen
Form
Reference / Source
Length
Range
Mean
Length
Std.
Dev.
Area
Range
Mean
Area
Std.
Dev.
Triteleia laxa
Fresh
Tuber
G. Scholze / Colusa
Co., CA
10 - 30 22.16 3.24 100 - 450 278.79 72.44
Starch Granule Shape and Form – Triteleia starches are often laterally symmetrical, occurring in teardrop, pear, ovoid, or sub-round form. Some smaller

81 granules are ellipsoid or irregular globose in form and may resemble other geophyte granules, although they are appreciably larger in size. Larger grains with their teardrop or pear-shaped form, with more restricted neck, and occasionally knobby protuberances on the end of the grain, are unique, however, to this taxon.
Other Characteristics – Other prominent characteristics include a moderately eccentric hilum, a strong degree of polarization, hilum vacuole, pronounced ring-like lamellae, and web-like cracking or fissures on top of the teardrop or pear-shaped examples. In polarized view, the arms of Triteleia grains extinction crosses are usually straight with a slight bend or curve towards the granule margin. Protuberances in the form of knobs at the end of the grain are also found on more irregular, ovoid granules
(Fig. 4.4). These may result from the separation of larger compound granules, the protuberances representing the remains of former attachment areas. In fact, knob-like protuberances never occur on compound granules, lending a certain degree of support to this interpretation.
Whether a single or compound granule and regardless of shape, Triteleia starch grains have a vacuole at the hilum. This was used to determine the number of grains comprising compound granules. Such granules showed no evidence of faceting, but most of the grains had lamellae rings around the hilum. Some were more centric than others, which approximated the form of the granule be it teardrop or pear-shaped.

Figure 4.4 – Top and Profile View of Triteleia laxa Starch.
82
In short, Triteleia starch granules are characterized by a distinctive set of traits that make them readily identifiable from other taxa in the comparative sample. These include their large size, pear or teardrop form, eccentric hilum, presence of lamellae, and knob-like protuberances. Further analysis will be necessary, however, to determine the frequency of these traits and whether other Triteleia species differ in starch morphology.
Oak ( Quercus sp.)
Fifty grains each were examined from fresh acorn nuts obtained from two species of oak found in northeastern California, Quercus garryana and Quercus kelloggii .
Although neither of these trees grows in the project area, their nuts may have been

83 transported into the region from nearby areas. Though a less likely subsistence resource, acorns were analyzed to provide a comparison to geophyte starches. As reported by
Rumold (2010), when various characteristics are combined for the “grain population signature” of acorns, their starch too can be identified (Piperno and Holst 2004). Acorn starch has some of the same characteristics as Perideridia (e.g., length and area), but is appreciably different in shape and form.
Starch Granule Size – Quercus starches range in length from 6-25 microns, with a mean of 12.92 microns. In area, they range from 25-200 square microns, with a mean of
89.58 square microns. This makes these starches slightly larger and more variable than
Perideridia (Table 4.8), providing one of the ways they can be distinguished on the basis of size.
Table 4.8 – Length and Area of Quercus sp. Starch (μm).
Taxonomic
Identification
Specimen
Form
Reference /
Source
Length
Range
Mean
Length
Std.
Dev.
Area
Range
Mean
Area
Std.
Dev.
Quercus kelloggii
Fresh Nut
G. Scholze (N.E.
California)
6 – 25 12.92 3.87 25 - 200 89.58 50.12
Quercus kelloggii
Fresh Nut
Gleason (2001)
Upper Klamath
Valley, California
2.0 – 2.8* 2.27* N/A N/A N/A N/A
Quercus garryana
Fresh Nut
Gleason (2001)
Upper Klamath
Valley, California
2.6 – 7.4* 5.04* N/A N/A N/A N/A
* Gleason (2001) slides were analyzed uncalibrated. These slides were reanalyzed under the calibrated microscope producing different starch grain lengths than those she reported.
Starch Granule Shape and Form – Quercus starches vary in shape, incorporating triangular ellipse, triangular ovate, circular, and teardrop-shaped forms. Larger, more

84 mature grains are typically triangular ellipse or triangular ovate in form, as distinct from the conical or bell-shaped varieties characterizing Perideridia . In three dimensions, most are triangular ovoid or triangular ellipsoid in shape providing another distinctive criterion
(Fig. 4.5).
Other Characteristics – Other distinctive traits of acorn starch are protuberances on the triangular ends of the grain. On larger, mature granules either one or two protuberances are present, sometimes creating the triangular form. Many of the granules also exhibit evidence of lamellae or concentric rings around the hilum that are lacking on
Perideridia .
Figure 4.5 – Profile and End View of a Quercus kelloggii Starch.

85
Pinyon (Pinus monophylla)
Fifty grains of Pinus monophylla starch were analyzed as part of the comparative study. Pinyon does not grow in the study area, but the only species found in reasonable proximity is the single-leaf variety, with other species ( P. edulis and P. quadrifolia ) located in distant parts of the eastern and southern Great Basin. Many features of pinyon starch grains (e.g., size, form, surface, and polarization) differ from those of geophyte and seed crops, making them easy to distinguish. These include the small, irregular grain size, oval discoid to reniform shape, off-centered protuberance wrapping around the starch grain, large open vacuole in the hilum, and curving polarization arms on the shortlegged cross.
Starch Granule Size – Pinus monophylla starches range in length from 4-10 microns, with a mean of 6.81 microns. In area, they range from 10-60 square microns, with a mean of 29.50 square microns (Table 4.9). In comparison to other starches, pinyon starch granules are substantially smaller, with the range of measurements also more restricted.
Table 4.9 –Length and Area of Pinus monophylla Starch (μm).
Taxonomic
Identification
Specimen
Form
Reference /
Source
Length
Range
Mean Std.
Length Dev.
Area
Range
Mean
Area
Std.
Dev.
Pinus monophylla
Fresh Nut
G. Scholze
(Hwy 89, CA)
4 - 10 6.81 1.10 10 - 60 29.50 8.30
Starch Granule Shape and Form – Pinyon granules can be of various shape, but larger, more mature examples tend to be oval or reniform in shape and ovoid, discoidal,

or reniform in form. As with other analyzed starches, smaller, less mature granules are less distinctive displaying spherical or globose shapes (Figure 4.6).
Figure 4.6 – Profile View of Pinus monophylla Starch.
86
Other Characteristics – Other distinguishing characteristics comprise a collar-like protuberance wrapping perpendicularly around three sides of the starch granule and/or an embryo tail-like feature. Some larger spherical grains have a pitted surface texture, but most of the granules are relatively smooth. Another distinctive feature found on 90% of the analyzed grains was an open vacuole within the hilum that reaches three to four microns in length. None of the other starches analyzed possess this large open hilum.
Lastly, the maltese cross on pinyon was centric in hilum orientation, but the legs were rather short and the central spot of intersection elongated, giving the impression of eccentricity.
CONCLUSIONS
In sum, analysis of various starches suggest that larger examples of biscuitroot, brodiaea, epos, and sego lily are morphologically distinctive and readily identifiable, as are certain important seed/nut crops, including acorns and pinyon. Characteristics

87 specific to each of these starches are summarized in Table 4.10. Although many of the starches overlap to some extent in size, when other aspects of their morphology (e.g., form, hilum eccentricity) are considered, most can be readily separated (Figure 4.7).
Other metric traits, such as the minimum polarization angle, seem to be of limited value for identifying different starches, although additional work should be done before they are entirely dismissed.
What emerges from this pilot study is that proper identification of starches needs to rely on a broad array of traits applied to a population of specimens (Piperno and Holst
2004; Rumold 2010) in conjunction with more unique features that distinguishes certain taxa.

88
Table 4.10 – Summary of Diagnostic Starch Grain Traits.
Taxonomic
Identification
Grain Shape and Form
Grain Size (μm)
Hilum, Fissure, and
Lamellae (μm)
Surface and
Margin
Polarization
Cross
Strong
Brodiaea
Triteleia sp.
Simple or
Compound;
Teardrop, Pear,
Irregular Ovoid
Mean Length – 22.16
Mean Area - 278.79
Aspect Ratio - 1.42
Eccentric Hilum - 2.06
Smooth Surface
Hilum Vacuole Present
Facet Absent
Arms Straight
Protuberances
Polarization with slight
Lamellae Present on end curve P-
Cross Angle -
87.2
Strong
Sego Lily
Calochortus sp.
Simple;
Conical, Bell, and Triangular
Ovoid
Mean Length – 17.34
Mean Area - 147.47
Aspect Ratio - 1.67
Eccentric Hilum - 3.18
Smooth Surface
Hilum Vacuole Absent
Facet Absent
Arms Straight
Lamellae Absent
Coil-effect
Polarization with slight
Surface Feature curve P-
Cross Angle -
49.3
Simple;
Biscuitroot
Conical, Bell,
Lomatium sp. and Triangular
Ovoid
Mean Length – 13.16
Mean Area - 125.53
Aspect Ratio - 1.19
Smooth Surface Strong
Centric Hilum - 1.26
Large Hilum Vacuole
Lamellae Absent
Flat Basal
Facet,
Sometimes
Multiple Facets
Polarization
Arms Straight
P-Cross
Angle - 80.6
Epos
Perideridia sp.
Simple;
Conical, Bell, and Triangular
Mean Length – 11.28
Mean Area - 84.18
Aspect Ratio - 1.28
Ellipsoid
Centric Hilum - 1.26
Large Hilum Vacuole
Lamellae Absent
Strong
Smooth Surface Polarization
Flat Basal
Facet,
Sometimes
Arms Straight or slightly curved P-
Multiple Facets Cross Angle -
69.8
Weak to
Oak
Simple;
Triangular
Ovoid or
Quercus sp. Ellipsoid, Teardrop
Mean Length – 12.92
Mean Area - 89.58
Aspect Ratio - 1.59
Centric/ Slightly
Eccentric Hilum - 1.28
Smooth Surface
Facet Absent
Moderate
Polarization
Arms straight
Hilum Vacuole Absent Pointy with slight
Lamellae Present or Protuberances
Absent on ends curve P-
Cross Angle -
69.4
Pinyon
Pinus monophylla
Simple; Ovoid,
Discoid,
Reniform
Mean Length - 6.81
Mean Area - 29.48
Aspect Ratio - 1.32
Centric Hilum - 1.22
Hilum Vacuole Present
Lamellae Present
Smooth to
Irregular
Bumpy Surface
Protuberances
Weak along side,
Polarization
Arms Straight
P-Cross
Angle - 86.7 when Present

89
Start
Figure 4.7 – Northeastern California Starch Grain Key.
Yes
Conical to
Bell-shaped form
With flat facet on
Basal end
(Apiaceae Family)
Yes
No Vacoule; Length < 6 microns
Area < 50 microns
Presence of Large Vacoule
(1/3 length of grain);
Length > 18 microns;
Area > 200 microns
Yes
No
Pear-shaped
To Teardrop
In form
(Liliaceae Family)
Yes
Highly Eccentric; No Lamellae;
When Grain Length ? 15 Microns,
Hilum Aspect Ratio ? 2.5
Yes
No
Knobby Protuberances On end of
Grain; Ring-like Lamellae;
When Grain Length ? 25 Microns,
Hilum Aspect Ratio ? 2.5
Yes
Triangular Ellipse
Triangular Ovate
In Shape
No
Yes
Protuberances on Triangular ends;
Presence of Lamellae
Yes
Perideridia sp.
Lomatium sp.
Calochortus sp.
Triteleia sp.
Quercus sp.
Ovoid, Discoid,
Or Reniform In Shape
Yes
No
Small Size; Length < 10 Microns
Area < 60 Microns
Out of Key
Yes
Pinus monophylla

Figure 4.7 (continued) – Northeastern California Starch Grain Key.
A Key to Northeastern California
1. Starch Grain is Conical to Bell-shaped in Form
(Flat Facet on Basal End)……………………………………… 2 (Apiaceae)
1´.
Starch Grain is not Conical to Bell-shaped in
Form (No Flat Facet on Basal End)…………………………… 3
2. Starch Grain contains Large Vacuole (1/3 length of Grain) and/or
Grain ≥ 18 microns, Area ≥ 200 microns………………………
Lomatium sp.
2´. Starch Grain contains No Vacuole and/or
Grain ≤ 6 microns, Area ≤ 50 microns…………………………
Perideridia sp.
3. Starch Grain is Pear-shaped to Teardrop in Form………………4 (Liliaceae)
3´.
Starch Grain is not Pear-shaped to Teardrop in Form…………..5
4. Starch Grain does not hold Lamellae; Highly Eccentric;
When Grain Length ≥ 15 Microns, Hilum Aspect Ratio ≥ 2.5...
Calochortus sp.
4´.
Starch Grain holds Ring-like Lamellae; Knobby Protuberances
On end of Grain; When Grain Length ≥ 25 Microns,
Hilum Aspect Ratio ≤ 2.5……………………………………… Triteleia sp.
5. Starch Grain is Triangular Ellipsoid or
Triangular Ovoid in Form……………………………………… 6
5´. Starch Grain is Not Triangular Ellipsoid or
Triangular Ovoid in Form……………………………………… 7
6. Protuberances on Triangular Ends of Grain;
Possible Presence of Lamellae…………………………………
Quercus sp.
6´. Other…………………………………………………………… Out of Key
7.
Starch Grain is Ovoid to Discoidal or Reniform in Form……… 8
7´.
Starch Grain is not Ovoid to Discoidal or Reniform in Form…..Out of Key
8. Starch Grain is Small; Length < 10 microns; Area < 60 microns
Open Vacuole Length = 3 – 4 microns……………………… Pinus monophylla
8´. Other…………………………………………………………… Out of Key
90

