..-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

...Eu-El

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 
 

 

 

 

 

 

 

 

 

 

 

 
 

 

 

 

 

 

 

.
V .V . . . . . .. . . . .
. . V. V a . . . . . . . . . . V .
. V . . . . . . . . . u . . . . .
. . .V s V. .V. . .. V .. . . V . V . . . .
. V . . V V o . .V. . . . .. . . . . .V . . .V .
_. . . . . . V . . V.V.. . . . V . V . . .
. V. .. . .. . .. V.. . . V. .V V. V . . . . V .
.. V V.V... . . .. . . _. . . . V . - V . .. . . ...
. .. . V . . . . . .V .- . . V .V . . . . .
V V V . . ...... .V V . . . : . . . . . .
.. . . . V . V V. . V V. V .. . V . V. . V . V .. . . V . V
V. V V .. ~ .V c . .. V. .V . V o . .. .. V . .. . V . . V . . .
. . V. .V . V . V.. . .. . V V . .. .. V. . . . V V . V .
V . . ..z. o V V V V. .. . .. . . .. . .. . . .. . . . .
. . . V... . V . V ._ a . .. . . . . . V . V.
V .. o . . V . o . o . V c . . . . V. c . .. u . . V
. .V . V ..q . V . . V .V V. V V V V . . V V . V . V
. . . . . V. . . . V. V. . . . . . . . . V . . . .
c . .. . . .. . V . . . . . V . .V . . V . .
... V V. . V. n . . . . . . . . .V .. V .. .V. . .. .
Q . .V . . .V V .V . .... . . . . V. . V . . . V V . V .
. V. . .... .. . uV V .. . o . . . . . . . . . .. V .
. . V. ..V . . . . ... . . .. . . .. . . V . . . V . .. V.
.. . V... . I. V . . . . . V V o T . V . . . . . .
.a . . . . V. . . .. -. .... V . . .. . . . .V V V
. .V . . . V .. V . . . V .V . V. . V . . . V V . . . . . V . . V .
. V. . V . . . ... . . V. . .. . . .V . . o . . V.
. .. . .V.. V . . . . V .. . . . . . . . V . . .V .
. . . V .... . . u. .. . V .. .. .. V . .V . V .V V .
. . .. V V. V . . . V. . V. . V V V ... V . . . . .V . V. V ..
. . V . . . V . . V V .. V .V. u . . ~ .
. . . . . . .. ... . . . .. . . .. . V V . . . .
.. . . . .V.. . V. .. V.V . . . . . V . .V . . . . . ..
... V V. . V . _ V . . . . V. nV . V . . . V. . . . .. .
. . . . Vs .. V. . .V .. .. . . . ~ ~ . V V. . . . V. V
. .V. . V ._ .V.. . V. . .. V . . V .. . . . ... . V . u
.. . n. .. . VV . V . . . .. . . u .. . .V . . . . . ..
. . V. V. . V . . . . .. . . . t . . .. . . . ... . V . . . u .. .
. . .V V. . V . . . . V . . V . . . ..
.. V. . .. . V . . . . V ... .V V- V .. V . V . .
. . V .. . . . V . . . V. . V.. . V. . . . . . .
. . . .. . . . V . . . . .V V . . . . V. . V. .
. . . I .. V. . . . . . ... ... V V .. . . .V . .... V . . .
. .V . . . V . .V. . . .. . V . . . ..V . V . . V c . V
.. . V . V. V. .V .. . . a . . u . .. . .. . . . . .V . . V .. .
. n . . V.. V .V . . V. V .. . .V . . ._ . . . V V. . V V . .
. . . V . V. . V. V. .. . . V.V .V . .. . . V . V V . .
V . V.. .V V. .. . V . V V . . . .n . . . . . . ... V.. V
. . V. .V. .. V.V . . . .V . V n . . V . . . . . . . . ..
. . . V . . . .. V . . V . . . . V.. . . V ...
. . .. . .V .. . . . V . V. v V. . . . . V . ..
. . V V. . .V. . . . .. V . . . V . . . V. .. . . V . . V . .
V .. . V . V V . V .. . . V .. . .. V V . v
. . .. . V . . . . . .... . . a . . V .
V . V . .. .. . . . V . .. .. . . . . . . . . V. . .
. . .. ... . .. . . . . V . . V. V . . . . . . V . . . V
. . _ V. . V . , . .V . n V. . . . . V . . V . . . .
V . VV .. . . . V V q . u . V. . .V. ..
.V . .V V . VV . . . . V. . V . . . . V
. V V. . . . . . . ... V .V ..
. V V .. . . . .. . ... .. . . . V V V . . V V
V . . .. V. .V . . . .. V . . . . .V . . . . V.
.. . .... . VV . . . . ... . . V . . . . . V .. . . .
. V . V. V .V.. . . . . V. . . .V . ... . . V . . . .
. . o . . V . V . V V. . . . V . . . n V . . V . .. V . V
n. V . . V. . . ... . V . . . . o - . V . .
. . . .. .V . V . . . . V .. . . .. o . . . u . u.
. . . v . . . .. . .V .. .. . . V . .
. V . V. . . .. . V V V V . . V. .. . n. V . V ..
V. .. .V . .V V . . . ... .. . t . . V . V . . V n
. V .. . . . V V. . V V V . . . . . V .
. . .V . V V . . . . .. . V ... V V V V . .V
V. . V V V .. .V . . .. . V . . .. . . .
V . I .V V . V . v. .V .V . . .
V V . .. .- V . . . V. .. .. . . .
. .V V . . .V . . V V . .V . u. a .
. . V. ... . . V . ... . V V. . VV . . . . V .. . V
. .. . . .. .. . I .V. V V . .. . .. .. V . V. . . . . .
. . . V V V V. V . . ... .. .V. .
. . V V . V . . .V .V V . . . . V
V . o . .V . V . . . . .. V . .. . . V
V . V. .V . .. V . . . V . . . . . .. .
. .. . V .. V. .. ..
. V . . . . . V. .. . V . V .. . . . . .
V . V V g. . . . V . .
V . V . . .. . V .. V . V . V . . V. V. ..
.V V . . . . . . ~ .. .
V . .. V . .. V. . . . V . n. V . . .
V V. . V . .. V . V . . . . . V .
. . . . . . . . V. . V .. .. . .. .
V V . .. .. . . . . .
V V . . n
V . .. V . . . . V. . V . . . V
. . . V . . . . V V . .
. V . V . . . . . .
. . . .. . . V .. . . . . . . . V .
. V V . . V V . . . . . . . . . V. . V
. . . . V. . V . .
V . V . . . ..
. . . . . .. V
. . . V . V V . V V . . .
. _ . .. . . .
. . . . . . V V
. . . . .
. . V . V . . .
. . . . . .
. . . V.

 

 

 

 

 

I
—

 

.
. . . .
_
V . .
. V
. . . .V
. .
. . .V . V. ... .
V. o . . . . .
. .V. . . V
. . . . . V V. V
c a V . . V
o . V
. . . . .
V . u .
V . . . ..
.. . V . . .
V. V. .. .
< V . . . .
V . .. . V
.. V .. .
.. . .V. V V. .V .
. . ... . . V
. .. . . .
. .. . . V .
V. V .. .V. ..
.. V. . . . V. .
. . . V
. . .. . ..
. . .V .
. . . . V.
. . . . V
. . . V . . ‘ V .
. . . .V. V .
. . . . . .
.. . . . . . .
V . V. .
. V . ~ V
V .. . .. .
V . a V . . V
. . . .
. V
. V. . . V .
V . ~.
. V . .
VV . . . .V . .
V . . V. V
. . . V
. V
. . V
. V. V V .
. . .
.
.
V .. V..

LIBRARY
Mic‘mgan State
University

—‘.

This is to certify that the
dissertation entitled

ASSESSING CARBONIZED ARCHAEOLOGICAL COOKING
RESIDUES: EVALUATION OF MAIZE PHYTOLITH TAPHONOMY
AND DENSITY THROUGH EXPERIMENTAL RESIDUE ANALYSIS

presented by

Maria E. Raviele

has been accepted towards fulfillment
of the requirements for the

PhD. degree In Anthrdpology

 

  

Lot ILI

Major Professor’ 8 Signature

‘11 goam‘e £20”)

Date

MSU is an Affirmative Action/Equal Opportunity Employer

— ¢ ..—._‘—.—.-.-.-.-.-.-.----.-.-.----.---.-.--.-.-.-a-.-.—

 

 

 

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE DATE DUE

 

AUG 2 O 2012

 

0
r‘ I. u
A

fiE‘L‘ if a 71310..

 

‘-

1124 14

 

.‘I'JB 1. 0 201.5

 

AUG 9352 2015

 

 

 

 

 

 

 

 

 

5’08 KilProlechroflCIRCIDateDuejndd

 

 

ASSESSING CARBONIZED ARCHAEOLOGICAL COOKING RESIDUES:
EVALUATION OF MAIZE PHYTOLITH TAPHONOMY AND DENSITY
THROUGH EXPERIMENTAL RESIDUE ANALYSIS
By

Maria E. Raviele

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Anthropology

2010

ABSTRACT
ASSESSING CARBONIZED ARCHAEOLOGICAL COOKING RESIDUES:
EVALUATION OF MAIZE PHYTOLITH TAPHONOMY AND DENSITY
THROUGH EXPERIMENTAL RESIDUE ANALYSIS
By

Maria E. Raviele

The timing of initial cultigen utilization by hunter-gatherers and early
horticultural populations remains a central research issue in understanding the transition
to low level food production. Continuing refinement of crop histories leads to a better
understanding of the processes, timing, and speed of domestication as well as human use
of cultigens and potential cultigens. The adoption of domesticates also speaks to inter-
and intra- group social interactions; knowledge of a cultigen does not necessarily arise
independently in each hunter-gatherer group which utilizes it. The latter is especially true
if the domesticate is non-native, as maize is to eastern North America, and must adapt to
an environment if/when it is introduced.

To understand these processes this research focused on two major research
goals. First, was to further establish the timing of the adoption of domesticates,
specifically maize, in the western Great Lakes region of eastern North America by using
a case study from the Saginaw Valley of Michigan. The use of phytolith and starch
analysis on ceramic residues offers another line of direct evidence for paleodiet. Through
phytolith and starch analysis of ceramic residues from Middle (200 BC-AD 500) and
Late Woodland (AD SOO—AD 1400) period sites in Michigan, I tested the current

hypothesis that maize horticulture is not ubiquitous in the Great Lakes until c. AD 1000.

Secondly, experimental residues were created to test for the presence of
phytoliths and starches. The goal of these experiments were threefold: to determine
phytolith and starch are presence in residues created for experimental purposes, to
determine if quantification of the percentage of maize present is possible through
phytolith density, and to determine if one can differentiate between the use of green and
mature maize through the presence of phytoliths and starches. Residues were created
using increasing 20% increments (0-100%) of maize. Green and mature maize were also
used separately in experimental residue creation. Results from the experimental residues
were applied to the interpretation of the archaeological residues used in this study.

Analysis and systematic direct dating of the archaeological residues indicates
that maize is present during the early Middle Woodland in the Saginaw Valley prior to c.
AD 1000. However, application of the experimental results to the interpretation of the
archaeological results indicates that it is a dried version, either kernels or flour; there is
no evidence for low level maize production. This inference is corroborated by
environmental evidence which indicates increased incidence of flooding during the early
Middle Woodland which would have made flood plain maize cultivation untenable. It is
hypothesized that the presence of maize c. 2000 years ago is due to down-the-line
exchange with nearby regional populations. The subsistence shift towards more intensive
maize cultivation, occurring during the Late Woodland period c. 1000 AD, appears to be
due to environmental change brought on by the Medieval Warm Period. Environmental .
change caused by the Medieval Warm Period resulted in a shift in the structure of

available resources requiring a change in subsistence strategies.

Copyright by
MARIA ELENA RAVIELE
2010

To the memory of my grandfather James W. Bain, one of the wisest men I have known.

“You have to earn it. ” James W. Bain

ACKNOWLEDGMENTS

There are many people that need to be thanked for their help in the completion
of this dissertation. First and foremost is my dissertation committee. My chair and
advisor Dr. William Lovis was incredibly supportive throughout the entire process of my
dissertation. I also would not have made it to this final step in the graduate program
without Bill’s guidance and support fi'om the beginning of my graduate career. I have
subjected Bill to office drop-ins (with sometimes complicated questions), reading
numerous paper drafts and panicked (and not so panicked) emails in regards to classes,
funding, and career opportunities. Throughout all this he has provided feedback and
guidance with humor and patience. I am happy to consider Bill not only a mentor but also

a friend.

The other members of my committee include Dr. Jodie O’Gorman, Dr. Lynne
Goldstein, and Dr. Catherine Yansa. Both Jodie and Lynne have challenged me to think
more holistically in regards to the interpretation of archaeological data. Their feedback
has helped make me a better archaeologist. Catherine has played a huge role in my
understanding of environmental reconstruction. I am also exceedingly grateful to
Catherine for her generosity in sharing lab space. This enabled me to store materials and
most importantly, perform the required processing. Dr. Alison Rautman aided in
clarifying my thinking on the transition to horticulture and the processes involved therein.
Other department faculty who supported me in various ways includes Drs. Helen Pollard,
John Norder, Marietta Baba, and the late Peg Holman and Susan Krouse. External to

MSU, Dr. John Hart offered constructive feedback and I need to thank Drs. Deborah

vi

Pearsall and Susan Mulholland for their willingness to train me in phytolith and starch

extraction and identification without which this project would never have happened.

I must also acknowledge the institutions and staff which granted me permission
to use their collections for my research as well as providing funding. This includes Karen
O’Brien at the University of Michigan Museum of Anthropology who was more than
accommodating, Barbara Mead at the Office of the State Archaeologist in Lansing who
helped identify residues which might be usefitl in my study, and Jeff Sommer at the
Historical Society of Saginaw County who suggested collections which might be useful
as well as allowing me to use materials from his ongoing excavations in the Shiawassee
National Wildlife Refuge. Major funding for my work was provided through grants from
the National Science Foundation and the Wenner-Gren Foundation while the American

Indian Studies Program at MSU provided funding for initial research trips.