91
Chapter 5
METHODS AND RESULTS OF ARCHAEOLOGICAL STARCH ANALYSIS
This chapter presents the methods and results of archaeological starch analysis on ground and flaked stone tools from the Pit River Uplands, Madeline Plains, and Secret
Valley, California. It begins with an outline of the sampling procedures, chemical extraction of starch residues, and analysis of archaeological starch grains. This is followed by a discussion of the results of the starch grain identification employing the reference material discussed in the previous chapter.
Results of the analysis indicate that starches preserved on previously excavated stone tools can be reliably retrieved and identified, providing previously unobtainable information on prehistoric subsistence patterns, tool use, and other behavior. Thirty-four of 35 analyzed ground and flaked stone tools produced a total of 667 starch grains, demonstrating the efficacy of the technique. Although some of the grains could not be identified, others were readily identifiable to specific genera or families. These include
Perideridia , Lomatium , Quercus , Triteleia , and more generic examples belonging to the carrot or Apiaceae family (i.e., Perideridia / Lomatium ).
SAMPLE SELECTION
The artifacts examined for starch grains derive from six archaeological sites in the
Pit River Uplands, Madeline Plains, and Secret Valley on the Modoc Plateau (Delacorte
1997a, 1997b; McGuire 1997, 2000a). All of the sites contained components dating to

92 the Late Archaic/Terminal Prehistoric period (circa 1300 B.P. to contact) and sometimes earlier with all examined artifacts found in well-dated context. The site settings vary from open, stream terrace to ephemeral playa habitats, but the preservation of starch grains should be similar at all the localities. All sites were originally excavated as part of the Alturas Transmission Line (McGuire 2000a) and Tuscarora Gas Transmission Line projects (Delacorte 1997a; McGuire 1997), the artifacts subsequently curated at the
Nevada State Museum in Carson City.
ARTIFACT SELECTION
Proper selection of an artifact sample must consider the ethnographic and biogeographic context of the research area. This includes the type and distribution of geophytes and other plant resources. This information was used to generate a list of plants for inclusion in the comparative starch grain collection for the study area.
Differences in the way roots were processed also need to be considered. Some roots required minimal processing (e.g., Triteleia sp. or Perideridia sp.) and are correspondingly harder to identify in the archaeological record. Other root crops (e.g.,
Camassia sp. or Allium sp.) contain an indigestible carbohydrate, inulin, which is produced in place of a starch granule and necessitates extensive processing (i.e., roasting) prior to consumption. As such, the archaeological signature of these later plants should be more apparent, producing roasting ovens and various flaked and ground stone tools.
From a practical standpoint, ethnographic, biogeographic, and technological factors need

93 to be considered to establish the archaeological signatures of geophyte use and how starch was entering the archaeological record.
Both flaked and ground stone tools were employed in root processing and examined as part of this study. This includes milling equipment such as millingstones, mortars, and handstones and other implements for the peeling or cleaning of roots (e.g., flake-based tools, bifaces, and broken projectile points). Flaked and ground stone artifacts from both the Alturas Intertie and Tuscarora projects exhibited organic residues that may preserve starch grains. Starch grains can enter the archaeological record in various natural and cultural ways, but analysis of plant processing tools should help to narrow the pathways responsible for their introduction.
With the preceding in mind, a representative sample of tools was analyzed from sites in each of the environmental sub-regions (Pit River Uplands, Madeline Plains,
Secret Valley). A total of 30 ground and five flaked stone tools was processed (Table
5.1). Because the processing and analysis time for each of these artifacts is quite high, only three to eleven tools were examined from each site. These included both grinding and scraping implements that were employed during different stages of plant processing or to process different kinds of resources.

94
AREA
Pit River Uplands
MOD-3150
MOD-3153
MOD-3448
Madeline Plains
LAS-1623
Secret Valley
LAS-206
LAS-1705
Table 5.1 – Summary of Analyzed Tools.
Hndst Mllgst Mor Pstl
2
2
2
2
2
2
1
2
2
Flkst
5
# OF ARTIFACTS
4
3
11*
4
3
2
2 1
1
2
2
8
5
9 5 35 Total # of Artifacts 13 6 2
* Includes five flaked stone artifacts
METHODS
The methods used to extract and analyze starch were adapted from Fullagar et al.
(1998), Horrocks et al. (2004), Lentfer et al. (2002), Rumold (2010), and Therin (1994).
The extraction of starch from tool surfaces is a multi-step process. After starch grains are recovered from tool crevices, they must be isolated from adhering soil and other residue, and concentrated for viewing under a polarizing transmitted light microscope for identification. Ultrasonic cleaning of artifact surfaces is the best way to recover starch grains. This can be done in either an ultrasonic washing tank or with an ultrasonic toothbrush.

95
Unlike artifacts brought directly from the field for processing, those in curated museum collections have generally been washed and analyzed before. This can produce contamination from previous handlers and/or storage, requiring a thorough rinse with distilled water before the artifact is ultrasonically cleaned. To prevent new contamination, un-powdered latex gloves were worn when handling and processing artifacts and starch samples, and all equipment (beakers, centrifuge tubes, etc.) thoroughly cleaned between each use.
Heavy Liquid Starch Extraction
Starch extraction included the ultrasonic cleaning of artifacts, starch residue concentration using heavy density liquid flotation (i.e., Sodium Polytungstate), and dilution of the flotation medium for final sample preparation.
Ultrasonic cleaning - Machine Method
The first step in sample preparation was to place the artifact in a clean glass beaker of appropriate size and cover it with distilled water. The beaker was placed in an ultrasonic washtub filled with regular tap water to the same level as the distilled water in the beaker. The sonic bath was then operated for approximately 10 minutes to remove any dirt, starch grains, or other residues adhering to the artifact. The longer the tool is cleaned, the greater the potential for the recovery of starch grains (Perry 2001:110;
Rumold 2010:259). If the entire artifact could not be submersed, it was rotated and the process was repeated so that every part of the artifact was cleaned for the allotted time.

96
When specific artifact surfaces like the grinding facets on milling tools were the target of analysis, other surfaces were kept out of the ultrasonic bath, to minimize contamination by starches potentially unrelated to the tool’s original or primary use. This may occur from contact with the surrounding soil and/or post-depositional contamination before or after excavation. Artifacts were periodically repositioned in the beaker during the washing process because the distribution of sound waves is uneven within the bath.
Ultrasonic Cleaning - Manual Method
An alternative to the ultrasonic bath was to use an ultrasonic toothbrush. This was typically done for one of two reasons: the artifact was too large to fit in the ultrasonic tank or specific artifact surfaces like the grinding facet on a millingstone were being targeted. When possible, the artifact or surface of interest was submersed in distilled water and scrubbed with the ultrasonic toothbrush for approximately 10-
20 minutes. When the artifact could not be submersed, it was continuously rinsed with distilled water while it was brushed.
Following the ultrasonic cleaning, the tool and latex gloves were rinsed with distilled water over the beaker or container to recover any adhering particles. The tools were then dried and re-inked to Nevada State Museum standards and re-bagged.
Starch Residue Concentration
After washing, the aqueous sediment in the beaker was transferred to clean, 50 ml centrifuge tubes and centrifuged for ten minutes at 2000 rpm. The supernatant or

97 wastewater on the surface of each tube was decanted down to approximately 10 ml and the sediment residue consolidated into as many tubes as necessary, rinsing each tube of adhering sediment with distilled water. The centrifuging and decanting procedure was repeated until the entire sample was concentrated into a 15 ml centrifuge tube.
Heavy Liquid (Sodium Polytungstate) Flotation
The heavy liquid solution was prepared by mixing Sodium Polytungstate
(Na
6
[H
2
W
12
O
40
]) with distilled water to a specific gravity of 2.0 g/cm
3
. Sodium
Polytungstate was chosen over Cesium Chloride because of its non-caustic nature.
Although many starch studies have used solutions with a density of 1.8 g/cm
3
, a heavier
2.0 g/cm
3
solution was employed to insure that all of the starch was retrieved. Before the
Sodium Polytungstate was added, the 15 ml centrifuge tube was centrifuged for five minutes at 2000 rpm and as much water as possible decanted. With some liquid remaining in the tube, the higher specific gravity of 2.0 g/cm
3
solution allows the Sodium
Polytungstate liquid to remain above 1.8 g/cm
3
. Five milliliters of Sodium Polytungstate was added to the centrifuge tube, capped, shaken to mix the solution with the sediment, and centrifuged for five minutes at 2000 rpm. The supernatant light fraction liquid floating on the surface was drawn off the top with a pipette. Another 5 ml of heavy liquid was added to the tube and the centrifuging and decanting processes repeated.

98
Sodium Polytungstate Dilution
The supernatant and heavy liquid extract was split between two 15 ml centrifuge tubes. Each of the tubes was filled with distilled water to reduce the specific gravity.
The solution was vortexed and centrifuged for 10 minutes at 2000 rpm. Half of the supernatant was drawn off the top with a pipette and discarded. These steps were repeated twice more, first drawing off half of the supernatant followed by the majority of the liquid, leaving less than 1.5 ml of residue sediment to be placed in a 1.5 ml vial for storage. All of the discarded supernatant containing Sodium Polytungstate was placed in a sealed container to later be recycled.
Microscope Analysis
Once extracted and separated from the sediment, archaeological starches were microscopically analyzed. The methods for preparing starch residue microscope mounts, scanning of slides with both brightfield and darkfield illumination, and the recording and identification of starches are reviewed below.
Slide Mounting
Microscope slides were prepared one of three ways. To begin, a drop of starch extract was placed on the microscope slide with a pipette and a glass cover slip was placed on top with a drop of clear nail polish in the corner to hold the cover slip in place.
As noted by others (Dickau 2005; Rumold 2010), only a drop of solution should be placed on the slide, as any more simply leaks out and is unnecessarily lost. As the slide is

99 microscopically scanned, the distilled water-mounting medium begins to dry out making further analysis impossible. This can be remedied by pipetting a small amount of distilled water along the edge of the cover slip to rehydrate the sample.
An alternative method followed by Rumold (2010) allows for longer suspension time and a more permanent slide mount. A drop of starch extract was placed on the microscope slide and left to dry for approximately five minutes, or until the extract solution was almost completely dry. A drop of glycerol was added to the semi-dried extract and a glass cover slip was placed on top. Permount was used to seal the cover slip to the microscope slide, so the slide could be scanned and stored flat for future analysis.
Another cheaper option for a mounting medium is Karo syrup. This was used initially and found to be rather thick; even after being thinned, starches could not be rotated or tumbled effectively.
Slide Scanning
Prepared slides were placed under a compound transmitted light microscope.
Scanning was performed under a darkfield illumination through cross-polarization.
Starch granules were illuminated against a dark background, with the trademark
“maltese” or extinction cross identifying starch granules from other organic particles. As
Rumold (2010) noted, some starch grains will be missed due to the strength or weakness in the cross-polarization of ancient archaeological specimens. This nevertheless proved to be the most effective method, insofar as the scanning of each slide requires one to two hours depending on the number of starch grains present.

100
Slide scanning was performed at 100x power and increased to 400x when starch grains were located, the higher magnification permitting better description and identification of the starches. The process was started in the upper-left corner of the coverslip, working in transects across the slide from left to right. When the opposite side of the cover slip was reached, the subsequent transect scanning back across the slide was from right to left and two fields of view down the slide. This was done to insure that the same starch grains would not be recorded twice, given the movement of grains within the mounting medium. This process was continued until the bottom of the cover slip was reached. When completed, 50% of the solution placed on the microscope slide was thoroughly scanned for any starch grains present.
Starch Identification
As starch grains were observed, the microscope field of view was replaced by brightfield illumination to better examine and describe the distinguishing traits of each granule. All identified granules were photographed using a Sanyo color CCD digital camera mounted on the microscope. Each starch grain was described and sketched in a notebook and this information later transferred to an Excel spreadsheet. Descriptive attributes recorded for the starch grains duplicated those developed for the modern reference collection (refer to Table 4.1).
Several photographs of each granule were typically taken using polarized, unpolarized, and red filtered polarized light prior to the written descriptions, in case the grain moved out of view. When photographed, the cover slip was lightly pressed with a

101 probe to roll the granule to observe and record the three-dimensional form. Descriptions were recorded and the granule was given a unique catalog number. Some grains were strongly polarized and easily rolled requiring fewer photographs, while others needed numerous photographs from various angles to properly document their morphology. As much information as possible was recorded for each granule, insofar as relocating specific grains on the slide for reexamination is difficult or impossible. Slides can be reanalyzed, but once the mounting medium has dried, the granules are locked in place and cannot be rotated.
Starch grains from the northeastern California artifacts were identified using the modern reference collection described in the previous chapter. Preliminary identification of starches was made when they were recorded and the photographic and descriptive data rechecked when the collection was completed. Individual starch grains were identified to the finest taxonomic level possible based on the attribute key developed from the modern reference material. Some granules had few diagnostic traits matching the comparative collection and were categorized as “unknown,” highlighting the need to expand the comparative collection during future studies. In addition to the “unknown” category were starches that displayed only some of the traits of broader taxonomic divisions that could be identified only to family (e.g., Apiaceae), not specific species or genera. These are, however, useful in identifying the range of plant or resource types exploited in different archaeological contexts.

102
QUANTIFICATION OF RESULTS
As noted by previous researchers (e.g., Haslam 2004; Rumold 2010), many factors affect the recovery of starch. Some of these include starch grain integrity, tool attributes, and the soil condition artifacts and/or starch were deposited in. Other postdepositional factors can also affect the recovery of starch, including the excavation and curation methods, and laboratory extraction techniques. Each of these factors influence starch grain recovery differently, such that the present research focuses on the presence/absence of different plant taxa, more than the quantification of starch granules.
RESULTS
All 30 ground stone artifacts tested for starch generated positive results, with a total of 655 starch granules recovered. Of these, 30 (4.6%) were identified as epos
( Perideridia sp.), 10 (1.5%) as biscuitroot ( Lomatium sp.), 15 (2.3%) as brodiaea
( Triteleia sp.), and 6 (0.9%) as acorn ( Quercus sp.) (Table 5.2). Another 58 granules
(8.9%) were more conservatively identifiable as Apiaceae (carrot family) starch. These lack sufficient diagnostic attributes to be definitely identified as either Perideridia sp. or
Lomatium sp., but are clearly one or the other. When both family and genus specific identifications are combined, Apiaceae starch accounts for 15% of the documented granules. Many of the unidentified grains were too small or damaged to have distinguishing features, and remained unidentifiable. Others were almost certainly identifiable, given their appearance, but not represented in the comparative sample and they could not be identified.

103
In addition to the ground stone artifacts were five flaked stone tools (biface, flake tool, formed flake tool) analyzed for starch. Four of these five tools from the Pit River
Uplands (CA-MOD-3153) produced a total of 12 unknown starches. All were thoroughly documented and may be identified in the future; they are excluded from the ground stone results discussed below.
Table 5.2 – Summary of Northeastern
California Starch Grain Results.
Perideridia
# Grains % Grains
30 4.6%
Lomatium
Apiaceae
Triteleia
Quercus
Unknown
Total
10
58
15
6
536
655
1.5%
8.9%
2.3%
0.9%
81.8%
100.0%
Perideridia Starch Grain Identifications
Thirty of the starch grains (4.6%) were identified as Perideridia . All but one of these are from artifacts in either the Pit River Uplands or Secret Valley, just a single example from the Madeline Plains. As with the comparative sample, the prehistoric grains are distinguished by a conical/bell-shaped form, flat facet on the basal surface, centric, closed hilum with no vacuole, and moderate to strong polarization arms and smooth surface.

104
Table 5.3 – Average Size, and Aspect Ratio of Perideridia (μm).
Mean
Length
Length
Range
Mean Aspect
Ratio
Mean
Area
Area
Range
Mean Hilum
Aspect Ratio
11.52 8 – 18 1.18 94.62 40 - 230 1.26
As Table 5.3 shows, both the size and form of the archaeological Perideridia starches are consistent with those in the reference collection. The mean length and area are slightly larger, but this may reflect nothing more than the fact that grains at the larger end of the size spectrum are more easily found and identified. The aspect ratio (1.18 μm) is similar to the comparative sample (1.28 μm), and the hilum aspect ratio (1.26 μm) identical in both cases. As previously stated, the most distinctive characteristic between
Perideridia and Lomatium is a large vacuole at the hilum of the latter. Nearly a quarter
(n = 7) of the Perideridia starch grains had a vacuole, but all were substantially smaller than those on Lomatium . Figure 5.1 depicts an archaeological Perideridia starch identified by its distinctive conical/bell-shaped form and flat basal facet.
Figure 5.1 – Perideridia Starch #1705-493-17 (magnification 400x).

105
Lomatium Starch Grain Identifications
Ten grains (1.5%) of Lomatium starch were identified from the Pit River Upland area. All are superficially similar to Perideridia , in that both are members of the
Apiaceae (carrot) family. Here again, the grains are simple conical/bell-shaped in form, with an open hilum and large vacuole, smooth surface, single or multiple facets, and moderate to strong polarization arms.
Table 5.4 - Average Size, and Aspect Ratio of Lomatium (μm).
Mean
Length
(μm)
Length
Range
Mean Aspect
Ratio
Mean
Area
Area
Range
Mean Hilum
Aspect Ratio
17.60 14 - 22 1.17 210.70 140 - 305 1.26
As with Perideridia , archaeological examples of Lomatium are, on average, longer (17.60 μm) and larger (210.70 μm) than their modern counterparts in the comparative collection (13.16 and 125.53 μm, respectively). All fall, however, within the length range of the reference sample. Much the same applies to the range in the area of the archaeological specimens (140-305 μm) that is greater than the modern sample, though only two of the prehistoric grains exceed the area of modern Lomatium starches.
The aspect ratio (1.17 μm) and hilum aspect ratio (1.26 μm) measurements of the archaeological starches mirror those of the modern reference collection, as does the uniform presence of large vacuoles, which distinguish these from Perideridia starch.
Figures 5.2 and 5.3 depict two archaeological Lomatium starch grains, with their conical/bell-shaped form, flat basal facet, and large vacuole.

Figure 5.2 – Lomatium Starch #3153-210-2 (magnification 400x).
106
Figure 5.3 – Lomatium Starch #3153-1547-12 (magnification 400x).
Triteleia Starch Grain Identifications
Fifteen grains of Triteleia starch were recovered. Eleven (73%) of these were extracted from ground stone artifacts excavated in Secret Valley. The four remaining specimens are evenly distributed between the Pit River Uplands and the Madeline Plains.
Triteleia starch consists of simple grains of ovoid to teardrop shape with a closed hilum, no vacuole, smooth-surface and eccentric hilum with moderate to strongly polarized arms.
Archaeological Triteleia starch grains are, on average, shorter (18.46 μm) and smaller (209.02 μm) than the modern examples in the comparative collection (22.16 and
278.79 μm, respectively). Their length and area, however, are generally consistent with

107 the reference collection, except for one of the prehistoric grains that has an area of 94.29 square μm. The aspect ratio (1.33 μm) and hilum aspect ratio (2.04 μm) measurements are likewise consistent with the modern sample, albeit slightly smaller (Figure 5.5).
Table 5.5 - Average size, and Aspect ratio of Triteleia (μm).
Mean
Length
Size
Range
Mean Aspect
Ratio
Mean
Area
Area
Range
Mean Hilum
Aspect Ratio
18.46 11 - 27 1.33 209.02 90 - 415 2.04
Eleven (73%) of the archaeological Triteleia starch grains had protuberances either on the side or end of the grain, which may have been present on the others, but impossible to see at the magnification employed or rotation of these particular grains.
Figures 5.4 and 5.5 depict archaeological Triteleia starches identified by their teardrop shape and eccentric hilum.
Figure 5.4 – Triteleia Starch #1623-400-6 (magnification 400x).

Figure 5.5 – Triteleia Starch #1705-493-14 (magnification 400x).
108
Quercus Starch Grain Identifications
Six grains (0.9%) of acorn ( Quercus sp.) starch were identified. All are from
Secret Valley sites at the southern end of the study area. One grain was found on a ground stone artifact at CA-LAS-1705 and the remaining five on milling tools from CA-
LAS-206. Quercus starch grains are simple triangular ovoid to teardrop in shape. They have a centric, closed hilum with generally no vacuole, pointy protuberance on the grain ends, and moderate polarization that reveals the smooth surface of the grains.
Five of the present grains are triangular ovoid in form with pointy protuberances at the ends (Figure 5.6). The only exception (#1705-1334-5) is a teardrop-shaped grain that has a small hilum vacuole and cracking or fissure through the hilum. These features may be attributable to physical wear (grinding) or heat treatment (cooking) of the starch
(Henry et al. 2009; Messner and Schindler 2010). As reported in Table 5.6, all but one of the archaeological Quercus starch grains matches the metric attributes of the modern reference collection, the one exception (#206-2514-19) being of somewhat greater area, though similar length.

Table 5.6 - Average Size, and Aspect Ratio of Quercus (μm).
Mean
Length
Size
Range
15.92 13 – 23
Mean Aspect
Ratio
1.42
Mean
Area
Area
Range
142.37 110 - 240
Mean Hilum
Aspect Ratio
1.16
Figure 5.6 – Quercus Starch # 206-2514-19 (magnification 400x).
109
Apiaceae (Carrot Family) Starch Grain Identifications
Fifty-eight starch grains (8.9%) from the various study localities were identified as Apiaceae. These presumably incorporate both Perideridia and Lomatium and possibly other members of the carrot family that have some of the same characteristics. Attributes common to the Apiaceae family include conical/bell-shaped forms, centric hilum with and without hilum vacuoles, flat basal facet, smooth surface, and moderate to strong polarization. Forty-eight (82.8%) of the Apiaceae grains were retrieved from artifacts in
Secret Valley and eight (13.8%) from tools in the Pit River Uplands. The former probably represent Perideridia starch grains that lack definitively diagnostic traits, since
Lomatium is absent in Secret Valley, but Perideridia reasonably common. Both
Perideridia and Lomatium are equally common in the Pit River Uplands, however, and

110 thus the Apiaceae grains from there might be either of these plants. Three of the Pit
River Upland’s grains are, however, larger than the range of Perideridia and might be attributed to Lomatium on the basis of grain size. Table 5.7 provides the average measurements for the archaeological Apiaceae starch granules.
Table 5.7 - Average Size, and Aspect Ratio of Apiaceae (μm).
Mean
Length
Size
Range
Mean Aspect
Ratio
Mean
Area
Area
Range
Mean Hilum
Aspect Ratio
13.86 11 - 23 1.31 129.04 60 - 315 1.14
As might be expected if multiple genera are represented, the average measurements for the Apiaceae grains span a wider range of dimensions than those for an individual genus. Figure 5.7 portrays the largest example of an Apiaceae grain, with a length of 22.37 μm and an area of 311.31 square μm. It is conical/bell-shaped in form, with a basal facet and no (large) vacuole, as distinguishes Lomatium , but is substantially larger than Perideridia . Its large size may reflect the time of year the plant was harvested, as starch grains can vary in size depending on the time of year (see Chapter 3).
Ethnographically, most Native American groups collect geophytes in the early spring when they are largest (Fowler 1986; Garth 1978; Housley 1994). This coincides with the seasonal swelling of starches in the underground storage organs. Overall, ninety-eight
(82.4%) of the 119 taxonomically attributable archaeological starches were identified as
Apiaceae, demonstrating the importance of these plant resources.

Figure 5.7 – Apiaceae Starch #1705-1333-6 (magnification 400x).
111
Unknown Starch Grain Identifications
The unknown category is the largest group of starches comprising 536 (81.8%) of the starch grains identified (Table 5.8). These grains are either unidentified or sufficiently damaged that they lack distinguishing morphological and/or metric attributes to determine their taxon. The unknown category includes starches of various size and form, the more prominent of which are ovoid/triangular ovoid, lenticular, globose/spherical, and irregular shapes.
Unknown Ovoid/Triangular Ovoid Starch Grain Identifications
A total of 162 ovoid/triangular ovoid starch grains were identified. They are highly variable in both morphological and metric attributes – more than the variability within or between taxa. This suggests that many of these specimens consist of immature grains of several analyzed and/or unanalyzed plant taxa.

112
Table 5.8 - Unknown Starches (N = 536).
Pit River
Uplands
Madeline
Plains
Secret
Valley
Ovoid/ Triangular Ovoid
Lenticular
Globose/Spherical
Irregular
Other
Total
48
21
20
10
24
123
5
3
23
11
1
3
103
29
61
75
122
390
Unknown Lenticular Starch Grain Identifications
Total
162
51
84
90
149
Fifty-one starch grains were classified as lenticular forms. Most were from artifacts in either the Pit River Uplands or Secret Valley. As with other unknown categories, lenticular grains are highly variable in size, ranging from 80 to 800 square μm in area. Morphologically, they are smooth-surfaced simple grains of circular shape, moderately to strongly polarized, and have a centric hilum with straight polarization arms. None of the identified plant taxa approach this form.
Unknown Globose/Spherical Starch Grain Identifications
Eighty-four of the unidentified grains were classified as globose or spherical in form. These were recovered in all of the study localities. As with ovoid/triangular ovoid forms, globose/spherical shapes can be found in all of the analyzed plant taxa. They are typically found at the lower end of the size spectrum and are presumed to be immature granules that have yet to develop the diagnostic characteristics of fully developed grains.