Of course, none of this would have been possible without the support of my
friends and family. My parents have always supported me in my desire to become an
archaeologist and in attaining my PhD and for that I am eternally grateful. My friends
have helped keep me sane throughout the entire process of graduate school. I especially
need to thank Megan (and the pocket advisor) and Charlotte for PFLG/general silliness,
Marita for always being there with popcorn and wine, and Marieka for a Shared
perspective. Finally, many many thanks to Josh for putting up with general grumpiness,
late night writing frenzies, and long work weeks. I know I was not always pleasant to live
with during my writing year and I am everlastingly gratefiil for Josh’s patience and

awareness that this was only a temporary situation. I hope I am as kind in the future.

vii

TABLE OF CONTENTS

LIST OF TABLES .................................................................................. xi
LIST OF FIGURES ................................................................................ xiii
CHAPTER I
INTRODUCTION .................................................................................. I
Introduction ................................................................................. I
Background on Terminology ............................................................... 4
Research Problem .......................................................................... 12
Predicted Outcome and Expected Research Gain ..................................... 15
Organization of the Dissertation ......................................................... 18
CHAPTER 2
ORIGINS OF AGRICULTURE IN EASTERN NORTH AMERICA ..................... 19
Introduction ................................................................................. 19
Overview of Eastern North America .................................................... 20
Brief History of Research into the Origins of Agriculture in ENA ........ 20
Crop Histories: Eastern Agricultural Complex ............................... 23
Crop Histories: Maize ............................................................ 26
History of Origins of Agriculture Research within Michigan ........................ 29
Geologic and Geographic Setting of Michigan ............................... 29
Research on prehistoric agriculture in Michigan .............................. 34
CHAPTER 3
EVIDENCE FOR PLANT USE .................................................................. 42
Introduction ................................................................................. 42
Direct Evidence ............................................................................. 43
Macrobotanicals ................................................................... 43
Pollen ............................................................................... 47
Starch ............................................................................... 49
Phytoliths ........................................................................... 54
Indirect Evidence ........................................................................... 57
Stable Isotopes ..................................................................... S7
Radiocarbon Dating ............................................................... 61
CHAPTER 4
EXPERIMENTAL WORK ........................................................................ 64
Introduction .................................................................................. 64
Principles of Phytolith Production ....................................................... 64
Research Design ............................................................................ 66
Experimental Residues: Methods and Materials ....................................... 67
Processing of Each Maize Form ......................................................... 68
Hand Ground Maize Flour ....................................................... 68

viii

Lightly Pounded Dried Kernels ................................................. 70

Whole Dried Maize Kernels ..................................................... 70
Green Maize Kernels Removed by Hand ..................................... 71

Whole Cob Green Maize ......................................................... 7 I

Laboratory Processing Procedures ...................................................... 72
Experimental Residue Results ............................................................ 74

Discussion: Phytolith and Starch Taphonomy .......................................... 77
Conclusion .................................................................................. 79

CHAPTER 5

ARCHAEOLOGICAL RESIDUE ANALYSIS ................................................ 83

Introduction ................................................................................. 83

Site and Ceramic Descriptions .......................... . .................................. 86
Liberty Bridge (ZOBY77): Middle and Late Woodland.......................86

Fletcher (ZOBY28): Middle and Late Woodland .............................. 90

Kantzler (ZOBY30): Middle and Late Woodland ............................. 93
208A1276: Middle and Late Woodland ....................................... 96

ZOSAIZS 1: Middle and Late Woodland ....................................... 98

Schultz (ZOSAZ): Middle and Late Woodland ................................ 99

Tobico Marsh Sites (ZOBYI92 and ZOBYZIO): Late Woodland .......... 105

Simons 249 (ZOSA66S): Late Woodland ................................................. 108

Solms (ZOSA57): Late Woodland ............................................................ I 10
Stadelmeyer (208A 195): Late Woodland ................................................ I I 1

Foster (ZOSA74): Late Woodland ............................................. I 13

Clunie (2OSA722): Late Woodland ........................................... I 15

Summary of New Phytolith and Starch Subsistence Data ........................... I 16

CHAPTER 6

SAGINAW DRAINAGE CONNECTIONS TO THE GREAT LAKES REGION ...... I 19
Introduction ............................................................................... I I9
Explanations for Maize in Residues .................................................... 121
Contamination ................................................................... 121

Low-level Cultivation of Maize in the Saginaw Drainage ................. 122

Exchange of Maize in Dried Fonn(s) ......................................... 124

Evidence for Middle Woodland Maize ................................................ 126

Southern Ontario ................................................................. 126

Ohio ............................................................................... 127

Indiana ............................................................................ 128

Illinois ............................................................................. 129

Evidence for Interregional Influence in Michigan .................................... 130
Summary ................................................................................... 133

CHAPTER 7

CONCLUSION .................................................................................... 138
Research Results .......................................................................... 145

Future Research ........................................................................... 148

ix

APPENDICES .................................................................................... I 5 I

Appendix A: Laboratory Protocol for Phytolith Extraction ........................ 151
Appendix B: Artifact Photographs ...................................................... 156
Appendix C: Phytolith and Starch Analysis Results.....................................I94
Appendix D: Radiocarbon Date Results ............................................... 218
REFERENCES .................................................................................... 221

LIST OF TABLES

Table 2-1: Early cucurbit finds in Eastern North America .................................... 25
Table 2-2: Summary data of early maize ......................................................... 28
Table 4-1: Maize form and processing used in experimental residues ....................... 68
Table 4-2: Microbotanical results of experimental residues ................................... 75
Table 4-3: Probable utilization of maize based on presence/absence of diagnostics ....... 79
Table 5-1: Middle and Late Woodland sites examined for residues .......................... 85
Table 5-2: Macrobotanical data for Liberty Bridge ............................................ 90
Table 5-3: Phytolith and starch data for Liberty Bridge ...................................... 90
Table 5-4: Macrobotanical data for Fletcher .................................................... 92
Table 5-5: Macrobotanical data for Kantzler ................................................... 95
Table 5-6: Phytolith and starch data for Kantzler .............................................. 96
Table 5-7: Phytolith and starch data for 208A 1276 ............................................ 98
Table 5-8: Macrobotanical data for Schultz .................................................... 103
Table 5-9: Phytolith and starch data for Schultz .............................................. 103
Table 5-10: Radiocarbon samples from Schultz ............................................... 105
Table 5-11: Macrobotanical remains from ZOBYZIO ......................................... 108
Table 5-12: Phytolith and starch data for ZOBY210 .......................................... 108
Table 5-13: Phytolith and starch data for 208A665 .......................................... 110
Table 5-14: Phytolith and starch results for Solms ............................................ 110
Table 5-15: Macrobotanical data for Stadelmeyer ............................................ 113'
Table 5-16: Phytolith and starch data for Stadelmeyer ....................................... 113
Table 5-17: Phytolith and starch data for Foster .............................................. 114

xi

Table 5-18: Macrobotanical data for Clunie ................................................... 116

Table 5-19: Phytolith and starch data for Clunie ............................................. 116
Table 5-20: Radiocarbon dates with plants identified from split residues ................. 118
Table C-I: Phytolith and starch analysis results ................................................ 194
Table D-l: Radiocarbon date results ............................................................ 215

xii

LIST OF FIGURES

Figure l- 1: Saginaw Drainage study area ........................................................ 2
Figure 2-1: Approximate division of ecotones in Michigan ................................... 30
Figure 3-1: Major starch grain features (recreated after Gott et a1. 2006) ................... 52
Figure 4-1: Student volunteers hand grinding maize ........................................... 69
Figure 4-2: Carbonization of experimental residues (in progress) ........................... 69
Figure 4-3: Diagnostic maize rondel ............................................................. 77
Figure 4-4: Undamaged diagnostic maize starch ................................................ 78
Figure 4-5: Maize starch with evidence for grinding ........................................... 79
Figure 5-1: Site locations (figure courtesy of Jon Carroll) .................................... 87
Figure 5-2: Maize starch with possible evidence for grinding ................................ 98
Figure 5-3: Diagnostic wild rice sheath phytolith from Schultz ............................. 104
Figure 5-4: Maize starch, polarized and bright field views .................................. 104
Figure 5-5: Yellow Pond Lily ..................................................................... l 11
Figure 6-1: Sites with maize dating to the Middle Woodland period (figure courtesy of
Jon Carroll) ......................................................................................... 120
Figure 6-2: River routes from southwest Michigan to the Saginaw drainage .............. 136
Figure BI: N10E8.ISE-4 ......................................................................... 156
Figure 82: 2.NOE0.28E-3 ........................................................................ 156
Figure B3: 2.NOE0.2NE-2 ........................................................................ 157
Figure B4: 77127 .................................................................................. 157
Figure BS: 77112 .................................................................................. 158
Figure B6: 77102 .................................................................................. 158

xiii

Figure 87: 77087 .................................................................................. 159

Figure B8: FS 394 .................................................................................. 159
Figure B9: FS 390 ................................................................................. 160
Figure BIO: FS 356 ................................................................................ 160
Figure 811: FS 305 ................................................................................ 161
Figure 812: FS 192 ................................................................................ 161
Figure 813: FS 153 ................................................................................ 162
Figure 814: Shiawassee #2 .............................................................................................. 162
Figure BIS: 65888 ................................................................................. 163
Figure 816: 65826 ................................................................................. 163
Figure 817: 65557 ................................................................................. 164
Figure 818: 65409 ................................................................................. 164
Figure 819: 64718 ................................................................................. 165
Figure 820: 64676 ................................................................................. 165
Figure 821: 64622 ................................................................................. 166
Figure 822: 66361 ................................................................................. 166
Figure 823: 66160 ................................................................................. 167
Figure B24: 66020 ................................................................................. 167
Figure 825: 66004 ................................................................................. 168
Figure 826: 66003 ................................................................................. 168
Figure 827: 65961 ................................................................................. 169
Figure 828: 65920 .......................................................................... i ....... 169
Figure 829: 65908 ................................................................................. I70

xiv

Figure 830:
Figure 831:
Figure 832:
Figure 833:
Figure 834:
Figure 835:
Figure 836:
Figure 837:
Figure 838:
Figure 839:
Figure 840:
Figure 841:
Figure 842:
Figure 843:
Figure 844:
Figure 845:
Figure 846:
Figure 847:
Figure 848:
Figure 849:
Figure 850:
Figure 851:

Figure 852:

67373 ................................................................................. 170
67345 ................................................................................. 171
67674 ................................................................................. 171
67469 ................................................................................. 172
67275 ................................................................................. 172
67207 .......................................... ‘ ....................................... 173
66839 ................................................................................. 173
66656 ................................................................................. 174
66656 ................................................................................. 174
69552 ................................................................................. 175
69477 ................................................................................. 175
69276 ................................................................................. 176
69030 ................................................................................. 176
63567 ................................................................................. 177
68559 ................................................................................. 177
67916 ................................................................................. 173
67396 ................................................................................. 179
67232 ................................................................................. 130
F29 .................................................................................... 130
69771 ................................................................................. 131
69743 .......................................................................... . ....... 131
69760 ................................................................................. 132
69666 ................................................................................. 132

XV

Figure 853:
Figure 854:
Figure 855:
Figure 856:
Figure 857:
Figure 858:
Figure 859:
Figure 860:
Figure 861:
Figure 862:
Figure 863:
Figure 864:
Figure 865:
Figure 866:
Figure 867:
Figure 868:
Figure 869:
Figure 870:
Figure 871:

Figure 1372:

69606 ................................................................................. 183
208Y210 ............................................................................. 183
Vessel 92 ............................................................................. 184
SAC 70 #1 ........................................................................... 184
SAC 70 #2 ........................................................................... 185
SAC 35 .............................................................................. 185
Vessel 435 ........................................................................... 186
SAC 31 .............................................................................. 187
SAC 15 #1 ........................................................................... 187
SAC 15 #2 ........................................................................... 188
76986 ................................................................................. 188
73995 ................................................................................. 189
73970 ................................................................................. 189
73949 ................................................................................. 190
73942 ................................................................................. 190
73931 ................................................................................. 191
76639 ................................................................................. 191
76643 ................................................................................. 192
2861 ............................................................................................................. 192
FS 433 ................................................................................ 193

xvi

CHAPTER 1
Introduction

Introduction

The research addressed within this dissertation consists of two major
components. The first component seeks to explore the transition to horticulture in eastern
North America (ENA), specifically with regard to the inception and use of maize,
through phytolith and starch analysis of ceramic residues. Phytoliths are microscopic
deposits of silica that form in and between the cells of plants and can be found to varying
degrees in each plant structure (Mulholland and Prior I993; Pearsall 2000; Pipemo
2006). Starch is produced in all plants and starches derived from long-term storage
organs are often diagnostic (Gott et a1. 2006; Torrence 2006). I investigate the timing of
the adoption of domesticates in mixed subsistence horticultural-hunter/gatherer societies
(referred to as low-level food producers) and the concomitant changes, if any, in the
resulting seasonal settlement-subsistence round. The adOption of cultigens or
domesticated plants by a hunter-gatherer society can have impacts on numerous aspects
of hunter-gatherer life including, but not limited to, scheduling of the seasonal round,
gender roles and division of labor, and social relations. The Saginaw drainage of
Michigan (Figure 1-1) will be used as a case study to explore these changes associated
with the transition to maize horticulture. It Should be noted that while this research
largely focuses on the use of maize, the analysis identifies all diagnostic plant phytoliths
and starches.

The second major component of this research is an analysis of experimentally

created residues utilizing various forms of maize. The incorporation of experimentally

Figure N: Saginaw drainage study area

 

 

 

. . _. A...”
.1 ." . ”.4...
Lakes V - I‘
Meters Above Sea Level - _. . ' . ~
014:550 ‘ r ;.' ‘ .

I 501-550 ' . * H
- 451-500 - . , , _ “’0“

fi 401-450 “"‘j

[__1 351-400 ~ , . _.
. "9’: it
301.350 Lake ’ '
LI 251.300 Michigan
[_] 201-250 a, - .
.0200 - -. j. - ' __ v ._ z.

 

 

 

 

 

I t . I.
21.!" -.t I
.: Ii?"
’1' ‘
I ‘ ii
’1 .‘r

0 40 80 . -. . ~ = ..r
Kilometers I I ' r -'

 

produced residues is a direct result of the work done by Drs. William Lovis, John Hart,
Gerald Urquhart, and Robert Feranac and their analysis of the masking of C. North

American resources (i.e. maize) by C3 North American resources. The experimental

 

residues present the opportunity to determine if the percentage of maize being cooked in
a pot, based on phytolith concentration, can be predicted. If quantification is possible,
this would provide a baseline to assess the percentage of maize in archaeological
residues. In general, however, the experimental residues will allow for the assessment of
phytolith and starch taphonomy and its implications for understanding the interpretation
of data generated via phytolith and starch residue analysis. It should be noted that at the
time of research design, starch identification was not a component of this project. Starch
identification was added once it was determined through analysis of the experimental
residues that starch remained in the samples.

Since the issue of maize timing is one of the important variables included in this
work, there is substantial employment of AMS dating. While the advent of AMS dating
of macrobotanicals has substantially refined the chronology, timing and history of maize
adoption, the systematic regional utilization of phytolith and starch analysis as
independent lines of evidence has not been as widespread. Examination of cooking
residues present on ceramics for phytoliths and starches provides the possibility of
producing earlier direct evidence of cultigen use not found in the macrobotanical record
and is an alternative means to investigating low-level food production (Hart et al. 2007;
Mulholland and Prior 1993; Mulholland, Rapp, and Ollendorf 1988; Pearsall 2000;
Pipemo 2006).