113
Other Unknown Starch Grain Identifications
In addition to other unknown starches were 239 grains of various forms. Specific morphologies occur in typically limited numbers and include irregular, diamond, discoid, rectangular, reniform, and square shapes. Some of these do not conform to any particular shape or size and might reflect immature, damaged or unknown grains from any of the analyzed or other plant taxa. Included in this category are 124 (18.9%) clearly damaged grains that are both morphologically and metrically indiscernible. These may be damaged for various reasons including cultural behaviors (grinding, roasting), natural processes (temperature fluctuations, soil conditions), and laboratory activities (extraction, rehydration).
Of these unidentified starch grains, 166 (31%) are lenticular, discoid, or irregular in form, forms not found in the comparative sample and 124 (23%) are highly damaged and more problematic in identification. The remaining 246 (46%) are ovoid/triangular ovoid, globose, or spherical in form. These forms are found as immature grains of the analyzed plant taxa comparative sample (Table 5.9). This highlights the possible increase of analyzed plant taxa based on the additional indistinguishable forms.
Comparative
Sample Forms
Non-Comparative
Sample Forms
Table 5.9 – Unknown Starch Forms.
Pit River
Uplands
Madeline
Plains
Secret
Valley
68
37
14
7
164
122
Damaged Grains 18 2 104
Total
246
166
124
Total 123 23 390

114
PIT RIVER UPLAND STARCH GRAIN DATA
Thirteen ground stone artifacts from three different sites (CA-MOD-3150, CA-
MOD-3153, CA-MOD-3448) in the Pit River Uplands were analyzed for starch (Figure
5.9). These produced a total of 154 starch grains. As Table 5.12 indicates, 11 grains were identified as Perideridia (7.1%), 10 (6.5%) as Lomatium , and two (1.3%) as
Triteleia . In addition to these were eight (5.2%) grains attributable to the Apiaceae family that lacked the diagnostic traits to be identified to a particular genus. Most of these are probably Perideridia or Lomatium , more definitive examples of which were identified at all the Pit River Uplands sites. If both genus specific and family-level identifications are combined, Apiaceae starch represents 18.8% of the Pit River Uplands sample. Apart from the Apiaceae and traces (1.3%) of Triteleia starch, 123 (79.9%) of the Pit River Upland starch grains remain unidentified. Of these unidentified starch grains, 68 (44%) are a more generalized immature form and 18 (12%) are damaged, resulting in only 37 (24%) accepted as non-comparative sample forms. This highlights the need to expand the existing comparative collection, such that hopefully more of these grains will be identifiable in the future.

Figure 5.8 – Select Pit River Artifacts.
115

116
MADELINE PLAINS STARCH GRAIN DATA
Four ground stone artifacts from CA-LAS-1623 on the Madeline Plains were analyzed (Figure 5.9). These produced a total of 28 starch grains (Table 5.12). Of these, one (3.6%) was identified as Perideridia , two (7.1%) as Triteleia , and two (7.1%) as
Apiaceae. Given the lack of Lomatium and at least some Perideridia , the Apiaceae grains are more likely Perideridia , though it would be difficult to confirm this. That said, if the Perideridia and Apiaceae starch are combined, the total for Apiaceae climbs to
10.7% of the Madeline Plains starch grains. This is only marginally greater than Triteleia
(7.1%), comparatively little of which was found in the Pit River Uplands. In addition to the identified Madeline Plains starches were a substantially greater number (n=23) of unidentified grains that account for the bulk (82.1%) of the sample. Of these, 14 (61%) are forms found in the comparative taxa sample and two (9%) are damaged beyond recognition. This allows only seven (30%) to be identified as a form not found in the analyzed plant taxa.

Figure 5.9 – Madeline Plains Artifacts.
117

118
SECRET VALLEY STARCH GRAIN DATA
Thirteen ground stone artifacts from two sites (CA-LAS-206, CA-LAS-1705) in
Secret Valley were analyzed for starch, yielding a total of 473 grains (Figure 5.10).
These include 18 (3.8%) identified as Perideridia , 11 (2.3%) attributable to Triteleia , six
(1.3%) with the morphology of Quercus , and 48 (10.1%) identified to the broader
Apiaceae category (Table 5.12). As before, the local presence of Perideridia and lack of
Lomatium suggest that most, or all of the Apiaceae starch is probably the former, though it cannot be confirmed. When the Apiaceae and Perideridia starch are combined, the
Apiaceae family accounts for 14% of the Secret Valley sample. Secret Valley is the only study area to produce Quercus starch, which is probably due to the closer proximity of oak trees that can be found 50 km to the southwest. In fact, charred Quercus macrofossils have been recorded in archaeological deposits of similar age in the Honey
Lake Basin (CA-LAS-1676/H, CA-LAS-1756) just 30 km to the south (McGuire 2000b,
Milliken and Hildebrandt 1997).

Figure 5.10 – Select Secret Valley Artifacts.
119

120
TOOL CATEGORY STARCH GRAIN DATA
When the starch grains data are examined by tool category (handstone, pestle, millingstone), several interesting patterns emerge. Most 371 (56.6%) were recovered from pestles, somewhat fewer 212 (32.4%) from handstones, and the lowest number 72
(11%) from millingstones (includes millingstones, mortars, and hopper mortar). These data indicate that most (89%) of the starch grains were recovered from handstones and pestles (Table 5.10). When the number of starch grains is adjusted per tool type, there are 80% more found on handstones than millingstones and more than twice as many grains on pestles as handstones (Table 5.11).
Perideridia
Lomatium
Apiaceae
Triteleia
Quercus
Unknown
Total
Table 5.10 – Starch Grain Results by Tool Type.
Handstones Millingstones
19 -
3
8
1
1
4
-
1
-
178 69
212 / 32.4% 72 / 11.0%
Pestles
11
6
49
10
6
289
371 / 56.6%
Total
30
10
58
15
6
536
655
Table 5.11 – Number of Starch Grains per Artifact.
Tool Type
Starches Per
Artifact
Handstones 16.3
Millingstones
Pestles
9.0
40.4

121
This suggests that root processing, or evidence of it, relied more extensively or is better preserved on handstones and pestles than millingstones. In fact, all of the
Perideridia and Lomatium starch grains were recovered from handstones and pestles, save a single Lomatium grain (#3150-210-2) from a millingstone. In a broader sense,
98% of all Apiaceae or carrot family starch grains were found on the surfaces of either handstones or pestles. Similarly, all of the Quercus starch grains came exclusively from pestles that were traditionally used to process this resource. The presence of starch on some but not other artifact categories may relate to various things. These include plant processing techniques, plant taxa and tool type relationships, and differential preservation of starches on certain tool categories.
SUMMARY OF RESULTS
The analysis of archaeological starch grains from late prehistoric sites in northeastern California indicates the processing and use of various plant taxa. These include Perideridia , Lomatium , Triteleia , and Quercus . As discussed in the previous chapter, analysis of modern geophytes and seeds disclosed that starches from many wild root crops are distinguishable on the basis of a few distinctive traits. All of the archaeological starch grains identified were morphologically and metrically consistent with modern plant samples. Both the preservation and recovery of ancient starches are influenced by various factors, but differences in the number and/or proportion of various plant taxa can be identified on a regional scale.

122
In all, 119 (18.2%) starch grains were identifiable at some level, with the number and proportion varying between different regional contexts (Table 5.12; Appendix B).
Perideridia , Triteleia , and Apiaceae starch grains were the only taxa ubiquitous across all areas examined. This probably reflects their widespread use throughout northeastern
California. Epos ( Perideridia sp.) was the most abundant starch (7.1%) in the Pit River
Uplands, followed by biscuitroot ( Lomatium sp. [6.5%]). This is in keeping with the abundance of these species in this higher elevation setting. Only one Perideridia and two
Apiaceae starch grains were identified from the Madeline Plains (Table 5.12). This suggests that Perideridia may have been less abundant at this locality and other geophyte or plant resources targeted.
Table 5.12 – Number of Starches by Tool Type in each Study Area.
Madeline Plains
Handstones
Millingstones
Total
Pit River Uplands
Handstones
Millingstones
Pestles
Total
Secret Valley
Handstones
Millingstones/
Mortar
Pestles
Total
Grand Total
Perideridia Lomatium Triteleia Apiaceae Quercus Unknown
1 - 1 1 - 12
# of
Starches
15
- - 1 1 - 11 13
1
11
-
-
11
7
-
11
18
30
-
3
1
6
10
-
-
-
-
10
2
-
-
2
2
3
-
8
11
15
2
1
-
7
8
1
-
47
48
58
-
-
-
-
-
-
-
6
6
6
23
85
5
33
123
84
53
253
390
536
53
325
473
655
28
105
6
43
154
95

123
Brodiaea ( Triteleia sp.) is a widely available resource, with different varieties growing throughout California. It does not appear, however, to have been collected extensively in the study area the way that Perideridia was. Triteleia starch grains were most abundant in the Madeline Plains sample (7.1%), and comparatively rare in the other study areas. This matches the dearth of Perideridia , Lomatium , and Apiaceae starch from the Madeline Plains and suggests that Triteleia may have been more abundant and heavily exploited in this environment.
Acorn ( Quercus sp.) starch grains were only found in Secret Valley. All were extracted from three pestles from the two analyzed sites (CA-LAS-206, CA-LAS-1705).
That Quercus starch was recovered exclusively from pestles matches the traditional association of intensive acorn use with mortar/pestle technology. Oak trees are not found in any of the study areas including Secret Valley, but are present fifty kilometers to the southwest near the western edge of Honey Lake and Susanville area. This suggests that acorns or the tools used to process them were transported to Secret Valley, probably the former given that the implements would have been reused at their destination.
On the whole, there is an impressive shift in archaeological starches between each of the study localities (Figure 5.11). Certain plant taxa are specific to each region, such as Lomatium in the Pit River Uplands, Quercus in Secret Valley, and most of the
Triteleia on the Madeline Plains. This suggests that the frequency of identified starch grains largely corresponds with the local availability of plant resources in different settings.

124
Figure 5.11 – Frequency of Identified Starches between Study Localities.
Percentages of Identified Archaeological Starches
12.0%
10.0%
8.0%
6.0%
4.0%
2.0%
Perideridia
Lomatium
Apiaceae
Triteleia
Quercus
0.0%
Pit River Uplands Madeline Plains Secret Valley
Northeastern California Environmental Regions
Up to now, little has been said about the starch recovered from the five flaked stone tools examined as part of the study. Although none of the flaked stone artifacts produced much starch, and all of the grains were unidentifiable, the presence of starch is nevertheless intriguing. Minimally, it suggests that flaked stone technologies were more diverse or multi-functional than often conceived, including use as scrapers/peelers for the preparation of certain geophyte or other plant resources. They may have also been used

125 for non-subsistence plant related items including the preparation of various basketry materials and other woodworking activities.
The starch grain data generated here are extremely promising, but point to the need for further work if the technique is to reach its full potential. This includes the development of more complete comparative collections to identify other numerous unidentified starch grains that may derive from both root and other subsistence resources.
The identification of Perideridia , Lomatium , Triteleia , and Quercus starch confirms, however, the late prehistoric use of these plants in northeastern California, resolving a number of outstanding cultural historical and other regional issues.

126
Chapter 6
SUMMARY AND CONCLUSIONS
The starch grain evidence explored in this study indicates that late prehistoric populations relied extensively on certain plant resources. Previous research has suggested that human settlement and subsistence patterns in northeastern California changed substantially around 1300 B.P., shifting from a highly mobile to a residentially tethered lifeway tied to more intensive use of geophytic root crops. Various indirect indicators (e.g., rock-lined pits, milling equipment, scrapers/peelers) have been cited to support this hypothesized shift in settlement-subsistence patterns. The relocation of settlements to productive root-producing areas and lack of alternative storable resources further supports the expanded use of geophytic plants during the late prehistoric interval.
Starch grain analysis was employed to directly test the purported intensification of storable root crops within the region, and the connection between their use and certain artifact categories.
Archaeological starch grain data speak to various aspects of geophyte plant use and settlement-subsistence patterns in northeastern California. First, the presence of starch grains attributable to specific plants establishes direct evidence for their prehistoric use. Second, variations in the starch grains recovered from different contexts and the seasonal availability of plants provides a richer understanding of how specific resources were targeted within settlement-subsistence systems. Lastly, starch grain data from

127 specific tool categories furnishes new insight on plant processing techniques and the types of tools that best preserve undamaged starch grains.
The 667 starch grains recovered from flaked and ground stone tools indicate that starch can be retrieved in meaningful quantities from even curated collections. In fact, future starch grain studies will certainly expand our understanding of plant use in the current and other regions, shedding new light on these often archaeologically invisible resources.
The current research indicates the Perideridia starch is most frequently found on handstones and pestles wherever it was exploited. It appears most abundant, however, in the Pit River Upland sample, where this plant is especially abundant and could have been collected and stored in bulk. Triteleia was also recovered in each of the areas examined, but found in greater abundance on Madeline Plains artifacts. Though not especially abundant in the area today, Triteleia may have been more prevalent in the past, prior to historic grazing and other modern impacts to the flora. If nothing else, the enormous quantities of milling equipment in this area indicate that now limited plant resources were substantially more abundant in the past, with the dominance of Triteleia starch making it a prime candidate. By contrast, Lomatium was exclusively found in the Pit River
Uplands, suggesting that economically significant quantities of it may have been restricted to this habitat or exploited in conjunction with Perideridia .
As with any wild plant resource, there is a limited seasonal window to effectively and efficiently procure root crops. Unlike seed crops, geophytes are best collected during or prior to the flowering period. These underground storage organs are largest during the

128 winter season, such that the early flowering period is the most efficient and lucrative time to gather them. As different plants flower at different times, the scheduling of gathering activities is crucial to efficiently exploiting these foods (Table 6.1).
Table 6.1 – Flowering Periods of Storable Geophytes.
Plant Taxon Flowering Period
Triteleia
Camassia
Perideridia
Lomatium
March - April
April - June
May - July
May - July
As brodiaea ( Triteleia sp.) flowers earliest, it may have been preferentially gathered at this time. As more abundant crops, such as Lomatium , Camassia , and
Perideridia become available at higher elevations, groups would have moved to these areas to exploit those resources.
Although the starch grain data cannot speak to the utilization of camas ( Camassia quamash ), which lacks granules, it may have been another critical component of the aboriginal subsistence pattern. Camas and wild onion are so-called back-loaded resources (Bettinger 2009) that require intensive processing after they are collected, and need to be further investigated if their archaeological identification and place in the late prehistoric adaptation are to be adequately understood.
Acorn ( Quercus sp.) starch was only identified in Secret Valley, probably a function of its closer proximity to oak trees on the western edge of the Honey Lake
Basin. Significantly, acorn macro-botanical remains were never identified in Secret

129
Valley flotation samples, with the starch grain evidence expanding out understanding of plant use in the area. All of these grains were recovered from pestles at two different sites, pointing to perhaps intensive acorn processing with a mortar and pestle technology.
If true, acorns may have provided an alternative storable resource in Secret Valley that was not available in other parts of northeastern California. It may also provide a clue to the territorial extent of different Native American groups (Achomawi, Northern Paiute, and Mountain Maidu), who are thought to have shared parts of the Honey Lake region during late prehistory (Delacorte 1997a; McGuire 2000a).
Differences in the number and types of starch grains recovered from different tool categories (Figure 6.1) will allow future researchers to focus on the most productive tool classes and parts of assemblages. This will save time and resources, increasing our knowledge of the archaeological record and understanding of regional prehistory.

130
Figure 6.1 – Frequency of Starch Grains by Tool Category.
56.2 %
60
50
40
32.8 %
30
20
10
11 %
0
Handstones Pestles
Tool Classes
Millingstone/Mortar
FUTURE DIRECTION OF STUDY
While this thesis answers many questions regarding late prehistoric resource use and settlement-subsistence patterns, other questions remain. This includes the application of starch grain analysis to artifacts from earlier time-periods to clarify the history and purportedly late prehistoric intensification of root crop use. Another avenue of future research is to expand the number of artifact categories analyzed (e.g., battered cobbles, mortars, etc.) and to evaluate additional sites in each of the present and other parts of the

131 region (e.g., Pit River Valley, Honey Lake Basin, etc.). The current reference collection also needs to be expanded to include other plant species such as seed-producing grasses and annuals that may account for some of the presently unidentified starches. More needs to be learned as well about damaged versus undamaged starch granules, since many of the former are probably overlooked when slides are scanned under darkfield polarized light. This may shed light on the processing of various plant taxa and the characteristics of processed starch grains. Lastly, starch grain analysis cannot identify camas, a highly exploited root crop, whose complex carbohydrate structure lacks starch grains. However, new advancement hold considerable promise for efficiently identifying camas use through plant microfossils (Laurence and Thoms 2011; Thoms et al. 2011).
This will go far in furthering our understanding of camas use in prehistoric economies and the role of earth ovens, storage pits, and artifacts that may be associated with it.
Ultimately, starch grain analysis is a relatively new method of archaeological research that has been incorporated in studies across the globe to improve our understanding of ancient plant-based settlement-subsistence patterns. In California, there are no published studies and little research that incorporate starch grain analysis. This study hopes, if nothing else, to rectify that omission by establishing a foundation for the regular and standardized use of starch grain identification to further our understanding of the past.

APPENDICES
132

APPENDIX A
Comparative
Starch
Attributes
133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

APPENDIX B
Archaeological
Starch
Table
152

153
Artifact type Art. #
# of
Starch
Unknown Perideridia Lomatium Apiaceae Triteleia Quercus
Handstone 400
Handstone 402
Millingstone 397
Millingstone 399
Handstone 58
7
19
Handstone 488 23
9
6
6
Pestle
Handstone
Handstone
Pestle
43 21
148 4
413 33
14 7
Pestle 20 2
Handstone 1547 26
Handstone 1766 22
Pestle
Pestle
117 7
783 13
Millingstone 239
Millingstone 210
Handstone 1328
4
2
5
Handstone 1620 12
Handstone 2705 19
Millingstone 3432 13
Millingstone 1352
Hopper
Mortar
Pestle
3189
4120
9
18
246
Pestle 2514 47
Handstone 1333 42
Handstone 908 17
Bowl Mortar 958 13
Pestle
Pestle
493 18
1334 14
Total
684
13
4
24
6
2
20
15
5
7
5
15
7
7
5
6
12
17
13
9
4
1
2
18
180
45
37
16
13
15
13
536
1
1
4
2
4
1
2
10
3
1
1
30 10
3
1
5
1
47
1
58
3
3
1
1
1
1
5
1
1
2
15
1
1
1
1
2
4
1
1
6

154
Adams, J.L.
1996 Manual for a Technological Approach to Ground Stone Analysis. Center for
Desert Archaeology, Tucson.
Aikens, C.M., and D.L. Jenkins
REFERENCES
1994 Environment, Climate, Subsistence, and Settlement: 11,000 Years of Change in the Fort Rock Basin, Oregon. In Archaeological Researches in the
Northern Great Basin: Fort Rock Archaeology Since Cressman edited by
C.M. Aikens and D.L. Jenkins, pp. 1-20. University of Oregon
Anthropological Papers 50. Eugene.
Anderson, R. S.
1990 Holocene Forest Development and Paleoclimates within the Central Sierra
Nevada, California. The Journal of Ecology , Vol. 78, No. 2. pp. 470-489.
Ames, K.M., and A.G. Marshall
1981 Villages, Demography, and subsistence Intensification on the Southern
Columbia Plateau. North American Archaeologist 2(1): 25-52.
Anderson, P.
1980 A Testimony of Prehistoric tasks: Diagnostic residues on stone tool working edges. World Archaeology 12: 181-193.
Atchison, J. and R. Fullagar
1998 Starch Residues on Pounding Implements from Jinmium Rock-Shelter. In A
Closer Look: Recent Australian Studies on Stone Tools , edited by R. Fullagar.
Pp. 109-125. Sydney: Archaeological Computing Laboratory, University of
Sydney, 1998.
Babot, M. D. P. and M. C. Apella
2003 Maize and Bone: Residues of grinding in northwestern Argentina.
Archaeometry 45(1): 121-132.
Banks, W. and C. T. Greenwood
1975 Starch and Its Components . Edinburgh University Press, Edinburgh.
Barrett, S.A.
1910 The Material Culture of the Klamath Lake and Modoc Indians of Northeastern
California and Southern Oregon. University of California Publications in
American Archaeology and Ethnology 5(4): 239-292.

155
Barton, H.
2005 The case for rainforest foragers: The starch record at Niah Cave, Sarawak.
Asian Perspectives 44: 56-72.
2007 Starch residues on museum artefacts: implications for determining tool use. Journal of Archaeological Science 34: 1752-1762.
Barton, H., R. Torrence, and 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: 1231-1238.
Basgall, M. E.
1988 Archaeology of the Komodo Site, an Early Holocene Occupation in Central-
Eastern California . Nevada State Museum Anthropological Papers 21.
Carson City.
Baumhoff, M.
1978 Environmental Background. In California , edited by R. F. Heizer, Pp. 16-24.
Handbook of North American Indians, Vol. 8, W. C. Sturtevant, general editor. Smithsonian Institution, Washington D.C.
Baumhoff, M. and R. F. Heizer
1965 Postglacial Climate and Archaeology in the Desert West. In The Quarternary of the United States . edited by H.E. Wright, Jr. and D.G. Frey: Princeton
University Press, Princeton, N.J.
Beck, W. and R. Fullagar
1985 Starch Granule analysis: Potential applications in Australian archaeology.
Poster Presented to the Australian Archaeological Association annual conference, Valla, New South Wales, December 2 – 6.
Bedwell, S.F.
1973 Fort Rock Basin: Prehistory and Environment. University of Oregon,
Eugene.
BeMiller, J. N. and Whistler, R. L.
1996 Carbohydrates. In Food Chemistry . edited by FennemaOR. New York: Marcel
Dekker, Inc. Pp. 157–224.
Bennyhoff, J. A., and R. E. Hughes
1987 Shell Bead and Ornament Exchange Networks between California and the
Western Great Basin. Anthropological Papers of The American Museum of
Natural History . Vol. 64: Part 2, NewYork.

156
Bettinger, R.L.
2009 Hunter-Gatherer Foraging: Five Simple Models . Eliot Werner Publications,
Clinton Corners, NY.
Bettinger, R. L. and R. E. Taylor
1974 Suggested Revisions in Archaeological Sequences of the Great Basin and
Interior Southern California . Nevada Archaeological Survey Research Paper
5: 1-26.
Bouey, P. D. and M. E. Basgall
1984 Trans-Sierran Exchange in Prehistoric California: the Concept of Economic
Articulation. In Obsidian Studies in the Great Basin , edited by R. E. Hughes,
Pp. 135 – 172. Contributions of the University of California Archaeological
Research Facility No. 45. Berkeley.
Boyadjian, C. H. C., S. Eggers, Reinhard, K.
2007 Dental wash: a problematic method for extracting microfossils from teeth.
Journal of Archaeological Science 34: 1622-1628.
Briuer, F. L.
1976 New Clues to stone tool function: Plant and Animal residues. American
Antiquity 41: 478 – 484.
Brocklebank, J. H. and G. A. F. Hendry
1989 Characteristics of Plant species that store different types of reserve carbohydrates. New Phytologist 112(2): 255-260.
Chandler-Ezell, K., D. M. Pearsall and J. A. Zeidler
2006 Root and Tuber Phytoliths and Starch Grains Document Manioc ( Manihot esculenta ), Arrowroot ( Maranta arundinacea ), and Lleren ( Calathea allouia ) at the Real Alto Site, Ecuador. Economic Botany 60(2): 103-120.
Clewlow, C. W.
1968 Surface Archaeology of the Black Rock Desert, Nevada . University of
California Archaeological Survey Reports 70.
Connolly, T.J.
1986 Terrestrial and Riverine Economies in Southwest Oregon: A View from
Camas Valley. Association of Oregon Archaeologist Occasional Papers 3:
190-218.
Cook, S. F.
1943 The Conflict between the California Indian and White Civilization I.
University of California Publications: Ibero-Americana No. 21.

157
Cortella, A. R. and M. L. Pochettino
1994 Starch Grain Analysis as a Microscopic Diagnostic Feature in the
Identification of Plant Material. Economic Botany 48(2): 171-181.
Couture, M.D.
1978 Recent and Contemporary Foraging Practices of the Harney Valley Paiute.
Unpublished M.A. thesis, Department of Anthropology, Portland State
University.
Couture, M.D., M.F. Ricks, and L. Housley
1986 Foraging Behavior of a Contemporary Northern Great Basin Population.
Journal of California and Great Basin Anthropology 8(2): 150-160.
Cressman, L.S.
1942 Archaeological Researches in the Northern Great Basin . Carnegie Institution of Washington, Publication 538.
1956 Klamath Prehistory . American Philosophical Society Transactions (N.S.)
46(4).
1986 Prehistory of the Northern Area. In Great Basin
, edited by W.L. D’Azevedo,
Pp. 120-126. Handbook of North American Indians, Vol. 11, W. C.
Sturtevant, general editor. Smithsonian Institution, Washington D. C.
Cressman, L. S., H. Williams, and A. D. Krieger
1940 Early Man in Oregon; Archaeological Studies in the Northern Great Basin .
University of Oregon Monographs, Studies in Anthropology, No. 3. Eugene.
Crowther, A.
2005 Starch Residues on Undecorated Lapita Pottery from Anir, New Ireland.
Archaeology in Oceania 41: 62-66.
Cuthrell, Rob
2009 Personal Communication
Delacorte, M.G.
1997a Culture Change Along the Eastern Sierra Nevada/Cascade Front. Volume I,
Research Design.
Report prepared for Tuscarora Gas Transmission
Company, Reno.
1997b Culture Change Along the Eastern Sierra Nevada/Cascade Front. Volume III,
South Fork Valley and Madeline Plains.
Report prepared for Tuscarora Gas
Transmission Company, Reno.