For the purpose of this research, low-level food production is defined as a
mixed subsistence economy which utilizes cultigens as a supplement to wild plant and
animal resources. The term agriculture is not utilized as this implies the intensive field

cultivation of crops. While the adoption of cultigens may result in either situation, the

nature of the use of domesticates in the proposed study area of the Saginaw drainage of
Michigan is more in line with the definition of low-level food production given above.
As this project focuses on the ‘middle ground’ (Smith 2001) between hunter-gatherers
and full scale agriculturalists, the time periods to be studied are the Middle (2008C—AD
500) and Late Woodland (AD 600-AD I650) periods. It should be noted that the Middle
Woodland period coincides with the Hopewell manifestation throughout the Midwest.
This manifestation varies but is generally characterized by Si gnificant mound
construction, exotic raw materials (obsidian, Gulf coast shell, copper, minerals such as
galena and mica), and elaborate burial customs (Maxwell 1952; Morgan 1952; Seeman

I992; Wymer 1987).

Background on Terminology

When discussing the transition to horticulture, a distinction between primary
and secondary origins of agriculture is required. As defined by Cowan and Watson
(1992:207), a primary origin area is one in which plant domestication and
horticulture/agriculture occurs without the outside introduction of crop staples. An area
of secondary origin has two dominant patterns; one pattern is defined by the introduction
of crop staples (either wholesale or one by one) which replace a foraging system while
the other pattern is characterized by the introduction of new domesticates into an area that
is already dependent on cultivated plants to some extent (Cowan and Watson:207-208).
The latter pattern characteristic of a secondary area for the origins of agriculture is
directly pertinent to ENA where EAC cultigens are replaced by maize.

In addition to the distinction between primary and secondary origins, there are

three separate and often related processes in the transition to horticulture; the process of

plant domestication, the transition itself (i.e. adoption), and subsistence intensification.
The process of plant domestication, in a primary agricultural origin context, involves the
isolation of the planted/raised vegetation from the wild members of their species and the
continued annual cycle of harvesting and storing seed (Hayden 1995:274; Smith 1995;
Wills 1988). Involved in this process is the selection of desirable characteristics, such as
larger seed size and thinner seed coat, by the people raising these plants (Hayden 1995).
Most importantly, it is the intentionality of this process that is significant (Hayden 1995;
Smith 1995; Wills I988). Domestication does not, as Smith (2001:17) states, “...define
the boundary of either agriculture or hunting and gathering...” but instead acts as a point
from which to view either side.

The decision to utilize (i.e. adopt) cultigens, initially as a supplement to a
hunter-gatherer subsistence system, can be considered a characteristic of the transition to
horticulture for either a primary or secondary agricultural origin area (Smith 2001; Wills
1988). This transition is better described as low-level food production (Smith 2001)
instead of horticulture; reasons why are discussed below. Often, but not always, tied to
the adoption of cultigens and low-level food production is the process of intensification,
either on these cultigens or on an already utilized wild resource. Intensification within a
subsistence system may also include using smaller animals, more diverse species, and the
storage of resources with abundant production but a limited window of access (Wills
1988:44). In general, intensification occurs on secondary resources, which fit easily into
the already existing subsistence system (Lovis et a1. 2001; Wills 1988). As Wills
(1988:36) states, “. . .foragers adOpt domesticated plants not to become farmers but to

remain effective foragers.” These resources act to diversify subsistence and buffer against

scarcity (Lovis et a1. 2001; O’Shea 1989).

An important distinction to make is the difference between cultigen and
domesticate. These are both descriptors for the intensity of use of a plant species. These
descriptors can have a variety of meanings depending on its context and the theoretical
orientation of the researcher. However, the use of the term domesticate implies a plant
primarily under human control and reliant upon human intervention for survival and
reproduction. The changes present in a domesticate manifest themselves genetically and
phenotypically (Ford 1985; Harris 1996; Rindos 1984). Domesticate most often refers to
plants that have become morphologically and genetically distinct from their wild
ancestors (Vrydaghs and Denham 200727). These morphological distinctions are a result
of intentional selection by humans for certain desirable traits, which commonly include
larger seed size, more compacted seed heads, thinner seed coats, and a brittle rachis to
enable easier harvesting (Vrydaghs and Denham 2007:3).

Cultigen, however, is used in reference to a variety of plant foods ranging from
domesticates (in the sense described above) to wild plants which may be tended and/or
encouraged by people. These wild plants do not rely on human intervention for survival;
rather humans will create environmental conditions, sometimes intentionally but
sometimes not, which are beneficial to the grth and prOpagation of these wild plants.
Not all human interaction and/or intervention with plants results in domestication and not
all human intervention in the life cycle of a plant is the result of agricultural behaviors.

Hunter-gatherers are often active managers of resources (Wagner 2003: 1 27).

Ethnographic accounts and archaeological evidence indicate hunter-gatherers
and low-level food producers encouraged wild plants via pruning, clearing, and the
occasional low-intensity burning to enable better growth, clear trails for hunting and
travel, or to increase browse for game animals (Wagner 2003). For these reasons,
cultigen is the preferred term used in this dissertation when discussing plant management
techniques and strategies.

The issue of intensification brings with it a consideration of how to identify
between three macro-level distinctions in regards to reliance on cultivated plants. From
most reliant/intensive to least reliant/intensive, these distinctions are agriculture,
horticulture, and low-level food production. Each term has distinct implications in
regards to how reliant a pOpulation is on cultivated plant resources and has been
characterized in a number of different ways (c.f. Denham, Iriarte, and Vrydaghs (eds.)
2007; Cowan and Watson (eds.) 1992; Price and Gebauer (eds.) 1995; Smith 1992;
Struever (ed.) 1971) in the literature on the origins of agriculture. The presence of
domesticates and wild cultigens (hereafter referred to collectively as culti gens) does not
always indicate the same level of subsistence reliance and/or intensification on these
plants. As demonstrated in the discussion below, the use of culti gens can result in a
variety of subsistence adaptations.

Agriculture is most commonly thought of when one thinks of domesticated
plant use and does not include wild cultigens. There is an intrinsic implication of
intensive field cultivation of crops, which may or may not include the use of draft
animals and the plow. Intensive field cultivation, often of monocrops, can result in faster

soil depletion than what might naturally occur. The archaeological implication for

 

intensive field cultivation is that agriculturalists may have to clear fields more often or
rotate fields in a fallow cycle to allow for soil regeneration. In addition to formal large-
scale cleared field cultivation, other cultural characteristics often associated with
agriculture include sedentism, permanent storage structures, social complexity, larger
population density, and (in some areas) large group aggregation (Bender 1978; Harris
1989, 2007; Keeley 1995). Agricultural communities rely predominantly on crop staples
and intensively cultivate the surrounding available arable land. Domesticates are the
primary means of subsistence for agricultural communities and dominate food remains
identified in the archaeological record.

Next on the scale of reliance/intensity from agriculture is horticulture.
Horticulture is the less intensive cultivation of crops, primarily as garden-level food
production. Plot size for cultivation is smaller than that found in agriculture.
Horticulture also includes greater biodiversity in regards to the types of crops grown.
Instead of monocropping, horticulturalists grow a range of plants in their gardens, which
better mirrors the biodiversity of the natural environment. Plants grown in these gardens
are cultigens, both domesticated and wild/weedy plant species, used for a variety of
purposes and notjust subsistence needs (Harris 2007; Keeley 1995).

The wild and/or weedy plants grown by horticulturalists are generally
characterized as “adventitious and spontaneously occurring” (Harris 2007:24).
Wild/weedy plants are valued because they have multiple useful parts including edible
fruit, flowers, or leaves, parts used for dyes or fibers, and medicinal applications (Harris
2007:24). As a result of this greater biodiversity, soil depletion progresses at a slower

rate than that which occurs with intensive agriculture and the need for clearing new plots

or allowing plots to lie fallow is diminished. Cultural characteristics that are often
associated with horticultural groups can include incipient social complexity, some form
of long-term storage, and some level of sedentism and group aggregation on an annual
basis (Bender 1978, 1981; Rosenberg I990).

The least intensive/reliant-upon-cultigens subsistence adaptation is low-level
food production. Smith (2001) coined this term to characterize societies which continue
to practice hunting and gathering but which also incorporate some level of plant
cultivation into their subsistence regime. In the traditional hunter-gatherer/farmer
subsistence dichotomy, these types of societies often proved to be a categorical
conundrum. Most often when societies characterized as in-between hunter-gatherers and
farmers have been considered archaeologically, this transition has been seen as a rapid
and short-lived cultural manifestation with farming as a permanent end product. Once
the line from hunting and gathering to farming had been crossed, there was no going back
(Smith 2001 :2-5). While Smith (2001) is certainly not the first archaeologist to consider,
what he terms, this ‘middle ground’ between hunter-gatherers and farmers (c.f. Ford
1985; Harris 1996, 2007; Rindos 1984; and Zvelebil and Lillie 2000), his
conceptualization of this grey area is the most straightforward and uncomplicated.

In general, hypotheses on the transition to horticulture have been framed in
terms of either the adoption of cultigens or the conditions necessary for the
transition/adoption to occur. These hypotheses include factors such as responses to social
relations with other groups, to environmental stress or climatic changes, to rising
population density, to resource abundance, to risk buffering strategies, and on the

availability of protodomesticates to a group (Bender 1978, 1981; Clark 1989; Hayden

1995; Price and Gebauer 1995; Stone and Downum 1999; Winterhalder and Goland
1997). While all of the factors listed above can play a role of variable importance in the
transition to horticulture, in each scenario, people’s initial use of cultigens/domesticates
is as a supplement to an existing hunter-gatherer diet (Wills 1988).

To deal with this period of mixed subsistence economy, Smith (2001) correctly
argues that there needs to be a tripartite (hunter-gatherer/low-level food
production/agriculturalist) instead of dichotomous (hunter-gatherer/agficulturalist)
framework for thinking about the origins of agriculture. It is important to note that these
transitional states, represented by hunter-gatherers as low-level food producers who
utilize cultigens as a supplement to their existing diet, or to be put more concisely, mobile
low-level food producers, have supporting evidence to indicate they were rather stable
and long-term adaptations instead of rapid and unstable, as has been usually assumed in
the literature (Smith 2001).

These are societies which utilize both wild (plant and animal) and domesticated
resources. They are not dependent on either resource form exclusively, nor is it possible
to determine their trajectory of development (i.e. ‘reverting’ back to hunting and
gathering or becoming farmers). Low-level food production is characterized by the
utilization, intensification and management of multiple types of resources in the
surrounding environment in an effort to maximize risk buffering strategies (Smith 2001;
Wills 1988;0’Shea 1989). Archaeological identification of low-level food producers
would demonstrate evidence for the use of multiple resources zones (aquatic/terrestrial,
uplands/lowlands), the use of resources both intensively and extensively (demonstrating a

broader resource base), and the identification of cultigens (whether domesticated or wild)

IO

(Smith 2001 :36). With the variety and abundance of resources present within ENA,
Smith’s criteria for low-level food producers offers a framework in which to reexamine
evidence for the useof domesticates in what have often been considered archaeological
backwaters. These regions have been characterized as such due to the perception that
nothing interesting (i.e. socially complex and stratified agriculturalists) really happened
there prehistorically.

As an example, evidence for low-level food production is demonstrated in
Parker’s (1996) discussion on the use of cultigens in the southern Great Lakes, which
demonstrates maize cultivation as part of a mixed strategy of crop production and wild
plant exploitation. Late Woodland inhabitants of the Great Lakes incorporated varying
elements of the Eastern Agricultural Complex (EAC) crops into traditional subsistence
patterns when it had previously been assumed that little use was made of cultigens
(Parker 1996). Low-level food production by hunter-gatherers as a type of subsistence
strategy is also corroborated by the chronological refinement on the use of cultigens,
especially EAC crops such as squash and chenopod in ENA. The dating of domesticated
squash to the Archaic period (8,000-5008C) is especially pertinent to the evidence for
mobile low-level food producing populations (Lovis and Monaghan 2008).
Chronological refinement on the use of cultigens as part of a nominally hunter-gatherer
diet has also consistently demonstrated an earlier use of culti gens.

A demonstration of this earlier use is in the study by Hart, Thompson and
Brumbach (2003) where phytolith analysis and AMS dating of cooking residues from
ceramics was used to find evidence of maize, squash, wild rice, and sedge being cooked

together during the 7th century AD. These data place maize and squash in New York at

11

least 350 years earlier than previously thought. The previously accepted date, of AD
1000, was based on macrobotanical evidence. Further research from Thompson et al.
(2004) in New York demonstrates the use of maize as part of a long trajectory of
intensification, rather than as a catalyst for intensification.

Maize is a domesticated imported non-native cultigen that requires tending,
protection by humans, and adaptation to a more northerly climate. The demonstration of
its earl ier-than-expected presence in New York extends the time period during which
non-native cultivated plants were being used in association with wild subsistence items
and native cultigens. Both types of cultivated plants, native and non-native, were acting
as a supplement to a traditional subsistence regime. A similar long pattern of co-use is
seen with the presence of squash in ENA during the Late Archaic (2000-500 BC).
Squash is utilized during the Late Archaic but does not become a significant food source
until the Middle Woodland (Lovis et al. 2001; Lovis and Monaghan 2008; Monaghan et
al. 2006; Scarry 2003; Wymer 1987). In sum, evidence from ENA seems to support the
idea of the gradual incorporation of cultigens and non-native domesticates into a hunter-
gatherer diet over a long period of time (thus causing them to be low-level food
producers), with subsequent increased intensification of their production (Kidder I992;

Lovis et al. 2001; Monaghan et al. 2006).

Research Problem

A long standing research question on the origins of agriculture/transition to
horticulture has been the initial timing of domesticate adoption and utilization by
(mobile) low-level food producers. Further refinement of crop histories leads to a better

understanding of the processes, timing, and speed of domestication as well as human use

12

of those plants that eventually became cultigens. The adoption of domesticates also
speaks to interactions between groups since knowledge of a cultigen does not
independently arise in each group which utilizes it. This is especially true if the
domesticate is non-native, as maize is to ENA, and must adapt to an environment if/when
it is introduced to a new area. The work conducted here clarifies the timing of the
adoption of domesticates, specifically maize, in the western Great Lakes region of ENA
through the case study of the Saginaw drainage of Michigan. The use ofphytolith and
starch analysis on ceramic residues offers other lines of direct evidence, besides
macrobotanicals, for paleodiet (See Table 1-1 for a summary of cultigens present at sites
in research area). As previous studies in New York have demonstrated (see Hart et al.
2003 and Thompson et al. 2004), phytolith analysis has pushed back the date for the
earliest use of maize in northern areas of eastern North America. Through phytolith and
starch analysis of ceramic residues, 1 will be testing the current accepted hypothesis that
maize horticulture is not ubiquitous in the Great Lakes until c. AD 1000.