158
2000 Late Prehistoric Resource Intensification in the Northwestern Great Basin. In
Archaeological Investigations Along the California-Great Basin Interface:
The Alturas Transmission Line Project. Volume I, Prehistoric Archaeological
Studies: The Pit River Uplands, Madeline Plains, Honey Lake and Secret
Valley, and Sierran Front Project Segments , by K.R. McGuire. Report submitted to Sierra Pacific Power Company.
2002 Late Prehistoric Resource Intensification in the Northwestern Great Basin. In
Boundary Lands: Archaeological Investigations along the California-Great
Basin Interface , edited by K. R. McGuire, pp. 41-48. Nevada State Museum
Anthropological Papers No. 24. Nevada Department of Cultural Affairs,
Carson City.
2008 Desert Side-notched Points as a Numic Population Marker in the West-Central
Great Basin. In Avocados to Millingstones: Papers in Honor of D.L. True , edited by Georgie Waugh and Mark E. Basgall, pp. 111-136. Monographs in
California and Great Basin Anthropology, No. 5. Archaeological Research
Center, California State University, Sacramento.
2009 Personal Communication
Delacorte, M. and D. C. Young
2000 Natural Setting. In Archaeological Investigations Along the California-Great
Basin Interface: The Alturas Transmission Line Project. Volume I,
Prehistoric Archaeological Studies: The Pit River Uplands, Madeline Plains,
Honey Lake and Secret Valley, and Sierran Front Project Segments , by K.R.
McGuire. Pp. 9-14. Report submitted to Sierra Pacific Power Company.
Dickau, R.
2005 Resource Use, Crop Dispersal, and the Transition to Agriculture in
Prehistoric Panama: Evidence from Starch Grains and Macroremains .
Unpublished Ph.D. Dissertation, Temple University, Philadelphia.
Dixon, R. B.
1905 The Shasta-Achomawi: A New Linguistic Stock, with Four New Dialects.
American Anthropologist 7(2): 206-217.
1908 Achomawi and Atsugewi Tales: Achomawi Myths. Journal of American
Folklore . Vol. 22, no. 81, Pp. 159-77.
Elston, R.
1979 The Archaeology of U.S. 395 Right-of-Way Between Stead, Nevada and
Hallelujah Junction, California . Archaeological Survey Report, University of
Nevada, Reno.

159
1982 Good Times, Hard Times: Prehistoric Culture Change in the Western Great
Basin. In Man and Environment in the Great Basin , edited by D. B. Madsen and J. F. O’Connell, Pp. 186-206. Society of American Archaeology, No. 2.
Washington, D.C.
1986 Prehistory of the Western Area. In Great Basin , edited by W. L. D’Azevedo,
Pp. 135-148. Handbook of North American Indians, Vol. 11, W. C.
Sturtevant, general editor. Smithsonian Institution, Washington, D.C.
Evans, W.S., Jr.
1978 Ethnographic Notes on the Honey Lake Maidu . (with Genealogies [Appendix
I] by F.A. Riddell) Nevada State Museum Occasional Papers No. 3(1).
Field, J.
2006 Reference Collections. In Ancient Starch Research . edited by R. Torrence and
H. Barton, Pp. 95-113. Left Coast Press: Walnut Creek, California.
Field, J. and R. Fullugar
1998 Grinding and Pounding Stones from Cuddie and Jinmium. In A Closer Look:
Recent Australian Studies of Stone Tools , edited by R. Fullagar, Pp. 95-108.
Sydney University Archaeological Methods Series 6, Sydney.
Foster-Curley, C.
2006 Adaptive Shifts in the Northwestern Great Basin: Prehistoric Culture Change in the Pit River Uplands . Unpublished M.A. thesis, Department of
Anthropology, CSU-Sacramento.
Fowler, C.S.
1986 Subsistence. In Great Basin
, edited by W. D’Azevedo, Pp. 64-97. Handbook of North American Indians, Vol. 11, W. C. Sturtevant, general editor.
Smithsonian Institution, Washington D.C.
Fowler, C. S., and S. Liljeblad
1986 Northern Paiute. In Great Basin
, edited by W. L. D’Azevedo, Pp. 135-148.
Handbook of North American Indians, Vol. 11, W. C. Sturtevant, general editor. Smithsonian Institution, Washington, D.C.
Fowler, D.D.
1993 Archaeological Investigations in Warner Valley, Oregon, 1989-1992: An
Interim Report.
Department of Anthropology Technical Report Series 93-1.
University of Nevada.

160
Fullagar, R.
1986 Use-wear and residues on stone tools: Functional analysis and its application to two southeastern Australian archaeological assemblages . Unpublished
Ph.D. Dissertation, LaTrobe University, Melbourne.
1988 Recent developments in Australian use-wear and residue studies. In S.
Beyries, ed., Industries Lithiques: Traceologie et Technologie , Pp.133-45.
British Archaeological Reports International Series 411.
1989 The Potential of lithic use-wear and residue studies for determining stone tool functions. In P.P. Gorecki and D. Gillieson, eds., A Crack in the Spine:
Prehistory and Ecology of the Jimi-Yuat Valley, Papua New Guinea , Pp.
209-23. Queensland: James Cook University, Townsville.
1992 Lithically Lapita: Functional analysis of flaked stone assemblages from West
New Britain Providence, Papua New Guinea. In Poterie Lapita et
Peuplement , edited by J.-C. Galipaud, Pp. 135-43. Noumea: ORSTOM.
2006 Starch on Artifacts. In Ancient Starch Research . edited by R. Torrence and H.
Barton, Pp. 177-202. Left Coast Press: Walnut Creek, California.
Fullagar, R., T. Loy and S. Cox
1998 Starch Grains, Sediments, and Stone Tool Function: Evidence from Bitokara,
Papua New Guinea. In A Closer Look: Recent Australian Studies of Stone
Tools , edited by R. Fullagar, Pp. 49-60. Sydney University, Sydney.
Fullagar, R., B. Meehan and R. Jones
1999 Residue analysis of ethnographic plant-working and other tools from northern
Australia. In Prehistory of Agriculture; New Experimental and Ethnographic
Approaches .
edited by P. Anderson-Gerfaud, Berkeley: Institute of
Archaeology. UCLA. Monograph 40: 15-23
Galliard, T. and P. Bowler
1987 Morphology and composition of starch. In Starch: Properties and Potential , edited by T. Galliard, Pp. 55-79. Critical Reports on Applies Chemistry, volume 13. Brisbane: John Wiley and Sons.
Galm, J. R.
1985 The Avey’s Orchard Site in Regional Perspective. In
Avey’s Orchard:
Archaeological Investigations of a Late Prehistoric Columbia River
Community , edited by J.R. Galm and R.A. Matsen, Pp. 321-334. Eastern
Washington University Reports in Archaeology and History 100-42. Cheney.

161
Garth, T. R.
1953 Atsugewi Ethnography. University of California Anthropological Records
14(2): 129-212. Berkeley.
1978 Atsugewi. In California , edited by R.F. Heizer, Pp.236-243. Handbook of
North American Indians, Vol. 8, W.C. Sturtevant, general editor. Smithsonian
Institution, Washington D.C.
Gatschet, A.S.
1890 The Klamath Indians of Southwestern Oregon . Department of the interior,
U.S. Geographical and Geological Survey of the Rocky Mountain Region.
Contributions to North American Ethnology 1. Washington.
Gleason, S.M.
2001 In Search of the Intangible: Geophyte Use and Management along the Upper
Klamath River Canyon . Unpublished Ph.D. dissertation, University of
California, Riverside.
Gott, B., H. Barton, S. Delwen and R. Torrence
2006 Biology of Starch. In Ancient Starch Research , edited by R. Torrence and H.
Barton, Pp. 35-46. Left Coast Press, Walnut Creek, California.
Grayson, D. K.
1973 The Mammalian and Avian Remains from Nightfire Island . Unpublished
Ph.D. Dissertation, Department of Anthropology, University of Oregon,
Eugene. University Microfilms, Ann Arbor.
Hall, C.
1944 The Hall Interview, dictated in 1873. Printed in Fall River Tidings 2/25/1944,
Pp. 11-12. Fall River Mills, California.
Hall, J., S. Higgins, and R. Fullagar
1989 Plant residues on stone tools. In Plants in Australian Archaeology , edited by
W. Beck, A. Clarke, and L. Head, Pp. 136-60. St. Lucia: University of
Queensland, Tempus 1.
Harden, D.
1998 California Geology . Prentice Hall, Upper Saddle River, New Jersey.
Hardesty, D. L. and S. Fox
1974 Archaeological Investigations in Northern California. Nevada Archaeological
Survey Research Paper , No. 4. University of Nevada, Reno.

162
Haslam, M.
2004 The Decomposition of Starch Grains in Soils: Implications for
Archaeological Residue Analyses. Journal of Archaeological Science 31:
1715-1734.
Heizer, R.F.
1966 Ancient heavy transport: methods and achievements. Science 161: 914-15.
1967 Human Coprolites from a Dry Nevada Cave. University of California
Archaeological Survey Reports 70: 1-20.
Heizer, R. F., and M. A. Baumhoff
1961 The Archaeology of Two Sites at Eastgate, Churchill County, Nevada.
University of California Anthropological Records 20(4): 119-149.
Heizer, R. F. and T. R. Hester
1978a Great Basin Projectile Points: Forms and Chronology. Ballena Press
Publications in Archaeology, Ethnology, and History 10. Socorro, N.M.
1978b Great Basin. In Chronologies in New World Archaeology , edited by R. E.
Taylor and C. W. Meighan, Pp. 147-199. Academic Press, New York.
Heizer, R. F., and A. D. Kreiger
1956 The archaeology of Humboldt Cave, Churchill County, Nevada. University of
California Publications in American Archaeology and Ethnology 47: 1-190.
Heizer, R.F., and L.K. Napton
1970 Archaeology and the Prehistoric Great Basin Lacustrine Subsistence Regime as seen from Lovelock Cave, Nevada . Contributions of the University of
California Archaeological Research Facility 10. Berkeley.
Henry, A. G. and 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:
1943-1950.
Henry, A. G., H. F. Hudson, and D. R. Piperno
2009 Changes in Starch Grain Morphologies from Cooking. Journal of
Archaeological Science 36: 915-922.

163
Hildebrandt, B.
2000 Summary of site CA-LAS-1705. In Archaeological Investigations Along the
California-Great Basin Interface: The Alturas Transmission Line Project.
Volume I, Prehistoric Archaeological Studies: The Pit River Uplands,
Madeline Plains, Honey Lake and Secret Valley, and Sierran Front Project
Segments . Report prepared for Sierra Pacific Power Company.
Hildebrandt W. R. and P. J. Mikkelson
1994 Vol. IIC. Summary Reports: Prehistoric sites, California. In Archaeological
Investigations PGT-PG&E Pipeline Expansion Project Idaho, Washington,
Oregon, California , edited by M.J. Moratto. Report submitted to Pacific Gas
Transmission Company, San Francisco.
1995 Projectile Point Typology. In Archaeological Investigations PGT-PG&E
Pipeline Expansion Project Idaho, Washington, Oregon, and California , vol.
V, by R. U. Bryson, C. E. Skinner, and R. M. Pettigrew, Pp. 1-1 to 1-40.
Report submitted to Pacific Gas Transmission Company, Portland.
Holanda, K.
2000 Summary of site CA-MOD-3448. In Archaeological Investigations Along the
California-Great Basin Interface: The Alturas Transmission Line Project.
Volume I, Prehistoric Archaeological Studies: The Pit River Uplands,
Madeline Plains, Honey Lake and Secret Valley, and Sierran Front Project
Segments . Report prepared for Sierra Pacific Power Company.
Holt, C.
1946 Shasta Ethnography . University of California Anthropological Records 3(4).
Berkeley.
Horrocks, M., and S. Bedford
2005 Microfossil analysis of Lapita deposits in Vanuatu reveals introduced Araceae
(aroids). Archaeology in Oceania 40: 67-74.
Horrocks, M. and P. D. Nunn
2007 Evidence for introduced taro ( Colocasia esculenta ) and lesser yam ( Dioscorea esculenta ) in Lapita-era (c. 3050-2500 cal. yr BP) deposits from Bourewa, southwest Viti Levu Island, Fiji. Journal of Archaeological Science 34: 739-
748.
Horrocks, M., G. Irwin, M. Jones, and D. Sutton
2004 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: 251-58.

164
Housley, L.
1994 It is in the Roots: Prehistoric Plants and Plant Use in the Fort Rock Basin. In
Archaeological Research in the Northern Great Basin: Fort Rock
Hughes, R.
Archaeology since Cressman , edited by C.M. Aikens and D.L. Jenkins, Pp.
561-572. University of Oregon Anthropological Papers 50. Eugene.
1977 The Archaeology of the Burrell Site CA-MOD-293: A Lowland Occupation
Site in the Goose Lake Basin, Northeast California . Report prepared for the
National Endowment for the Humanities, The Goose Lake Archaeological
Project.
1986 Diachronic Variability in Obsidian Procurement Patterns in Northeastern
Hunn, E.S.
California and Southcentral Oregon.
University of California Publications,
Anthropology 17. University of California Press.
1990 The Plateau Culture Area. In Native North Americans: An Ethnobotanical
Approach . Edited by D.L. Boxberger, Dubuque, Iowa: Kendall/Hunt
Publishing Company, Pp. 361-385.
Hunn, E.S., N.J. Turner, and D.H. French
1998 Ethnobiology and Subsistence. In Plateau , edited by D.E. Walker, Jr., Pp.
525-545. Handbook of North American Indians, Vol. 12, W. C. Sturtevant, general editor. Smithsonian Institution, Washington D.C.
Jenkins, D.L.
1994 Settlement-Subsistence Patterns in the Fort Rock Basin: A Cultural –
Ecological Perspective on Human Responses to Fluctuating Wetlands
Resources of the Last 5000 Years. In Archaeological Researches in the
Northern Great Basin: Fort Rock Archaeology since Cressman , edited by
C.M. Aikens and D.L. Jenkins, Pp. 431-484. University of Oregon
Anthropological Papers 50. Eugene.
Jenkins, D. L., and T.J. Connolly
1990 Archaeology of the Indian Grade Spring: A Special Function Site on
Stinkingwater Mountain, Harney County, Oregon . University of Oregon
Anthropological Papers 42. Eugene.
Jennings, C. W.
1977 Geologic Map of California. Division of Mines and Geology.
Jennings, J. D.
1957 Danger Cave . University of Utah Anthropological Papers 27. Salt Lake City.

165
1964 The Desert West. In Prehistoric Man in the New World , edited by J.D.
Jennings and E. Norbeck, Pp. 149-174. University of Chicago Press.
Jones, K. T. and D. B. Madsen
1991 Further Experiments in Native Food Procurement. Utah Archaeology , Vol. 4,
No. 1, Pp. 68-77.
Kamminga, J.
1979 The Nature of Use-polish and Abrasive Smoothing on Stone Tools. In Lithic
Use-Wear Analysis , edited by B. Hayden. Pp. 143-157, New York:
Academic.
Keeley, L.
1980 Experimental Determination of Stone Tool Uses: A Microwear Analysis .
University of Chicago Press, Chicago.
Kelly, I.T.
1932 Ethnography of the Surprise Valley Paiute. University of California
Publications in American Archaeology and Ethnology 31(3): 67-210.
Kelly, R. L.
1988 The Three Sides of a Biface. American Antiquity 53: 717-734.
Kniffen, F.B.
1928 Achomawi Geography. University of California Publications in American
Archaeology and Ethnology 23(5): 297-332. Berkeley.
Kolvet, R.C., and J.A. Eisele
2000 A Ground Stone Study and Classification System for the Great Basin .
Archives of Great Basin Prehistory, Coyote Press, Salinas, California.
Kroeber, A. L.
1925 Handbook of the Indians of California . Smithsonian Institution, Bureau of
American Ethnology Bulletin 78. Berkeley. Reprinted (1976) Dover
Publications, Inc., New York.
Kuhnlein, H. V., N. J. Turner, and P. D. Kluckner
1982 Nutritional significance of two important root foods (springbank clover and
Pacific silverweed) use by Native People of the coast of British Columbia.
Ecology of Food and Nutrition 12: 89-95.

166
Lamb, J.
2003 The Raw and the Cooked: A Study on the Effects of Cooking on Three
Aboriginal Plant Foods in Southeast Queensland.
Poster presented at the
2003 Australian Archaeological Association conference. NSW, AU.
Lamb, J. and 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:
1433-1440.
Landhausser, S. M. and V. J. Lieffers
2003 Seasonal changes in carbohydrate reserves in mature northern Populus tremuloides clones. In Trees – Structure and Function 17(6): 471-476.
Laurence, A. R. and A. V. Thoms
2011 Microscopic Revelations from Pre-Columbian Earth Ovens in Central Texas .
Poster presented at the 76 th
Annual SAA Meeting, Sacramento, California,
March 31-April 3.
Layton, T. N.
1970 High Rock Archaeology: An Interpretation of the Prehistory of the
Northwestern Great Basin . Unpublished Ph.D. dissertation, Department of
Anthropology, Harvard University.
Leach, M.
1988 Subsistence Intensification and Settlement Change among Prehistoric Hunters and Gatherers of the Northwestern Great Basin . Unpublished Ph.D. dissertation, University of California, Los Angeles.
Leach-Palm, L.
2000 Summary of Site CA-Las-1623. Archaeological Investigations Along the
California-Great Basin Interface: The Alturas Transmission Line Project.
Volume I, Prehistoric Archaeological Studies: The Pit River Uplands,
Madeline Plains, Honey Lake and Secret Valley, and Sierran Front Project
Segments . Report prepared for Sierra Pacific Power Company.
Lentfer, C., M. Therin, and R. Torrence
2002 Starch Grains and Environmental Reconstruction: a Modern Test Case from
West New Britain, Papua New Guinea. Journal of Archaeological Science
29: 687-698.
Loy, T.
1993 The artifact as site: An example of the biomolecular analysis of organic residues on stone tools. World Archaeology 25: 44-63.

167
1994 Methods in the Analysis of Starch Residues on Prehistoric Stone Tools. In
Tropical Archaeobotany: Applications and New Developments , edited by J. G.
Hather, Pp. 86-114. Routledge, New York.
Loy, T., M. Spriggs and 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: 898-
912.
Macdonald, G. A.
1965 Geologic map of the Harvey Mountain quadrangle, Lassen County, California.
United States Geological Survey.
Mack, J.
1983 Archaeological Investigations in the Salt Cave Locality: Subsistence
Uniformity and Cultural Diversity on the Klamath River, Oregon . University of Oregon Anthropological Papers 29. Eugene.
Mahar, M.
1953 Ethnobotany of the Oregon Paiutes of the Warm Springs Indian Reservation.
Unpublished B.A. thesis, Reed College, Portland.
Manuel, D.W.
1983 Class II Cultural Resource Inventory of the Alturas Resource Area, Susanville
District . Bureau of Land Management, Susanville District. Susanville,
California.
1989 Excavation and Analysis of CA-LAS-973: A Small Winter Village on the
Modoc Plateau of Northeastern California . Report on file, Bureau of Land
Management, Susanville, California.
Matson, R.G. and D. Alexander
1987 Report on the Potato Mountain Archaeological Project (1985) . Laboratory of
Archaeology, University of British Columbia. Report prepared for the Social
Sciences and Humanities Research Council of Canada.
Matthews, P. and R. Torrence
2006 The very broad temporal and geographic ranges of archaeological contexts in which starch had been preserved. Table 1.1. In Ancient Starch Research . edited by R. Torrence and H. Barton, Pp. 19. Left Coast Press: Walnut Creek,
California.

168
McCarthy, H.
1997a Ethnographic Background. In Culture Change Along the Eastern Sierra
Nevada/Cascade Front. Volume IV: Secret Valley . Report prepared for
Tuscarora Gas Transmission Company, Reno, Nevada.
1997b Ethnographic Background. In Culture Change along the Eastern Sierra
Nevada/Cascade Front. Volume II:, South Fork Valley and Madeline Plains.
Report prepared for Tuscarora Gas Transmission Company, Reno.
McGuire, K.
1997 Culture Change along the Eastern Sierra Nevada/Cascade Front. Volume VI:
Secret Valley.
Report prepared for Tuscarora Gas Transmission Company,
Reno.
2000a Archaeological Investigations Along the California-Great Basin Interface:
The Alturas Transmission Line Project. Volume I, Prehistoric Archaeological
Studies: The Pit River Uplands, Madeline Plains, Honey Lake and Secret
Valley, and Sierran Front Project Segments . Report prepared for Sierra
Pacific Power Company.
2000b Summary of Site Ca-Las-206. In Archaeological Investigations Along the
California-Great Basin Interface: The Alturas Transmission Line Project.
Volume I, Prehistoric Archaeological Studies: The Pit River Uplands,
Madeline Plains, Honey Lake and Secret Valley, and Sierran Front Project
Segments . Report prepared for Sierra Pacific Power Company.
Meilleur, B.A., E. S. Hunn, and R. L. Cox
1990 Lomatium dissectum (Apiaceae): Multi-Purpose Plant of the Pacific
Northwest. Ethnobiology 10(1): 1-20.
Merriam, C. H.
1926 The Classification and Distribution of the Pit River Tribes of California.
Smithsonian Miscellaneous Collections 78(3): 1-52. Washington, D.C.
Merriam, C. H, and Z. M. Talbot
1974 Boundary Descriptions of California Indian Stocks and Tribes .
Archaeological Research Facility Department of Anthropology, Berkeley.
Messner, T.
2008 Woodland Period People and Plant Interactions: New Insights from Starch
Grain Analysis . Unpublished Ph.D. dissertation, Temple University,
Philadelphia, PA.

169
Messner, T. and B. Schindler
2010 Plant Processing and their Effect upon Starch Grain Survival when rendering
Peltandra virginica (L.) Kunth, Araceae edible. Journal of Archaeological
Science 37, no. 2 (2010): 328-336.
Mifflin, M. D. and M. M. Wheat
1979 Pluvial Lakes and Estimated Pluvial Climates of Nevada. Nevada Bureau of
Mines and Geology, Bulletin 94 . Macay School of Mines, University of
Nevada, Reno.
Mikkelson, P. J. and R. U. Bryson
1997 Culture Change along the Eastern Sierra Nevada/Cascade Front. Volume II:
Modoc Uplands . Report prepared for Tuscarora Gas Transmission Company,
Reno.
Milliken, R. T. and W. R. Hildebrandt
1997 Culture Change along the Eastern Sierra Nevada/Cascade Front. Volume V:
Honey Lake Basin . Report prepared for Tuscarora Gas Transmission
Company. Reno, Nevada.
Neasham, E. R.
1957 Fall River Valley: An Examination of Historical Sources. In California
Indians III , edited by E. W. Voegelin and E. R. Neasham. American Indian
Ethnohistory: California and Basin-Plateau Indians, Pit River Indians of
California. Garland Publishing Inc., New York.
O’Connell, J. F.
1967 Elko Eared/Elko Corner-notched Projectile Points as Time Markers in the
Great Basin . University of California Archaeological Survey Reports 70:
129-140.
1971 The Archaeology and Cultural Ecology of Surprise Valley, Northeast
California . Unpublished Ph.D. dissertation, Department of Anthropology,
University of California, Berkeley.
1975 The Prehistory of Surprise Valley , edited by L. J. Bean. Ballena Press
Anthropological Papers No. 4, Ramona, California.
O’Connell, J. F. and P. S. Hayward
1972 Altithermal and Medithermal Human Adaptations in Surprise Valley,
Northeast California. In Great Basin Cultural Ecology: A Symposium , edited by D. D. Fowler, Pp. 25-42. Desert Research Institute Publications in the
Social Sciences 8.

170
Odell, G., and F. Odell-Vereecken
1980 Verifying the Reliability of Lithic Use-Wear Assessments by ‘Blind Tests’: the Low Power Approach. Journal of Field Archaeology 7(1): 87-120.
Olmstead, D. L., and O.C. Stewart
1978 Achomawi. In California , edited by R. F. Heizer, Pp. 225-236. Handbook of
North American Indians, Vol. 8, W. C. Sturtevant, general editor.
Smithsonian Institution, Washington D.C.
Pearsall, D. M., K. Chandler-Ezell and 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): 423-442.
Perry, L.
2001 Prehispanic Subsistence in the Middle Orinoco Basin: Starch Analyses yield
New Evidence . Ph.D. dissertation, Southern Illinois University, Carbondale.
Perry, L., R. Dickau, S. Zarillo, I. Holst, D. M. Pearsall, D. R. Piperno, M. J. Berman, R.
G. Cooke, K. Rademaker, A. J. Ranere, J. S. Raymond, D. H. Sandweiss, F. Scaramelli,
K. Tarble, J. A. Zeidler
2007 Starch Fossils and the Domestication and Dispersal. Science 315: 986-988.
Piperno, D. R.
2006 Starch grains and root crops. In Documenting Domestication: New Molecular and Archaeological Paradigms, edited by M. Zeder, D. Decker-Walters, and
B. Smith .
Washington: Smithsonian Institution Press.
Piperno, D. R. and I. Holst
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): 765-776.
2004 Crop Domestication in the American Tropics: Starch Grain Analysis. In
Encyclopedia of Plant and Crop Science , edited by R. Goodman, Pp. 330-332.
New York: Marcel Dekker.
Piperno, D. R., A. J. Ranere, I Holst and P. Hansell
2000 Starch Grains Reveal Early Root Crop Horticulture in Panamanian Tropical
Forest. Nature . 407: 894-897.
Piperno, D. R., E. Weiss, I. Holst and D. Nadel
2004 Processing of Wild Cereal Grains in the Upper Paleolithic Revealed by Starch
Grain Analysis. Nature . 430: 670-673.