The prevailing hypothesis on the use of maize horticulture in the Saginaw
drainage of Michigan states that it was not predominant until approximately AD 1000
(Brashler et al. 2000; Lovis et al. 1996, 2001; Parker 1996). This assertion is based
primarily on recovered archaeological macrobotanical material (predominantly from
flotation) and reconstructed environmental data based on pollen studies, soil types, and
modern maize climate requirements (Krakker 1983; Monaghan and Lovis 2005). Until
relatively recently, a lack of the systematic use of flotation has hindered the
understanding of the use of domesticates simply because seed evidence had been

primarily recovered when found in large concentrations, such as in storage pits, during

13

excavations.

The methods used to study the transition to horticulture in the Saginaw drainage
(and other areas of the western Great Lakes), have relied predominantly on the presence
of macrobotanical remains recovered during excavation or through the use of flotation.
Maize macrobotanicals recovered fi'om excavations within the study area have remained
at a low level of incidence despite the use of intensive flotation for the past thirty years.
Phytolith and starch analysis will offer new, independent lines of direct evidence to
document plant exploitation among the prehistoric inhabitants of the Saginaw drainage
by testing the current hypothesis that maize horticulture was not ubiquitous until ca. AD
1000 in the Saginaw drainage of Michigan. However, if maize is found to be regularly
present prior to AD 1000, several key questions arise:

I) In the terminology of Smith (2001) and Wills (1988), how much earlier is
maize being “plugged into” traditional subsistence practices? What implication, if any,
does this earlier use of maize have for the seasonal settlement-subsistence round
especially in regards to the division of labor, the location of people on the landscape, and
scheduling of resource procurement?

2) Is this earlier use continuing the trend of greater reliance on collective
resources, such as weedy annuals (a similar argument for collective use is the exploitation
of anadromous fish in the Great Lakes by Egan [1993])?

3) To what extent does the use of maize overlap with the utilization of Eastern
Agricultural Complex domesticates such as chenopod and squash in the westem‘Great
Lakes?

4) Is the low incidence of maize macrobotanicals anomalous or not? If it is not

14

anomalous, what does this mean for the relationship between mobile low-level food
producers and known maize horticulturalists from the surrounding area (Ontario, Ohio,

and Illinois)?

Predicted Outcome and Expected Research Gain

A primary purpose of this research is to further refine the understanding of the
settlement-subsistence system of Woodland period peoples as well as to refine the
chronology on the adoption of maize in eastern North America, specifically within the
southern and western Great Lakes region. The study area proposed for this research has a
different trajectory of domesticate utilization than that seen in the surrounding areas of
New York, Ontario, Ohio, Indiana, Illinois, and Wisconsin (Crawford and Smith 2003;
Scarry 2003). These regions all had pOpulations which utilized intensive maize
agriculture and examples include Mississippian, Fort Ancient, incipient Iroquoian
populations, and Oneota groups. Earlier adoption of the use of maize not only has
important implications for the crOp history, but also for the interpretation of mobile low-
level food producer behavior within the archaeological record. The use of phytolith and
starch analysis on ceramic residues offers other independent lines of direct evidence as to
the composition of paleodiet that does not have the same limitations as macrobotanicals
or the use of stable isotope analysis on residues. Phytoliths and starches also offer the
unique perspective of allowing the researcher to identify some of the plant foods being
cooked together. This allows for better understanding of nutritional composition and can
have implications for the scheduling of resource procurement and, perhaps, seasonality or
evidence for storage of specific food resources that may not be preserved as

macrobotanicals. When used in conjunction with extant faunal and floral data, a more

15

complete picture of Woodland period diet and the changes that occur throughout the
period is possible (e.g. Crawford et al. 2006).

The secondary purpose of this research is to determine if it is possible to
quantify the percentage of maize used/cooked in a pot as well as differentiate between the
use of green and mature maize. The utilization of controlled experimental residues test
this possibility. If there is a quantifiable difference in phytolith density and/or size
between green and mature maize, it may be possible to identify which is being utilized in
archaeological samples. The possibility of identifying which stage of maize is being
utilized could explain the low incidence of early macrobotanical remains as well as
refining the subsistence regime. The ability to determine the percentage of maize being
cooked in a pot has implications for the ‘importance’ of that cultigen in a prehistoric
mixed subsistence economy. The identification of starch in the experimental residues has
changed some of the original goals of the experimental residue analysis, but as is further
discussed in chapter four, identification of probable maize forms utilized is still achieved.

The working hypothesis of this study is that there will be archaeological
identification of maize prior to AD 1000 and this will extend the period of low—level food
production. Previous research in the Finger Lakes region of New York by Hart et al.
(2003) and Thompson et al. (2004) demonstrate the earlier presence of maize than
previously provided by macrobotanical evidence. Both the western Great Lakes and New
York were on a trajectory of resource intensification and it is highly likely that maize was
utilized earlier in the research area than the dates derived from macrobotanical evidence.
However, other outcomes are possible.

One predicted outcome is if early cultigen use is found, but cultigens were not

16

used intensively. Ideally my analysis of experimental residues for quantification of
phytolith densities would aid in addressing this issue of intensity as well. If quantification
of phytoliths is found to not be a viable means to assess intensity of use, an alternative
might be through the presence of cultigen phytoliths identified in ceramic residues but
absent as macrobotanicals in the site assemblage. The presence of phytoliths and/or
starches with the absence of macrobotanicals could indicate that while these cultigens
were utilized they were not intensively cultivated. This outcome could also be used to
argue for a scenario of maize exchange (either directly or down-the-line) between
foragers and maize horticulturalists.

A final outcome to consider would be if no cultigen plant phytoliths of any kind
are found. This would then fail to reject the current hypothesis that maize horticulture
was not ubiquitous until AD 1000. Finally, one other outcome that would be possible is
the presence of particular cultigens or intensified wetland resources with the absence of
phytolith producing cultigens that are expected to be present. This last outcome could
have implications for plant storage and seasonality depending on the site context.

While previous research has focused on the transition to horticulture (Egan
1993), and the use of maize in Michigan (Krakker 1983), the utilization of phytolith and
starch analysis as independent lines of evidence has not been employed. The presence of
macrobotanicals at a site is dependent on both site preservation processes as well as
recovery techniques. Analysis of ceramic residues for stable isotopes can provide an
overall indication of dietary composition but there is the drastic masking of the presence
of C4 plants by C3 plants (Hart et al. 2007). By analyzing the phytoliths and starches

present in ceramic residues, a direct analysis of what plants were being consumed is

17

 

possible. This is made even stronger when used in conjunction with the macrobotanicals
present at a site since not all food plants are phytolith producers and not all starches are
diagnostic (Pearsall 2000; Pipemo 2006). Direct AMS dating of the residue also allows

for tighter control on the chronology of the use of maize and other cultigens.

Organization of the Dissertation

To address the above questions, the dissertation has been organized along the
following lines. Chapter two is an overview chapter on the origins of agriculture within
eastern North America. This includes a discussion of both EAC and maize utilization, as
well as a Michigan specific overview regarding the explanatory mechanisms for low-
level food production and the switch to the reliance on maize production. Chapter three
is a discussion of the various lines of evidence, both direct and indirect, archaeologists
utilize in the study of paleodiet and subsistence. Chapter four encompasses the
experimental work conducted for this study while chapter five provides the background
for the archaeological sites used in this work and the results of my archaeological residue
analyses. Chapter six provides the larger archaeological context for the results of my
archaeological residue analysis. The results of the experimental work discussed in
chapter four are also incorporated in chapter six to aid in the interpretation of the
archaeological residue results. Finally, chapter seven is the conclusion with my final
thoughts and suggestions for future research; for me, this study has raised many more

questions than it has answered.

18

CHAPTER 2

Origins of Agriculture in Eastern North America

Introduction

The geographic definition and boundary of eastern North America (ENA) as it
relates to the archaeology of the region roughly follows the extent of deciduous forest
starting at the Atlantic coast and moving west to the Ozarks where grassland becomes the
predominant landscape. The northern limits of this area are characterized by a mixed
deciduous-conifer landscape with shorter growing seasons and colder winters south of the
boreal forest and sub-arctic (Gremillion 2003:19). Major plant communities include
mast-rich oak-hickory, oak-chestnut, northern hardwoods, southern oak-pine, and
numerous wetland and river edge species (Gremillion 2003220). Animal taxa present
reflect the available plant diversity and includes terrestrial game, migratory fowl, and
marine, lacustrine, and riverine fish and mollusk species.

Such diversity in both plant and animal resources enabled people to develop a
number of subsistence systems including sedentary, socially complex agriculturalists and
groups that remained hunter-gatherers until EurOpean-American contact. In addition to
these two ‘extreme’ ends of the subsistence spectrum, there is a range of groups which
practiced, to certain degrees, hunting and gathering in conjunction with the utilization of
domesticated plants. The groups located within this ‘middle ground’ of subsistence
activity are best characterized as low-level food producers (Smith 2001). This variety of
cultural adaptations and significant biodiversity makes the origins of agriculture here a

complicated and interesting case. This history includes both the independent

19

domestication of indigenous plants as well as introduced domesticates from
Mesoamerica. The adoption and utilization of both indigenous and exotic domesticates
overlaps throughout ENA and the initial introduction of exotic domesticates from

Mesoamerica did not have immediate effects on the extent subsistence systems.
Overview of Eastern North America

Brief Hijstoryof Re,s_e_farch into the Origins of Agriculture in ENA

Evidence for the presence of indigenous domesticates within ENA was first
discovered in the 1930s with the identification of cultigens at sites such as Salts Cave and
Cloudsplitter and Newt Kash Rockshelters, along with other Ozark and eastern Kentucky
sites. It was through the archaeobotanical work of Melvin Gilmore and Volney Jones
that caches of seeds found at these sites were posited to have been intentionally grown
(Gardner I987; Gremillion and Sobolik 1996; Smith 1992; Smith and Cowan 1987).
That the small, starchy-seeded annuals (such as chenopod and marshelder and dating to
around 4500-34008P) found at these sites were cached, lead Gilmore to suggest that
these plants were cultivated and not simply gathered from the wild. For Jones however it
was the large seed size of those cached plants, as compared to specimens found in the
wild, which suggested that they might be cultigens. The issue of whether these seeds
indicated ENA as a primary center of domestication (i.e. an independent in situ
development of agriculture through co-evolutionary cultural and agricultural systems
processes) was, however, only briefly speculated upon by both researchers (Smith 1992;
Cowan and Watson 1992). 2

It is not until Fowler’s (1957) work that the case for ENA as an area of

independent development of agriculture ‘reopens.’ Fowler, using Anderson’s (1952)

20

Dump Heap theory, as well as the work of Gilmore and Jones, prOposed that a Late
Archaic subsistence pattern, based on a seasonal settlement cycle, set up behaviors for
domestication through the gathering of small starchy seeded annuals such as arnaranth,
marshelder, and sunflower (Fowler 1971). He argued that this pattern of utilizing small-
seeded annuals continues virtually unchanged throughout much of ENA. The exception
to this pattern is modification to this indigenous system with the introduction of maize.

The previous work done by Fowler, Gilmore, and Jones is invaluable but it isn’t
until the 19608 and 705, that the evidence for ENA as an independent center of
domestication is systematically examined. Systematic examination and methodological
innovation within the field of paleoethnobotany was undertaken through the work of
Struever (1971), Yamell (1964), and Watson (1976; I997). Binford's call for more
standardized field and laboratory methodology during the 605 and 708 also influenced the
methods employed by paleoethnobotanists (as discussed in Chapter 3). During this
period, Struever also (1971) argued for the more rigorous use of fine-grained recovery
methods, especially flotation, in order to allow for better recovery of charred
macrobotanical remains. In conjunction with the use of flotation, Struever (1971)
suggested that feature contexts, especially hearths, are the best archaeological source for
potential subsistence remains to occur. These types of archaeological features are
targeted because it is unlikely for small seeds to have been preserved without charring of
some kind.

Another important methodological recommendation by Struever ( l 971) is the
need to compare recovered seeds of the same species from different archaeological sites.

This was required in order to (potentially) clarify issues of indicator traits for

21

domesticates. At this point in time, a major methodological question was how to identify
if these small starchy seeded annuals were cultigens. For Struever this was pa1ticularly
important due to the issues surrounding larger seed size as a potential indicator of plant
domestication. At the time, there were arguments against using larger seed size as an
indicator for domestication.

One line of reasoning against the argument for larger seed size as an indicator
for domestication was the potential for unintentional selection during harvesting of larger
plants with larger seeds resulting from growing in enriched soil (Yamell 1971). These
views were still prevalent during the 1970s despite Yamell’s (1971) study of seed size in
giant ragweed. In this study, randomly sampled seeds fiom a large plant and from a
small plant of giant ragweed were collected and compared for both seed size and seed
weight. Plant size was found to have no affect on seed size or weight.

However it wasn’t until further work on the markers of domestication by other
researchers, such as Asch and Asch (1977, I978, 1985), that confirmed these plants as
domesticates. Along with the identification of larger seed size as a marker of
domestication, the identification of thinner seed coat thickness as an indicator of wild or
domesticated status was also confirmed. When compared to their wild counterparts, it
was clear that the seeds found at those Kentucky cave sites had those clear markers of
domestication. This combined body of evidence led to the hypothesis that ENA may
have been a center of independent plant domestication and husbandry (Yamell 1993).

One question raised, however, is the mechanism by which these plants were
transformed from wild to domesticated species. The most likely model is that proposed

by Edgar Anderson (1952). Anderson (1952) proposes the “dump heap” model as a

22

theory on the process of domestication within ENA. This hypothesis states that dump
heaps and disturbed areas caused by human activity are ideal habitats for weedy, invasive
plants. These invasive plants are best characterized by many of the oldest cultigens (i.e.
the Eastern Agricultural Complex (EAC): squash, marshelder, chenOpod, and sunflower)
within ENA. This hypothesis has since been expanded upon, primarily by Smith (1995,
2002) who also utilizes the work of Harlan et al. (1973), to develop his theoretical model
on the role of domestilocalites in plant domestication.

According to Smith (2002), domestilocalities have four attributes important for
a new plant habitat. These include sunlight, soil fertility, soil disturbance, and the
continual introduction of seeds. Domestilocalities are open habitats caused through
anthropogenic disturbance of the surrounding environment (Smith 2002). These
disturbed areas, in conjunction with at first unintentional coevolutionary processes, bring
about domestication and eventually active encouragement of plant species (Smith 2002).
As weedy invasives, these plants would quickly move into an area disturbed by human
activity. Hunter-gatherers moving through their seasonal round and returning back to the
same or nearby areas would have noticed these plants and perhaps started to gather their
seeds (Smith 2002). The result is that eventually intentional planting or habitat creation

would have transformed EAC plants into domesticates.

Crop Historiesiastem Agiculturg Complex

As discussed in Chapter I, Cowan and Watson (1992) define ENA as both a
primary area and secondary area for the origin of agriculture. This is because of the two
separate crop systems in ENA, both of which became economically important at different

points in time. The first crop system is an indigenous suite of cultigens characterized by

23

small starchy and oily seeded plants that were originally weedy invasive species, as
described above. These indigenous EAC cultigens include chenopod (Chenopodium
berlandierr'), marshelder/sumpweed (Iva annua), sunflower (Helianthus annuus), and
squash (Cucurbita pepo). Depending on the region of ENA, other wild grasses which in
some areas became EAC cultigens include little barley (Hordeum pusillum), erect
knotweed (Polygonum erectum), and maygrass (Phalaris caroliniana) (Crawford and
Smith 2003 Scarry 2003; Smith and Cowan 2003).

The earliest evidence for the domestication of an EAC species is squash.
Evidence for domesticated squash species dating to the Middle and Late Archaic (time
periods: 8000-60008P and 5000-20008P) is prevalent throughout the Midwest and
Northeast. Most Middle Archaic sites come from the Midwest and Southeast and are
located around major riverine systems including the Missouri, Illinois, Mississippi and
Ohio River valleys. It is not until the Late Archaic that the earliest evidence for squash in
the Northeast (New England and the Great Lakes) is found. See Table 2-1 for a list of
sites with dates. Throughout all of ENA, however, the early domestication of squash
does not have a significant economic impact upon subsistence systems and it is tenuous
to argue that food production was occurring in these areas during the Middle and Late
Archaic periods due to the utility of squash fruits for non-food purposes such as floats for
fishing lines and nets or water containers (Fritz I999; Gremillion 2003; Hart et al. 2004).

The pattern of site locations (trending south to north) seen in early squash
remains is also true for chenopod, sunflower and marshelder. By the Late Archaic (ca.
5000-30008P) and Early Woodland (ca. 3000-20008P) in the Midwest and Southeast the

earliest domesticated remains of these species are found at several sites including Koster

24

Table 2-1: Early Cucurbit finds in Eastern North America

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Midwest & . . Calibrated
Southeast S'te:;:e & Radgezrgbon Radiocarbon Reference
Sites Age (2&ng
Ph‘II‘Ifigpmg’ 4440475 8P 4869-5290 8P Smith 2000
Koster IL 7100i300 BP; 7423-8482 8?; Conard et al. 1984;
’ 63204240 8P 7265-8074 BP Asch 1994
Anderson, TN 6990i120 8P 7606-8017 8P Crites 1991
Hayes, TN 54304120 8? 5936-6438 BP Watson 1935
“Name" 5130460 8P 5726-5995 8P Cowan 1997
Napo'w“ 70004240 8P 7435-3223 8P Conard et al. 1934
Hollow, IL
Northeast
Sites (Great
Lakes &
New
England)
Sharrow, ME 56954100 8P 6295-6678 8P Asch Sidell 1999
Memorial 54054552 8P; 4952-7337 8P; Hart and Asch
Park, PA 2625445 BP 2702-2352 8P Sidell 1997
Meadgxcmfir Ca. 3000 BP Cushman 1932
Lovis and
Schultz, MI 2320440 8P 2844-3064 8P Monaghan 2°08;
Monaghan et al.
2006
ngNfguber 2530460BP 2461-2795 BP Perkl 1993
Scaccia, NY 2905435BP 2947-3164 8P Hart et al. 2007

 

 

 

and Napoleon Hollow (IL), Hayes (TN), and Cloudsplitter (KY) (Asch and Asch I985;
Crites 1993; Smith and Cowan 1987). Evidence for these plants in the Northeast does not
occur during the Late Archaic. Instead the earliest evidence for chenopod, sunflower,
and marshelder in the Northeast is during the Early and Middle Woodland periods (ca.
3000-20008P and 2000-16508P). Use of EAC plants continued throughout ENA up to

contact with Europeans. However their peak period of use occurred during the Middle

25

 

Woodland. Evidence for intensive EAC use during the Middle Woodland is most
apparent in Hopewell cultural manifestations in Illinois and Ohio. It is after this period
that maize (Zea mays) cultivation becomes widespread and use of EAC plants declines,
or as is the case in the central Midwest and Southeast, disappears altogether (Crawford

and Smith 2003).

Crop Histories: Mag;

It is with the introduction of maize that ENA becomes an area of secondary
agricultural origin. The introduction of this crop staple eventually replaces the original
mixed foraging-horticultural subsistence system present in many areas of ENA. (Cowan
and Watson 1992:207-208). However, the replacement of indigenous domesticates by
maize is not represented as an immediate wholesale adoption of maize in favor of EAC
cultigens. The adoption of maize as a staple is a gradual occurrence with a number of
contributing factors. The primary factor in the lengthy period of time between the first
evidence of maize and its use as a staple is due to its initial limitations (environmentally
Speaking) as an imported trOpical non-native domesticate that required adaptation to a
more northerly climate. Other secondary reasons include the labor investment to tend
and protect maize during the growing season, processing requirements, and a general
unfamiliarity with the reliability of this plant to local indigenous pOpulationS (Smith and
Cowan 2003). For these reasons it took time to move north within ENA and make a
significant impact on the traditional subsistence system. However, it is also for these
reasons that research concerning the adoption and use of maize within ENA has been so

robust.

26

The timing of maize agriculture in ENA has been an almost continuous area of
research. Use of increasingly refined data recovery techniques, especially microbotanical
analyses such as phytolith and starch studies on ceramic residues, to approach this
problem has meant that the date of initial maize utilization is continually being pushed
back. The most recent early date (2270i358P; Hart and Matson 2009) on maize
utilization comes from central New York with the recovery of maize phytoliths from
AMS dated ceramic residues. Compared to the earliest AMS dated maize macrobotanical
remains which fall into the range of 2050-1 8008P from sites located in Ohio and
Tennessee (Smith and Cowan 2003) demonstrates some overlap. Earlier initial use for
maize is especially pertinent when found in northern areas of ENA, such as the early date
from New York, as this indicates that the process of maize adapting to more northerly
climes is happening faster than originally expected (See Table 2-2 for summary data).

Initial findings of maize remains in archaeological contexts generally involved
large caches of maize. One such cache example comes from the Juntunen (20MKI) site
located on Bois Blane Island, MI. Excavated by the University of Michigan in the 19603,
large caches of charred maize kernels and cobs were discovered. Caches of maize
throughout ENA were primarily found prior to the development and use of flotation as a
widespread means for recovering archaeobotanical remains. These caches dated to late in
the archaeological record of ENA for a number of reasons. First, in order to have caches
of plant remains visible to the naked eye, a large quantity of seeds are required. The
quantity needed also implied that by the time these caches were made, maize was
cultivated to the extent that it was a staple crop and no longer a dietary supplement.

Overall for ENA, and especially the Northeast, the dates for maize as a staple are late

27

Table 2-2: Summary data of early maize

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. . . Calibrated
Midwest & Site Name & Radiocarbon Rad iocarbon Reference
Southeast Sites State Date .
Ase Q-sama)
Harness 1730:1285 8P; 1483-1827 8P; Smith 1992
Mound, OH 17204105 BP 1403-1872 8P
Hol din IL 2017450 8P; 1875-2115 8P; Riley et al.
gr 2077470 8P 1331-2132 BP 1994
Icehouse Chapman and
Bottom, TN 17754100 8P 1509-1925 8P Crites 1987
The Pas
Reserve, ON 1590*” BP 1366-1579 BP 533; 3‘3: 0
(north western)
Northeast
Sites (Great
Lakes & New
Eggland)
Vinette, NY 1960423 8P 1864-1953 8P “$5053; 6‘
Grand Banks, 1550i150 8P; 1222-1817 8P; Crawford and
ON 1570490 8P 1302-1628 8P Smith 2003
Fortin 2, NY 1515427 8P 1337-1422 BP “0mm“ CI
al. 2005
Wickham, NY 1433431 BP 1295-1332 8P ”Iggy"
Eidson, MI 1700i70 8P 1479-1742 8P Garland 1990
Marquette Lovis et al.
Viaduct, M1 1700-1500 8P 1226-2115 BP 1996
208A1034 1400 BP Parker 1996

 

(AD 900-1000 or later). The combination of late dates and large quantities of maize with

little intermediary evidence for maize cultivation resulted in the initial hypothesis that the

introduction of maize was both late and resulted in rapid change from a hunter-gatherer to

an agricultural mode of subsistence.

The rapid adoption of agriculture model was modified, however, with the

advent of flotation as a method of data recovery. The use of flotation as a refined

recovery technique has produced vast quantities of charred macrobotanical remains that

28

 

would not have been found otherwise. As a result, a better understanding on the
incorporation of domesticates into a hunter-gatherer subsistence system developed.
Models on the origins of agriculture within ENA no longer incorporated a rapid transition
fi‘om a hunter-gather to a maize dominated agricultural subsistence system. Instead, it
was recognized that a period of indigenous crop cultivation occurred before the advent of

intensive maize agriculture.

History of Origins of Agriculture Research within Michigan

The study of the origins of agriculture within Michigan acts as a good case
study and microcosm which mirrors the history of this problem within the larger
geographic region of ENA. A short background on the geophysical landscape of
Michigan is provided before going into the history of research on the origins of

agriculture in Michigan.

Geologic and Geographic Settigof Michigag

Michigan’s lower peninsula represents the transition between two ecotones,
Northern Hardwood-Conifer forest in the northern part of the lower peninsula, and
Beech-Sugar Maple forest to the south. The floristic tension zone present within the
lower peninsula of Michigan (Figure 2-1) is primarily a product of weather and climate
but soil characteristics also play a role in the distribution of plant communities within the
state. These soil characteristics are directly related to the glacial history of Michigan in
regards to the deposition of gravel, glacial till, alluvium, and beach sand (Dorr and

Eschman I970; Monaghan and Lovis 2005).

29

Lake level extent and fluctuation during the pre-Holocene (end of the
Wisconsin glaciation, c. 14,500 BF) and Holocene (12,000 8P to present) were directly

influenced by glacial advance and retreat as well as by-products of glaciation such as

Figure 2-1: Approximate division of ecotones in Michigan

 

' . . . N
Lake Superior » .

     
 

I a
\ I m “f ‘ .. “1.3 l
, .. - ‘ ~.-a -‘ ~.-..-~
. g, . ‘
heel-IE— -- -«'.m—Am if y I. j
. {I It 1 -IJ I, ' ‘ ,. ‘2’ -'
—— Rivers 4" 1
"
I
Lakes .
I!
.1 ’ "I '0.
f
3

Meters Above Sea Level
044550 - . ' . _4

I 501-550 ‘ , -. — i- 1 ‘

a 451-500

, "CanadianBieiie. - 731
{:1 401-450 ‘

Province ' "I"

-.‘. ‘ , r -I 't‘ f‘ _.- :5. 5...“..9.”
l | 351-400 . - 1 - i 41 1

  
 

Lake -, -. 1.4“?“ 1 1,5“? ._
1 1251-300 1 Michigan , -- a ~ '

    
 

1"“

I_,

_ I "4.; ’ , I y‘u- ‘ I .'d
. 4- ‘ - with, 4,. ~. .
- --— 1 ~ -I_ ' ._-‘._ '0' g. I
l I 201 250 if? 1 ~ Lf 44%....»441'1. ‘ 1
- D. '3.“ . I .

 

 

I" Carolinian Biotic , ._ I
G 0-211] 1 1 1 Province ' ‘ “ '

 

I ~. _,. _

 

I ‘4’
.I
e, 9
If . . ;
We. . . 1 .
‘ Q l . .. ~ . 11“.?
5' I". ~ '1 -‘ 1 I 3 ' ' I
, -.' . -1) ' . .;' ' a" I
. (4:! I“ . . . ‘ -. -. ,- ‘ '- ..
U '0 , J . - ‘ 7 f" --' ‘1 i l 1‘ .
EU ‘I .1 ‘ l- '_ . '4 ~ ‘ ‘ e ‘
- J I ~ 4 '. , . - . .
l ‘ -

 

In; . . _, (a _.' ,
- i =4. -‘ 4. * ...
Kilometers .I‘L '

 

 

 

30

isostatic rebound (Dorr and Eschman 1970; Monaghan and Lovis 2005). During the
early Holocene isostatic rebound played the biggest role in regards to lake level
fluctuation in the Michigan and Huron basins primarily due to the opening and closing of
drainage outlets (Monaghan and Lovis 2005). Before retreat of the Laurentide Ice sheet
opened the outlet at North Bay, Ontario, drainage for the Michigan and Huron basins was
to the south at the Chicago and Port Huron outlets (Monaghan and Lovis 2005). Pre-
Holocene lakes went through a number of high and low water stages during glacial
retreat.

In the Lake Michigan Basin, pro-glacial Lake Chicago went through a number
of high-low-high lake level phases. The low phase was directly due to the retreat of the
Lake Michigan Ice Lobe which opened the outlet at the Straits of Mackinac, c. 13,500 8P
(Dorr and Eschman 1970; Monaghan and Lovis 2005). This changed drainage of Lake
Chicago towards the north and east through the Port Huron outlet and the now connected
Michigan and Huron basins formed Lake Arkona (Monaghan and Lovis 2005).

Reflective of the fluctuations in Lake Chicago, pro-glacial lake levels in the
Huron and Eerie basins mimicked glacial advance and retreat but were more prone to
significant changes in water levels (Dorr and Eschman 1970). Lake Maumee (c.14,000
8P), the oldest and highest of the major lake phases in the Lake Eerie basin had 3 stable
high-low-high water phases. Initial drainage of Lake Maumee was through the Fort
Wayne outlet, however glacial retreat Opened an outlet across southeast Michigan. The
result was Lake Maumee draining to the west via the Glacial Grand River channel to
Lake Chicago (Monaghan and Lovis 2005). Lake Maumee became a part of Lake

Arkona, mentioned above, once glacial ice retreat opened up the Straits of Mackinac. As

31

glaciers advanced during the Greatlakean stade, the Straits of Mackinac outlet was cut off
and glacial lakes Chicago (Michigan basin), Whittlesey (Eerie basin), and Saginaw
(Huron basin) were formed c. 13,000 8P (Dorr and Eschman 1970; Monaghan and Lovis
2005).

During the early Holocene the low water phases and drainage outlets for Lakes
Chippewa and Stanley (c. 10,300 8?) were primarily controlled by isostatic rebound at
North Bay, ON. This outlet was reopened and allowed water in the lakes to drain north
and east (Monaghan and Lovis 2005). Once the North Bay outlet was raised higher, due
to isostatic rebound, than the Straits of Mackinac, water level in Michigan and Huron
basins increased. With the continued uplift of the North Bay outlet, it was eventually
raised above the altitude of the southern outlets and drainage back to the southern outlets,
initiated Lake Nipissing at c. 5,000 8P (Monaghan and Lovis 2005).