171
Powers, S.
1877 Tribes of California . Contributions to North American Ethnology, Vol. 3.
University of California Press, Berkeley.
Prouty, G. L.
1994 Root Crop Exploitation and the Development of Upland Habitation Sites: A
Prospectus for Paleobotanical and Archaeological Research into the
Distributions of Economic Plants in the Fort Rock Basin. Archaeological
Researches in the Northern Great Basin: Fort Rock Archaeology since
Cressman , edited by C.M. Aikens and D.L. Jenkins, Pp. 573-598. University of Oregon Anthropological Papers 50. Eugene.
1995 Roots and Tubers: Prehistoric Plant Use, Settlement and Subsistence
Intensification and Storage in the Fort Rock Basin, Northern Great Basin,
Oregon . Unpublished Ph.D. dissertation, Department of Anthropology,
University of Oregon, Eugene.
Raven, C.
1984 Northeastern California. In California Archaeology , edited by M. J. Moratto,
Pp. 431-470. Academic Press, New York.
Ray, V.F.
1963 Primitive Pragmatists: The Modoc Indians of Northern California .
University of Washington Press, Seattle.
Reichert, E. T.
1913 The Differentiation and Specificity of Starches in Relation to Genera, Species, etc . Carnegie Institute, Washington D.C.
Riddell, F. A.
1956 Final Report on the Archaeology of Tommy Tucker Cave . University of
California Archaeological Survey Reports 35. Berkeley.
1958 The Eastern California Border: Cultural and Temporal Affinities . University of California Archaeological Survey Reports 42: 41-48. Berkeley.
1960a The Archaeology of the Karlo Site (Las-7), California . University of
California Archaeological Survey Reports 53. Berkeley.
1960b Honey Lake Paiute Ethnography . Nevada State Museum Anthropological
Papers 4. Carson City.

172
Rosenthal, J.
2000 Summary of Site CA-Mod-3150. In Archaeological Investigations Along the
California-Great Basin Interface: The Alturas Transmission Line Project.
Volume I, Prehistoric Archaeological Studies: The Pit River Uplands,
Madeline Plains, Honey Lake and Secret Valley, and Sierran Front Project
Segments . Report prepared for Sierra Pacific Power Company.
Rumold, C.
2010
Illuminating Women’s Work and the Advent of Plant Cultivation in the
Highland Titicaca Basin of South America: New Evidence from Grinding
Tool and Starch Analyses . Unpublished Ph.D. dissertation, University of
California, Santa Barbara, CA.
Sawyer, John O., Todd Keeler-Wolf, and Julie Evens
2009 A Manual of California Vegetation . California Native Plant Society,
Sacramento, CA.
Scholze, G.
2007 Native Root Use within the Northern California Environmental Zones . Paper presented at the 41 st
Annual Meeting of the Society for California
Archaeology, March 22-25, San Jose.
2010 Potential Role of Starch-Grain Analysis in Determining Geophyte Use within
Northeastern California. Proceedings of the Society of California
Archaeology . Vol. 24.
Shafer, H. J., and R. G. Holloway
1979 Organic residue analysis in determining stone tool function. In Lithic Usewear Analysis, edited by B. Hayden, Pp. 385-399. London: Academic Press.
Simmons, W. S., R. Morales, V. Williams, and S. Camacho
1997 Honey Lake Maidu Ethnography of Lassen County, California. Journal of
California and Great Basin Anthropology 19(1): 2-31.
Sivak, M.N.S. and J. Preiss
1998 Occurrence of starch. In Starch: Basic Science to Biotechnology, M.N.S.
Sivak and J. Preiss, eds., Pp. 1-12. Advances in Food and Nutrition Research ,
41. San Diego: Academic Press.
Spier, L.
1930 Klamath Ethnography. University of California Publications in American
Archaeology and Ethnology 30 . Berkeley.

173
Steward, J. H.
1933 Ethnography of the Owens Valley Paiute. University of California
Publications in American Archaeology and Ethnology 33: 233-350. Berkeley.
1938 Basin-Plateau Aboriginal Sociopolitical Groups . Bureau of American
Ethnology Bulletin 120. Smithsonian Institution, Washington, D.C.
Stewart, O. C.
1941 Culture Element Distributions, XIV: Northern Paiute. University of
California Anthropological Records 4(3): 361-446. Berkeley.
Theodoratus, D. J., F. A. Riddell, C. M. Blount, D. A. Bell, and M. E. Peters
1979 Anthropological Overview of a Portion of Lassen and Modoc Counties,
Therin, M.
Volume 1 . Report submitted to the U.S. Department of the Interior Bureau of
Land Management, Susanville, California.
1994 Subsistence through starch: The examination of subsistence changes on
Garua Island, West New Britain, Papua New Guinea, through the extraction and identification of starch from sediments . Unpublished B.A. (Hons) thesis,
Archaeology, University of Sydney.
Thomas, D. H.
1973 An Empirical Test for Steward’s Model of Great Basin Settlement Patterns.
American Antiquity 38: 155-176.
1981 How to Classify the Projectile Points from Monitor Valley, Nevada. Journal of California and Great Basin Anthropology 3: 7-43.
Thoms, A.V.
1989 The Northern Roots of Hunter-Gatherer Intensification: Camas and the
Pacific Northwest . Unpublished Ph.D. dissertation, department of
Anthropology, Washington State University, Pullman.
Thoms, A.V., L. R. Laurence, M. Kamiya, and V. M. Bryant, Jr.
2011 Experimentally Tacking Geophyte Microfossils from Raw to Baked-Residue
Status . Poster presented at the 76 th
Annual SAA Meeting, Sacramento,
California, March 31-April 3.
Torrence, R.
2006 Description, classification and identification. In Ancient Starch Research . edited by R. Torrence and H. Barton. Pp. 115-144. Left Coast Press: Walnut
Creek, California.

174
Torrence, R. and H. Barton (editors)
2006 Ancient Starch Research . Left Coast Press, Inc., Walnut Creek.
Torrence, R. and Conway, R.
2006 Classification using image analysis and multivariate techniques. In Ancient
Starch Research . edited by R. Torrence and H. Barton, Pp. 137-142. Left
Coast Press: Walnut Creek, California.
Torrence, R. and Therin, M.
2006 Experiments for testing starch extraction techniques. In Ancient Starch
Research . edited by R. Torrence and H. Barton, Pp. 156-158. Left Coast
Press: Walnut Creek, California.
Torrence, R., R. Wright, and R. Conway
2004 Identification of starch granules using image analysis and multivariate techniques. Journal of Archaeological Science 31: 519-32.
Turner, N. J., and H. V. Kuhnlein
1982 Two important "root" foods of the Northwest Coast Indians: springbank clover ( Trifolium wormskioldii ) and Pacific silverweed ( Potentilla anserina ssp. pacifica ). Economic Botany , 36(4), 411-432.
1983 Camas ( Camassia spp.) and Riceroot ( Fritillaria spp.): Two Liliaceous
“Root” Foods of the Northwest Coast Indians.
Ecology of Food and Nutrition
13: 199-219.
Ugent, D., S. Pozorski and T. Pozorski
1981 Prehistoric Remains of the Sweet Potato from the Casma Valley of Peru.
Phytologia 49: 401-415.
1982 Archaeological Potato Tuber Remains from the Casma Valley of Peru.
Economic Botany 36: 182-192.
1984 New Evidence for Ancient Cultivation of Canna edulis in Peru. Economic
Botany 38: 417-432.
1986 Archaeological Manioc (Manihot) from Coastal Peru. Economic Botany 40:
78-102.
Van Peer, P., R. Fullagar, S. Stokes, R. M. Bailey, J. Moeyersons, F. Steenhoudt, A.
Geerts, T. Vanderbeken, M. De Dapper, and 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: 187-193.

175
Vaughn, P. C.
1985 Use Wear Analysis of Flaked Stone Tools . University of Arizona Press,
Tucson.
Voegelin, E.W.
1942 Culture Element Distributions, XX: Northeast California. University of
California Anthropological Records 7. Berkeley.
1974 Pit River Indians of California. In California Indians III , edited by E. W.
Voegelin and E. R. Neasham. American Indian Ethnohistory: California and
Basin-Plateau Indians, Pit River Indians of California. Garland Publishing
Inc., New York.
Waechter, S.
2000a Summary of Site Ca-Mod-3153. In Archaeological Investigations along the
California-Great Basin Interface: The Alturas Transmission Line Project.
Volume I, Prehistoric Archaeological Studies: The Pit River Uplands,
Madeline Plains, Honey Lake and Secret Valley, and Sierran Front Project
Segments . Report prepared for Sierra Pacific Power Company.
2000b A Record of Cultural Upheaval: The Late-Period Residential Pattern in
Crooks Canyon. In Archaeological Investigations along the California-Great
Basin Interface: The Alturas Transmission Line Project. Volume I,
Prehistoric Archaeological Studies: The Pit River Uplands, Madeline Plains,
Honey Lake and Secret Valley, and Sierran Front Project Segments . Pp. 299-
317. Report prepared for Sierra Pacific Power Company.
Waechter, S. and D. Andolina
2005 Ecology and Prehistory in Sierra Valley, California: Excavations at CA-
PLU-1485 . Report prepared for Caltrans. On file at Far Western
Anthropological Research Group, Davis, CA.
Weide, M. L.
1968 Cultural Ecology of Lakeside Adaptation in the Western Great Basin .
Unpublished Ph.D. dissertation, Department of Anthropology, University of
West, G. J.
California, Los Angeles.
1997 Paleoenvironmental History of Secret Creek Marsh. In Culture Change along the Eastern Sierra Nevada/Cascade Front. Volume VI: Secret Valley.
Pp. 6-
7. Report prepared for Tuscarora Gas Transmission Company, Reno.

176
Wheeler-Voegelin, E.
1974 Pit River Indians of California. In American Indian Ethnohistory: California
Indians, 3: 6-184. Garland, New York.
Williamson, B. S.
2004 Middle Stone Age tool function from residue analysis at Sibudu Cave. South
African Journal of Science 100: 174-178.
Wilson, J., K. Hardy, R. Allen, L. Copeland, R. Wrangham, M. Collins
2010 Automated classification of starch granules using supervised pattern recognition of morphological properties. Journal of Archaeological Science .
37: 594-604.
Wohlgemuth, E.
1997 Vegetation and Plant Resources. In Culture Change along the Eastern Sierra
Nevada/Cascade Front. Volume VI: Secret Valley.
Pp. 5-6. Report prepared for Tuscarora Gas Transmission Company, Reno.
2000 Floral Remains of CA-MOD-3153. In Archaeological Investigations Along the California-Great Basin Interface: The Alturas Transmission Line Project.
Volume I, Prehistoric Archaeological Studies: The Pit River Uplands,
Madeline Plains, Honey Lake and Secret Valley, and Sierran Front Project
Segments . Report prepared for Sierra Pacific Power Company.
2010 Personal Communication
Young, D. C.
1997 Geology and Geomorphology. In Culture Change along the Eastern Sierra
Nevada/Cascade Front. Volume VI: Secret Valley.
Pp. 3-5. Report prepared for Tuscarora Gas Transmission Company, Reno.
Young, James, Raymond A. Evans, and Jack Major
1988 Sagebrush-steppe. In Terrestrial Vegetation of California . edited by Barbour,
Michael G. and Jack Major. California Native Plant Society, special printing
#9.
Yu, P.
2006 Pit Cooking and Intensification of Subsistence in the American Southwest and
Pacific Northwest . Unpublished Ph.D. dissertation, Department of
Anthropology, Southern Methodist University, Dallas Texas .
Zarillo, S. and B. Kooyman
2006 Evidence for Berry and Maize Processing on the Canadian Plains from Starch
Grain Analysis. American Antiquity 71, no. 3: 473-499.

Zdanowicz, C.M., G.A. Zielinski, and M.S. Germani
1999 Mount Mazama eruption: Calendrical age verified and atmosphere impact assessed. Geology 27, no. 7: 621-624.
177