Subsequent changes in Holocene lake levels during the mid-Holocene and later
were in part due to isostatic rebound but not to the extent it played during the early
Holocene (Larsen 1985; Monaghan and Lovis 2005). Starting with Lake Nipissing, and
continuing through to modern lake levels, lake level fluctuation was less due to isostatic
rebound but more to shifts in climate, primarily between cooler and wetter and warmer
and drier shifts in climate (Monaghan and Lovis 2005). Cooler climates resulted in less
water evaporation while wetter conditions contributed greater precipitation to the lakes;
the opposite mechanisms operated during periods of warmer and drier climate. Warmer
temperatures raised water evaporation rates while drier climatic conditions meant less
precipitation was available to replace water lost through evaporation (Monaghan and

Lovis 2005).

32

Such shifts in lake levels during the Holocene resulted in flooding of low lying
areas such as floodplains and the Shiawassee Flats during high water stages. lncising and
erosion of river beds occurred during the low water stages. These processes resulted in
the formation of wetlands and swamps during high water stages, the deposition and
subsequent erosion of gravels and alluvium, along with burial of archaeological sites,
during various points in time (Monaghan and Lovis 2005). The result of this complex
geologic history has played a direct role in the location of plant communities throughout
lower Michigan and in the Saginaw drainage due to the variable soil types and their
characteristics which resulted from glaciation and repeated episodes of flooding and
erosion. However soil characteristics are not the only factor which affects the location of
plant communities; weather and climate also play a role.

The most important aspects of weather and climate which influence the
distribution of plant communities within Michigan are a result of rain fall, solarization,
wind, and its proximity to the Great Lakes (Andresen and Winkler 2009). The Great
Lakes have an ameliorating effect on Michigan climate in the state’s higher latitudes,
especially along the Lake Michigan shoreline, which allows for a longer growing season
due to variations in microclimate which can warm soils faster and mitigate frost
formation (Andresen and Winkler 2009). It is due to these factors that species with
southern affinities, such as Quercus velutina (Black Oak) and Caiya ovata (Shagbark
Hickory), extend as far north as Missaukee and Manistee counties in Michigan (Harmon
2006). The same is true of species with more northern affinities, such as Tsuga
canadensis (Eastern Hemlock) and Pinus strobus (Eastern White Pine), which occur as

far south as Berrien county (Eastern Hemlock) and Bay county (Eastern White Pine) in

33

Michigan. In the instance of northern affinity species, it is the presence of patches of
sandy, acidic soils in conjunction with the ameliorating effects of the Great Lakes on
summer heat which allow for the southern extent of these species.

The location of the Saginaw drainage on the border of these two macro-pattems
of plant communities allowed for a varied subsistence base. Exploitation of plant
resources in both deciduous and coniferous forested zones allowed for a more diverse
woodland resource base. Recognizing that the floristic tension zone between deciduous
and coniferous forest in the Lower Peninsula varies across the state, it also makes ease of
discussion in regards to plant geography cumbersome. In order to ease discussion, the
area north of the tension zone will be referred to as the Canadian Biotic Province (i.e.
predominantly coniferous forest) while the area south of the tension zone will be referred

to as the Carolinian Biotic Province (i.e. predominantly deciduous forest; Figure 2-1).

Research on prehistoric agriculture in Michigag

Ideas and research on the use of agriculture in Michigan are necessarily
constrained by the local environment. Michigan is an area with a northerly climate and
maize had to already be adapted to a shorter growing season in order to be successfully
grown. This shorter growing season included adaptation to rainfall, solarization,
temperature, and number of frost free days. On average, maize requires 180 frost free
growing days and is intolerant to cold, especially in its early stages of growth where cold
growing conditions can stunt growth. Similar limitations occur with precipitation; too
much water early in its growth will also stunt maize. As a result, maize not only needs
warm climate conditions, but also well drained soils for ideal growing conditions

(Krakker 1983).

34

In addition to maize plant adaptation, there are human behaviors that would
have buffered against adverse environmental effects and aided in successfiil maize
horticulture (Krakker 1983). These behaviors include cultivation techniques, such as
using ridged fields or hills in which to plant crops, to garden plot location and soil
selection (Gartner 1999; Krakker I983). Planting cultigens in ridged fields or hills has a
number of benefits. Ridges provide drainage and water storage by aerating planting
surfaces and storing moisture at the base of ridge beds (i.e. in the field valleys). The
cultivation techniques associated with raised fields also improve crop habitat by
providing weed control as well as producing a loose well-aerated and fertile planting
surface. Planting surface fertility was maintained by using sediment collected in the
ridge ditches in conjunction with the burning of crop residues (in the case of maize, this
would be stalk remnants). Use of ridged fields also helps to counteract the dangers of a
deep spring flost, which can be more damaging to plants as they are in the initial stages
of plant growth and germination. The likelihood of a deep spring frost is often more
critical in terms of crop damage than the total number of frost-free days in a geographic
area (Gartner 1999). Locating garden plots on areas of higher elevation and in
conjunction with well-drained soils also guards against any subsequent growing season
flooding which may occur and allows for warmer soil temperatures (Krakker 1983: 100).

The Carolinian biotic province is characterized by deciduous forest and
temperate climate with moderate rainfall (Albert et al. 1986; Anderson and Winkler
2006; 8ames and Wagner 1981; Cleland 1966; Dice I943; Harmon 2009). Most of ENA
falls within this floristic tension zone and is considered to be very productive in terms of

available wild food resources. Most importantly, this biotic zone falls within the 180

35

fi'ost-free growing days maize requires (at a minimum). This is in contrast to the
Canadian biotic province which is dominated by conifer forest and a less temperate
climate with a shorter growing season.

The transition between these two zones within Michigan stretches from the
southeastern edge of Saginaw Bay on the east side of the state to the mouth of the
Muskegon River on the west side of Michigan. This transition zone stretches as far north
as Charlevoix on the Lake Michigan side and to the mouth of the Ausable River on the
Lake Huron Side; the ameliorating climatic effects of Lake Michigan allow for the
transition zone between the Carolinian and Canadian biotic provinces to be farther north
on the west side of Michigan than the east side. As a result, the potential for maize
horticulture is best found in the Carolinian biotic zone and the southern portion of the
transition zone from the Carolinian to the Canadian biotic province. This expectation, in
addition to the type of evidence found, guided early interpretations on the use of maize
and maize agriculture.

The model advocated by Yamell (1964), Cleland (1966), and Fitting (I969)
mirrors the early model within ENA of a rapid transition from hunter-gatherer to reliance
on maize agriculture. In this model the transition from the Late Archaic to the Woodland
period is marked by rapid widespread change from hunting and gathering to an
agriculturally based subsistence system (Egan 1990). Evidence for this rapid transition to
maize horticulture was based on the discovery of maize storage pits located within large
village sites such as Juntunen. Other evidence cited for the rapid transition from hunting
and gathering to farming is based on the presence of open woodland species at sites such

as Schultz and Moccasin Bluff. These types of species (examples include elk and

36

woodchuck) appear in Middle and Late Woodland strata. Evidence for the exploitation
of open woodland fauna is not found during the Early Woodland periods at these sites.
As a result, Cleland (1966) interprets the presence of these open woodland species as
evidence for the clearance of forest in conjunction with agricultural activities, primarily
clearance for agricultural fields. In conjunction with the presence of Open woodland
fauna, Cleland also cites the decreased importance of hunting, based on the increased
presence of fish remains coinciding with a decrease in mammalian remains, as evidence
for a more plant based economy. It is inferred that this change means hunting was no
longer the primary focus of subsistence during the Middle and Late Woodland periods
which indicates that a shift in subsistence had occurred (Cleland 1966:210).

While this model of quick subsistence change is no longer accepted, it is the
jumping off point for the two predominant models currently in use to explain the
transition to low-level food production incorporating maize within Michigan. One model
is that advocated by Lovis et al. (2001) and Egan (1993), which focuses on ecosystem
factors while the other is put forth by O’Shea (2003) and Milner (1991) and places
primacy on social relationships. Both models debate the mechanism for gradual
intensification on domesticates and cultigens.

O’Shea (2003) suggests that there are two populations in Michigan, an inland
forager group and a coastal dwelling maize horticulturalist population. He argues that
social interaction between these two groups was important in affecting each other’s
subsistence decisions. When maize horticulture was adopted by these coastal groups, the
inland population was cut-off from resource rich coastal zones (O’Shea 2003). In

response to this minimization of resources, O’Shea (2003) hypothesizes that inland

37

forager groups responded in two ways: 1) intensified exploitation of coastal-equivalent
inland resource zones (i.e. large inland lakes and wetlands) and 2) creation of inter-
regional social ties which enabled them to acquire both coastal resources and maize
through exchange. This relationship would have been beneficial to the coastal
populations since maize horticulture can vary in yearly production and this would have
allowed them access to potentially more stable inland resources. Milner (1991) also
argues for the importance of social dynamics in the western Great Lakes, in the form of
trade alliances, marriage ties, and mortuary ritual. She argues for short-term fluidity in
local group affiliation with longer term regional ties maintained via exogamous marriage
with other sub-regions. In this way, local group control over productive fisheries was
maintained while preserving access to inland resources (Milner 1991).

In the ecosystem and landuse model, (c.f. Egan 1993; Lovis 1990; Lovis et al.
2001; Robertson 1987) the focus is on factors such as reliability and abundance of wild
resources, where those resources are located on the landscape and how this effects site
location, and the intensified use of these reliable and abundant resources. In the Great
Lakes, reliable and abundant plant resources are characterized by wetland resources and
weedy annual EAC cultigens. Intensification on these plant resources sets up pre-
adaptive behaviors conducive to expanded low-level food production which incorporates
maize. Use of and planting and tending practices of EAC cultigens allows for a ‘plug—
and-play’ type of system wherein these extent practices can be adapted for use with new
types of cultigens; in this case maize.

Egan’s (1993) work also takes into account issues of gender and the allocation

of labor within the ecosystem model. However, cultigens were not included in her

38

macrobotanical analysis of sites within the Saginaw River valley. The exclusion of these
plant resources, under the assumption that they were not a large component of the hunter-
gather diet, may have had a drastic impact on the model subsequently developed for
settlement-subsistence patterns within the Saginaw drainage. For the time period
analyzed by Egan, the populations within the Saginaw drainage are best characterized as
mobile low-level food producers. Currently, there is ample evidence that domesticated
squash was present as early as the Late Archaic and domesticated chenopod by the
Middle Woodland (Lovis et al. 2001; Monaghan et al. 2006). The presence of these
domesticates suggests they were being incorporated into the diet of the local population
on some level. It is especially important that these were domesticates and not wild
species of these plants as this indicates they were being cared for in some way. As such,
we can characterize the population within the Saginaw drainage area as mobile low-level
food producers. The incorporation of domesticated plants into what has traditionally
been deemed a hunter-gather system also requires an explanation of which segment of the
population were tending these plants and how this division of labor was divided when
care for cultigens coincided with traditional hunting and gathering activities.

A combination of the two models presented above exists in the work of Howey
(2006, 2007). Howey combines landscape factors such as waterways, slope, and
vegetation cover in a GIS in order to model movement of different populations within
Michigan to the Missaukee Earthworks (20MAI 1-12). Missaukee Earthworks is a late
prehistoric ceremonial earthwork, in lower northern Michigan. Howey (2006, 2007)
modeled pathways to Missaukee from various locations in Michigan including the

Saginaw drainage, southeast Michigan, and southwest Michigan.

39

Pathways to Missaukee Earthworks from these various regions were modeled
since this site was known as a ceremonial integration site. The interpretation of
Missaukee as an area where dispersed populations convened is also supported through the
presence of lithics and ceramics from disparate areas of Michigan. Lithic material
includes Bayport chert (which outcrops in the Saginaw Basin) and Norwood chert (which
outcrops in Antrim county on Lake Michigan) while ceramic styles such as Juntunen and
Traverse Wares (northeastern Michigan), as well as what Howey terms Mixture Late
Traits (i.e. a mix of decorative styles which are not typical of one ware) are present.
Howey’s interpretation follows the O'Shea and Milner models by positing that the
different regional pOpulations congregating at Missaukee represents coastal farmer and
interior hunter-gatherer populations.

The research undertaken for this dissertation attempts to fill in gaps present in
the ecosystem and landuse model for the Saginaw drainage. Data collected via phytolith
and starch analyses of ceramic residues provides another line of evidence for various
subsistence activities. Microbotanical data often provides evidence for plant use not seen
in macrobotanical data due to different taphonomic processes undergone by
microbotanical data. Microbotanical data directly derived from artifact surfaces (either
ceramic or lithic artifacts) also provides direct evidence of which plants were utilized
together. The utilization of ceramic residues in this study specifically provides data on
plant mixes cooked together. These combinations not only provide dietary information
but also potential information on seasonality through the use of plants which are known
to be stored or those which are known to only be utilized during their season of

availability.

40

 

 

Additionally, a social component may be relevant to the interpretation of
microbotanical data if any exotic or non-local plants are identified. If such plants are
identified, their presence is most likely due to either down-the-line, or perhaps direct
exchange between populations. It is arguable, however, if this potential social interaction
represents coastal farmers and inland hunter-gatherers. For the purposes of this
dissertation, it is assumed that the various populations within Michigan do not represent a
coastal-farmer, interior hunter-gatherer dichotomy but rather distantly related groups

practicing mixed subsistence economies to varying degrees.

41

CHAPTER 3
Evidence for Plant Use
Introduction

As archaeological methods and techniques have become more refined and
innovative, the lines of evidence, both direct and indirect, for plant use have also become
more varied and refined. Direct evidence for plant use consists of the physical remains of
the plants themselves and includes macrobotanicals, such as seeds, cobs, and rinds and
microbotanicals such as pollen, starch, and phytoliths. While not all of these remains are
visible to the naked eye, they are all some form of physical remain which can be
extracted from various archaeological and environmental contexts. Included in direct
evidence is accelerated mass spectrometry (AMS) dating, which has enabled the
radiocarbon dating of small amounts of carbonized material (i.e. individual
macrobotanical remains or carbon located within phytoliths). The development of AMS
radiocarbon dating is what is most pertinent to the issue of diet since it has allowed for
refining the timing of plant use.

Indirect evidence for plant use consists of evidence which would indicate the
use of plants but does not come from the actual plant. Indirect evidence is most often
derived from the refinement of scientific techniques which have come to be applied to
archaeological data. The most important indirect dietary evidence is through stable
isotope analysis. Stable isotope studies, especially when done on bone, provide an
excellent indication for overall long-term diet composition. However, this is not always
the case as various artifact classes, such as grinding stones, can indicate plant use without

the physical remains of plants present.

42

When used in conjunction with each other, indirect and direct evidence for plant
use act as complementary lines of evidence. This enables archaeologists to better
interpret human behavior from the archaeological record and provide a more complete
picture of activities taking place, which segments of the population were likely doing

these activities, and the interaction between various groups on the landscape.
Direct Evidence

Macrobotanicals

To archaeologists studying the origins of agriculture in the early years of the
discipline, it was clear domesticates had to have developed at some point in the past since
modern industrial society is agriculturally based. As a result, macrobotanicals were the
earliest form of evidence archaeologists had for plant use. Macrobotanicals are any plant
part visible macroscOpically (i.e. to the naked eye). These include seeds and seed cases,
stems, leaves, roots, and fibers. However, methods of macrobotanical recovery have
always been an issue.

In the case of macrobotanicals, evidence for paleodiet and subsistence is
entirely dependent on the recovery of preserved floral remains. Initially, these remains
were recovered during excavations only if an excavator encountered a large cache of
remains or was sufficiently sharp-eyed enough to see various plant remains during
excavation. Additionally, early screening techniques also posed problems to
macrobotanical recovery. In early archaeological excavations, standardized screening
techniques were varied. Excavations of the same site year to year could utilize I‘/2” or ‘A”
mesh screens, or in the case of very early excavations, none at all. This resulted in many

potential remains, both artifact and macrobotanical, going uncaptured. With the

43

advocation for the use of (relatively) standardized excavation techniques by Binford
(1965) and others in the 1960s, the switch to '/4” mesh as the standard for screening aided
in some extent to counteract this recovery problem. However, it was not until the
development of flotation that macrobotanical recovery became commonplace, regular,
and standardized.

The ‘flotation revolution’ (Watson 1976), during the 1960s, led directly to the
systematic recovery of macrobotanicals and allowed for quantitative analysis of plant
remains (Pearsall 2000). Flotation started as a modification of a lab technique used by
botanists to clean botanical samples still mixed with soil. It was modified into a field
technique primarily by Struever for the express purpose of recovering small botanical and
faunal remains (Pearsall 2000:20).

The overall principle behind flotation is that botanical remains will float and
rise to the surface of water used in the flotation system while the soil is agitated through
the fine mesh screen at the bottom of the flotation bucket. Botanical remains can then be
collected, depending on the system in use, either by hand with a fine mesh screen or
floated off into cheese cloth or other fine mesh screen. Assuming that standardized
sampling for flotation samples was employed during field excavations, flotation recovery
of botanicals allows then for systematic analysis of the recovered remains. Adding an
exotic marker to each flotation sample, such as a standard number of charred poppy seeds
in the case of New World excavations, also enables statistical analysis of recovery rates
and acts as a check on the efliciency of flotation procedures (Pearsall 2000:93). The

incorporation of flotation as an archaeological sampling method has caused a drastic

44

increase in the number and types of macrobotanicals recovered as well as an increase in
the variety of plants recovered.

The number and types of plants recovered also increases if flotation soil
sampling during excavation is employed in all excavation contexts and is not simply
limited to feature contexts (Pearsall 2000:66). Pearsall (2000:66) refers to this strategy as
‘blanket sampling’. While sampling hearths and ashy deposits may seem the logical
choice for the recovery of charred macrobotanical remains, this does not always result in
macrobotanical recovery. Charred plant remains can occur through both deliberate and
unintentional processes and the use of blanket sampling allows archaeologists to
potentially capture both types of processes.

Charred remains can also be spread around and end up in secondary contexts
through the cleaning behaviors of prehistoric peoples (Pearsall 2000:66). Examples of
these secondary contexts include garbage middens and house floors. Sampling strategies
need to account for these potential taphonomic processes in order to best obtain a
representative sample. By utilizing blanket sampling, the ability to analyze and discuss
paleodiet dramatically increases since the paleoethnobotanist now has maximum
flexibility for interpretation since all contexts were sampled. This provides a means to
evaluate various site contexts and site functions (Pearsall 2000:67).

As the discussion above should indicate, the recovery of macrobotanicals is not
solely reliant on archaeological excavation techniques. Preservation of macrobotanicals
is affected by a number of variables. The variability of macrobotanical preservation can
be due to processing procedures (both in terms of initial food preparation prehistorically

and, as already discussed, archaeological recovery methods), prehistoric storage

45

techniques, soil acidity, and site formation and post-depositional processes (Pearsall
2000). Perhaps most importantly, the issues of prehistoric food preparation and storage
techniques are central to the recovery of macrobotanicals by archaeologists.

Prehistoric food preparation techniques which result in the potential for
macrobotanical preservation ubiquitously involve fire which to some extent chars the
plant material (Pearsall 2000:247). Many historic and ethnographic accounts (c.f. The
Jesuit Relations; Harrington 1908; Wilson 1987; others) describe the parching and drying
of seeds in a shallow basket or clay bowl/pot over coals. The movement of seeds is
required so as to prevent burning; this can be accomplished via continual movement of
the patching basket or clay pot or by moving seeds with a stirring paddle or stick.
Sievers and Wadley (2008) also describe experimental work wherein seeds are parched
through burial 5-lOcm directly underneath a fire. While their experimental archaeology
project was undertaken to explain parched seeds found at Middle Stone Age sites in
South Africa, it could be a method potentially used elsewhere. Burial of foods in a
container or wrapped in leaves is a common cooking technique worldwide and it would
not be unusual to assume that the parching and drying of seeds could be done in this
manner. The use of indirect heat in this way would allow for an individual to participate
or carry out other activities; similar arguments have been made for the heat treatment of
lithic material (Raviele 2007). Other drying techniques, such as sun and air drying of
parboiled seeds or mature seed pods/cobs would not result in preservation via fire. These
macrobotanical remains could however, be preserved in cache/storage pits.

The storage technology utilized prehistorically is also a factor in the recovery of

macrobotanicals. If subsistence items (plant and animal) were available year round, long-

46

term storage would be a relatively unnecessary technology. However, in an area where
drastic seasonal change occurs, as happens presently and prehistorically in the study area
for this project, long term storage of various plant and animal foods is required to survive
during the winter months. Storage pits encountered during excavation which are sampled
for flotation often yield seed remains from the plants that had been stored in these pits or
from the various grasses used to line storage pits. In these contexts evidence for the use
of subsistence and non-subsistence flora is gained.

While storage pits and hearths can represent the most likely circumstances
under which archaeologists can encounter plant remains, they do not account for all
situations in which plants were utilized by people. Not all plant use is economic in nature
nor are all economically important plants processed for storage. Fresh greens and benies
are examples of foods which are more likely to be eaten immediately instead of processed
for long-term storage. As a result, macrobotanical data produce variable interpretations
of prehistoric human subsistence-settlement rounds since clearly not all economically
important plants will be preserved. It is for this reason that multiple lines of evidence are

required when reconstructing subsistence systems.

Pollen

 

Pollen studies, while predominantly focused on reconstruction of past
vegetation and climate, have also been utilized in the reconstruction of paleodiet and
plant, specifically when analyzing coprolites and soil sarnples taken within
archaeological sites and anthropogenic contexts (Pearsall 2000:249-250). In addition to
taking soil core samples from nearby lakes or swamps for pollen analysis, soil samples

from archaeological sites can also be analyzed.

47

The best soil sampling technique for archaeological contexts is through the use
of contiguous sampling. This type of sampling removes a vertical block of soil so that
each soil horizon remains in contact with each other (Dimbleby 1985). Contiguous
sampling is recommended over spot sampling, which takes samples at intervals
throughout a profile, due to the likelihood of pollen accumulation of various ages within
these soil samples (Dimbleby 1985:20-21). Spot sampling often results in pollen
sequences which do not make sense in the context of known paleoenvironmental data.
While pollen rain from the surrounding local environment will be present in these
archaeological soil samples, pollen from subsistence plants utilized at the site is also
likely to be present. This especially true if plants are gathered with their flowers attached
and dried as in its whole form or if flowers are considered edible, as in the case of squash
blossoms.

Pollen data does, however, present other problematic issues. Pollen cores are
most often derived fi'om nearby lakes and swamps rather than directly from soil samples
at a given site (Bryant and Hall 1993; Bryant and Holloway 1983; Dimbleby 1985;
Pearsall 2000). Thus, pollen data largely provides information on the regional
environment. Pollen data derived from nearby lakes and swamps are also less likely to
preserve pollen from important food plants. These plants may not be represented due to
the distance of the lake or swamp fiom the archaeological site (Bryant and Hall 1993;
Bryant and Holloway 1983; Dimbleby 1985;Pearsa112000). Furthermore, if such pollen
grains are preserved, their importance may be underrepresented due to more abundant

Pollen producers in the local environment (Pearsall 2000).

48

Related to where pollen cores are most likely to be taken from (lakes and
wetlands), is pollen preservation. Pollen requires a continuously wet environment or an
environment arid enough--as in desert caves-that it does not degrade. Fluctuations in
temperature or moisture levels (cycles of wet and dry soils) will cause pollen grains to
degrade, sometimes to the point that they are unidentifiable. This may mean that pollen
grains from soil columns taken at archaeological sites may not be well preserved. The
case is also true of seasonal wetlands located adjacent to archaeological sites. From
personal experience taking a pollen core from a seasonal wetland adjacent to 20AN54
located in Antrim County, M1, the pollen grains recovered from this context were

unidentifiable from such severe fluctuations in moisture.

§_tar_§ji_

Often paired with phytolith analysis (discussed below) is the study of starches.
Starches are produced through photosynthesis as an energy store for the plant. All plants
produce some form of starch, of which there are two varieties. The first type is transitory
starch found in the leaves and stems and this is produced daily (Cortella and Pochettino
1994; Gott et al. 2006; Messner 2008). These starches generally do not have diagnostic
features and are often used the day they are created. As a result they are not useful in
archaeological starch grain analysis (Gott et al. 2006; Haslam 2004; Messner 2008).

The second type of starch produced by plants is reserve starch used as long term
storage. These starches are primarily found in fruits, seeds, rhizomes, corms, bulbs, and
tubers (Gott et al. 2006; Holst et al. 2007; Messner 2008; Messner and Dickau 2005;
Pearsall et al. 2004; Perry 2001, 2002a, 2002b; Perry et al. 2007; Pipemo and Holst 1998'

9

Pipemo and Holst 2004; Pipemo et al 2000; Zarrillo and Kooyman 2006). It is these

49

reserve starches that have diagnostic features which are useful in archaeological starch
grain analysis. Perhaps most importantly, it is reserve starch that is found in those plant
parts which are impOItant prehistorically, either for medicinal or economic purposes. It is
reserve starch that is likely to be preserved in residues found on stone tools or ceramic
containers (Gott et al. 2006; Messner 2008).

Starch grains are composed of a semi-crystalline structure, containing both hard
and soft layers. Grains are formed from glucose in two different forms, amylose and
amylopectin. It is these molecules which form the alternating hard and soft layers of the
starch granule. The ratio of amylose to amylopectin also aids in the resiliency of starch
to environmental stressors and has an effect on physical properties such as gelatinization
and reactions to staining (Gott et al. 2006). This ratio of amylose to amylopectin within a
starch granule is controlled by both genetic and environmental factors and it is this semi-
crystalline structure which has been demonstrated to preserve for long periods of time.
Preservation is due to a number of factors including insolubility in cold water and under
certain conditions, resistance to enzymic hydrolysis (Gott et al. 2006; Messner 2008).
While not completely understood, it is these starch characteristics which allow for the
recovery of starch granules from archaeological contexts.

Examples of archaeological starch studies include the analysis of residues
adhering to stone tools and ceramic sherds, recovery of starch from dental calculus, and
the retrieval of starch from archaeological feature context soils (F ullagar 2006; Messner
2008; Samuel 2006; Torrence 2006). Importantly, organic residues adhering to ceramic
sherd interiors can be processed for both phytoliths and starches. These samples can be

split so there is separate processing for each type of microbotanical, or they can be

50

processed together. In the case of processing for phytoliths and starch at the same time,
some starch attrition may occur but this is likely to be negligible (Boyd personal
communication 2008; Cummings 2006). In both cases, the processing for both starch and
phytoliths for dual identification provides multiple lines of evidence in a relatively easy
manner.

Diagnostic starches are identified via a number of starch grain features (see
Figure 3-1 for features described below). These include the location of the hilum, the
starch grain size and shape, fissures (if present and the type of fissure), lamellae (growth
layers), and the type of birefringence (if and what type of extinction cross is present when
the grain is viewed through cross polarized light). Included within the size and shape
category for identifying starch, is the specific granule type: simple, compound, and semi-
compound (Gott et al 2006:40-42).

The specific starch granule type is determined by how grains are formed in the
amyloplast. Compound granules are composed of several parts deemed subgranules or
granula; these form simultaneously during starch production. When ground, these
granula will separate. Semi-compound granules start as compound forms but then each
granula becomes fused together with a surrounding layer of amorphous starch. The result
is that a semi-compound grain has a single exterior surface but multiple hila (Gott et al
2006:40-42). Simple granules are those that form independently of one another and are
found as single starch grains.

Along with the granule type, starch granule shape can be part of the suite. of
diagnostic features. Diagnostic grains are found in a variety of shapes including the basic

round spherical shape but other common shapes include discs, ovals, elongated, rounded,

Sl

Figure 3-1: Major starch grain features (recreated after Gott et al. 2006)

Fissures
‘7
"a -—-Lamella
Hilum
(center)

Centric Hilum

J 649

Eccentric Hilum

 

 

Extinction Cross

kidney, and polyhedral. Grain size is also influenced by the amount of water in the grain,
age of the grain, and where in the plant it is formed. This last factor can also affect the
Shape of the starch granule (Gott et al 2006). As a result, plants may have multiple
associated starch types, of which only some are diagnostic. Prehistorically, if multiple
plant parts were economically usefirl, the variation in starch grains between them must be
examined. Therefore, it is critical that researchers have access to comparative collections
to aid in starch grain identification.

As stated previously, it is not well understood why some starch grains are
resistant to processes which will cause other grains to degrade (even those grains from

within the same plant species). However, this does mean that starch grains are left

52

behind in archaeological contexts. In order to understand how starch is incorporated into
the archaeological record, an understanding of not only which plants but which plant
parts were utilized prehistorically is needed. In conjunction with this, how they were
used and what types of processing (drying, grinding/milling, roasting, leeching, etc.) they
may have been subjected to is also required.

Experiments with starch taphonomy have been performed by a number of
researchers (cf. Babot 2006; Gott 2006; Lu 2006; Williamson 2006; Chapter 4 of this
dissertation). These experiments confirm that starch grains can exhibit damage due to
processing techniques (cracking, fractures, cavity-like depressions, clumping grains).
This damage can aid archaeologists in identifying which human behaviors were most
likely associated with plant use. Identification of starch on stone tool edges also provides
more detailed and often new information on how those tools were used while also
identifying which plants were processed. The analysis of starch fi'om tool edges may also
aid in confirming use-wear identification.

One other taphonomic issue in regards to starch studies is the potential for
contamination from the surrounding soil matrix or contact with other artifacts. Bonny
Williamson (2006) conducted a study wherein She buried a number of stone tools in
flower pots (at varying levels and left outside for varying lengths of time) to determine
under which conditions contamination would be most likely, if at all, to occur.
Williamson found that starch from the surrounding soil matrix was highly unlikely to
adhere to artifact surfaces and thus act as a source of contamination. However, when the
stone tools were buried with whole vegetables nearby, a significant number of starch

grains were found to have transferred from the vegetable material to the artifact surface.

53

As a result, the context of artifacts sampled for starch analysis should be taken into
account and “. . .a comparison with use-wear would be required to distinguish use from

contact,” (Williamson 2006:90).

Phytoliths

The use of phytolith analysis in archaeology as a line of evidence for the
identification of botanical remains has become increasingly common in the past 30 years.
Phytoliths are microscopic deposits of silica that form in and between the cells of plants
and can be found to varying degrees in each plant structure (Mulholland and Prior 1993;
Pearsall 2000; Pipemo 2006). Phytoliths have been known as important plant
constituents since the mid-nineteenth century but were not widely used by American
archaeobotanists as a fossil economic indicator until the late 19703 (Pipemo 2006). The
initial utilization of phytolith studies occurred in Meso- and South America where very
poor macrobotanical preservation required the exploration and use of alternative methods
for identifying domesticates and food plants (Pipemo 2006). As of late, it is increasingly
employed by North American archaeologists as an alternate means of identifying the
early presence of domesticates (c.f. Boyd et al. 2008; Hart et al. 2003; Thompson et al.
2004).

Phytolith analysis offers direct evidence of plant use which may not exist in
macrobotanical, palynological, or starch data (Hart et al. 2003; Pipemo 2006; Thompson
et al. 2004). The sources from which archaeological phytolith samples are derived
includes either ceramic or lithic residues or from soil samples taken within and around a
site. The result of sampling artifact residues or soil from directly underneath an artifact is

that phytolith analysis allows for the identification of plant remains directly associated

54

with an artifact(s). Direct artifact association enables paleoethnobotanists to identify
plant processing (from stone tools), suites of plants that may have been cooked and eaten
together (ceramic residues), and other human behaviors which may have only previously
been hypothesized about or inferred based on ethnographic analogy.

Archaeological phytolith analysis has primarily been performed using soil
samples taken within sites. One central assumption in the analysis of soil samples taken
within archaeological sites is that the phytoliths present are primarily the result of direct
human disposal of plants, resulting from a “decay-in-place mechanism” (Pipemo
2006:81). Soil sampling methods for within site analysis occurs in two ways. One is
through a vertical soil column removed from a finished, profiled wall while the other is a
horizontal sampling of deposits during excavation. Vertical soil columns are ideal when
a known chronology is in existence and the site represents multiple discrete occupations.

Horizontal sampling, however, is better suited for the study of various
contemporaneous areas within a site. This is particularly useful in the understanding of
different contemporaneous functional areas across a site and can add important data not
represented through macrobotanical blanket sampling, soil sample derived starch grain
data, or pollen analysis. However, even when horizontal sampling is being used, vertical
soil columns should still be employed to act as a stratigraphic and depositional control
(Pearsall 2006:82-83).

Phytolith analysis can also be used for a more direct line of data by analyzing
residues from lithic and ceramic contexts as well as from human remains such as teeth.
Phytoliths directly derived from artifacts can indicate plant processing functions and/or

techniques as well as a more nuanced study of paleodiet since plant food-specific

55

phytoliths are recovered together in a controlled context. One such context is ceramic
residues which result from cooking; phytolith analysis allows for determine which plants
may have been cooked and eaten together (Pipemo 2006:83). If these artifacts are
recovered in the field, control samples from the surrounding soil matrix should be taken
in order to differentiate background “noise” from specific functional data. Otherwise
careful examination of field notes to help determine artifact context can be used when
analyzing museum-based collections.

An aspect of phytolith analysis from artifacts is the utilization of museum
collections which may not have been reexamined since their initial recovery. While these
artifacts will not have the corresponding control samples taken from the surrounding soil
matrix, they can and do offer an important source of information not in use at the time of
excavation. As has been demonstrated a number of times (c.f. Hart et al. 2003; Lovis
1990; Thompson et al. 2004), the reexamination of already excavated material in light of
new analytical techniques offers the chance for new data from finite archaeological
resources.

Another plus to phytolith analysis is that phytolith preservation is not subject to
the same forces of decay as other botanical markers. Macrobotanicals and pollen both
require specific environmental conditions in order to preserve. Pollen requires a stable
wet or very dry environment, with minimal fluctuations between wet and/or dry
conditions in order for good preservation to occur (Bryant and Hall 1993; Pearsall 2000).
Generally good macrobotanical preservation also requires relatively stable environmental
conditions. Such requirements are not necessary in order for phytoliths to preserve; they

do not degrade as a result of climatic fluctuations.

56

Despite the number of drawbacks listed above for other lines of direct evidence
of plant use, it is not suggested that phytoliths should act as a stand-alone source of data
for paleodiet. One drawback to phytolith analysis is that the food plants in question must
be phytolith producers. Systematic studies of various plant families have identified that
not all plant families are phytolith producers (Pipemo 2006). Some subsistence plants,
such as amaranths and chenopods, fall into these families and so would not be identified
during analysis. In addition to this familial limitation, the plants which are phytolith
producers must have their phytolith producing parts (i.e. leaves, stem, roots, and
inflorescence) utilized in some form. Even then the phytoliths present can vary due to
inconsistent phytolith production between different plant parts; for example, a root will
produce fewer phytoliths than a leaf (Mulholland I993). As with all direct evidence for

plant use, taphonomy plays a role in the recovery of these indicators.

Indirect Evidence

 

In conjunction with the direct lines of evidence for paleodiet and subsistence
practices, indirect lines of evidence are also useful. Two of the most important in regards
to subsistence and paleodiet are stable isotope analysis and AMS radiocarbon dating.
This section provides an overview of each of these techniques, specifically focusing on

aSpects directly related to subsistence and paleodiet.

Stable Isotopes

In addition to direct botanical evidence, one indirect line of evidence for plant
use can come from stable isotope studies. These studies have been performed on human

remains (bone collagen) or residues present on either ceramic or lithic artifacts. Among

57

the various isotopes stable isotope studies are able to identify, both carbon and nitrogen
isotopes (”C and LN) are two of the most important isotopes in regard to dietary studies
and the identification of food sources (terrestrial versus aquatic/marine).

Analysis of carbon isotopes are used in examining the plant component of a
prehistoric diet. Carbon isotopes can provide information on the use of C4 versus C3
plants. The distinction between C3 and C4 plants has to do with the process of
photosynthesis used by a plant and how efficiently C02 is used in the photosynthetic
pathway. Plants characteristic of the C4 pathway tend to be tr0pical grasses, an example
of which is maize. The C3 pathway is what characterizes the majority of other plants.
There is, however, an exception to the C3 and C4 photosynthetic pathways. CAM
(Crassulacean Acid Metabolism) plants utilize both C3 and C4 photosynthetic pathways.
They will switch depending on their environmental conditions; if the weather is cooler, a
C3 pathway is used while if the weather is hotter the plant will switch to the C4 pathway.
CAM plants are most often succulents such as agave (Barbour et al. 1999). Due to the
differences in C02 use, the absorption of 13C will vary between C3 and C4 plants. C3
plants generally have a BBC index in the range of -22%o to -35%o while C4 plants have a
range of -9960 to 40%., for 5% (Barbour et al. 1999:422; O’Leary 1988). It is the
difference between the 8'3 C of C3 and C; which allows for the identification of these
plants based on carbon isotopes.

In addition to the use of carbon isotopes for plant identification, the ratio of
nitrogen isotopes (”N/MN) can be used to distinguish marine versus terrestrial based
animals (DeNiro 1987; Hastorf and DeNiro I985; Heron and Evershed 1993; van der

Merwe 1982). It is possible to distinguish marine versus terrestrial animals based on

58

nitrogen isotopes due to the increased ISN uptake of marine plants. This difference
translates up the food chain so that marine animals have a higher 'SN/MN ratio than
terrestrial animals (Schoeninger et al. 1983:1381). The 15N range for terrestrial animals
is +1 .9%o to +10% 8'5N while the range for marine animals is +9.4%o to +29%» SUN. It
is presumed that this higher l5N ratio will also be present in the bone collagen of humans
who subsist on a diet heavy in marine resources (Schoeninger et al. 1983:1381).

Isotope studies on human bone collagen and/or carbonate provide information
on both human consumption behaviors (i.e. dietary staples such as animal proteins and
grains) and location/region of habitation (Buikstra 1992; Hart 1999; Schoeninger and
Moore 1992; van der Merwe and Vogel 1978). These studies are useful in tracking
dietary change (by time period, between contemporary populations and within
populations) and population movement. One drawback to isotope studies on bone
collagen is the sample size. Most studies use a few individuals from a few sites to
extrapolate to an entire region (Hart 1999). As Rose (2008) demonstrates in her isotope
study of individuals from Middle Woodland through Mississippian period sites in west-
central Illinois, her results from contemporary populations demonstrate differential maize
use ranging from heavy reliance to non—existent. While Rose's study demonstrates the
need for a wider range of population sampling, isotope studies of human bone remain
promising and extremely important. But, how do these isotope studies compare to those
of ceramic or lithic residues?

Carbon and nitrogen stable isotope studies from residues can present a different
picture than that seen in bone collagen studies. Ceramic residue isotope studies reflect

the mixture of items which were cooked in the pot as well as probable accretion of

59

residue over time; it is silly to assume a pot would be used once and then discarded——
barring breakage or some other form of damage. These residues would then represent a
combination of plant and animal species. How are the contributions of C4 plants (if
present), C3 plants and their consumers, or aquatic species separated out from this mix?

Reber and Evershed (2004) discuss the combination of both stable isotope
analysis with lipid analysis of absorbed ceramic residues to differentiate between the
various components within a residue. The technique uses a combination of gas
chromatography -— combustion -— isotope ratio mass spectrometry (GC—C—IRMS) which
separates lipids while at the same time combusting the various compounds within the
residue to produce a C02 gas. This gas is then subjected to stable isotope analysis to
differentiate between C3 and C4 carbon isotopes (Reber and Evershed 200424).

For non-absorbed ceramic residues, i.e. those adhering to the surface of a sherd,
Morton and Schwarcz (2004) used a combination of nitrogen and carbon isotopes to
differentiate between potential residue contributors. Their findings from tests of a
number of northern and southern Ontario Woodland period sites indicate minimal to zero
amounts of maize contributing to the residue. This data does not corroborate the findings
of earlier bone collagen studies from Ontario sites which indicated maize as a significant
portion of the diet starting around AD 600 (c.f. Schwarcz et al. 1985). Morton and
Schwarcz (2004) offer the explanation that different cooking techniques were used
prehistorically and maize was only a minor component of those foods cooked in ceramic
pOtS.

As a result, standard views on isotope studies using ceramic residues set the

carbon isotope threshold for the presence of C4 plants at -22%o for 6‘3C and anything

60

below this represents C3 (Hart et al. 2007, 2009). However recent work by Hart, Lovis,
Urquhart, and Feranac (2007, 2009) challenge these thresholds. Systematic experiments
performed at twenty percent increments of C3 to C4 (in this case maize) contributors
indicate that C3 plants and their animal consumers often mask the present of C4 plants.
There is not a perfect linear correlation of percentage of C3 to C4 as previously assumed.
Hart et al.’s 2009 experiments further indicate that the use of carbon stable isotope
analysis to identify the presence of maize, as well as the percentage of maize, in a
ceramic residue is untenable.

From this discussion we see that the use of stable isotope analysis as a means
for reconstructing paleodiet must be used judiciously. Bone collagen isotope studies
reflect accurate representations of diet whereas residue isotOpe studies are more
problematic. Lipid analysis of absorbed residues can provide useful information;
however it is unclear if the use of GC—C—IRMS, as described by Reber and Evershed
(2004), is also subject to C3 masking of C4 plants. This is an area in which further testing

is required.

Rad'ocarbon Dating

The advent of radiocarbon dating in the 19505 caused a revolution in regards to
dating archaeological sites. Besides having relative dates through the use of seriation and
stratigraphy, absolute dates were now available. The availability of absolute dates now
provided more time depth in which culture change, especially within North America, had
occurred. -

Refinement of radiocarbon dating through Accelerated Mass Spectrometry

(AMS) has provided further time depth for archaeology. Since the carbon isotopes

61

themselves are directly counted, instead of indirectly with a Geiger counter, a smaller
amount of material is needed for radiocarbon dating (Thomas 1998). The direct result of
only needing a small amount of dateable material is that artifacts and faunal and floral
remains can be directly dated themselves; associated carbon from nearby hearths need no
longer be used. The result of this direct dating of artifacts (example: residues adhering to
ceramic sherds) and faunal and floral remains (example: a kernel of carbonized maize)
once again provided time depth to North American archaeological chronologies. Maize
was found to be older than previously assumed based on associated dates (c.f. Chapman
and Crites 1987; Crawford et a1. 1997; Crawford and Smith 2003; Ford 1987; Riley et al.
1994) as were/are some ceramic typologies (for Saginaw drainage specific examples of.
Lovis 1990; Sommer 2003a; Chapter 5 this volume).

One drawback with radiocarbon dating is the need for calibration due to the lack
of a one-to-one correlation of radiocarbon years to calendar years. The need for this
calibration is due to varying levels of atmospheric carbon throughout the earth’s history
(Reimer et al. 2004). Calibration curves are determined through tree rings
(dendrochronology) up until 12,400 years cal BP and after this period, through marine
records (Reimer et al. 2004). The marine records are converted to atmospheric
equivalents and are also subject to site-specific marine reservoir corrections. These
records extend the terrestrial record up to 26,000 years cal BP (Reimer et a1. 2004). All
radiocarbon dates must be calibrated to ensure accuracy.

The combination of the various direct lines of evidence described above, used in
conjunction with AMS dating, can yield data which help present a more robust

understanding of paleodiet and the changes that occurred over time and with the

62

introduction of domesticates. As presented later in this work, AMS dating of
archaeological ceramic residues, used in conjunction with phytolith and starch analyses,
provides new data on the timing of the use of maize, aquatic tubers and grasses.
Previously known macrobotanical evidence and associated standard radiocarbon dates

corroborate much of this data while some of it is challenged by the new findings.

63

CHAPTER 4

Experimental Work

Introduction

The experimental component of this research was initially undertaken to test if
there is a noticeable difference in maize (Zea mays) phytolith production during the stage
of plant development (i.e. younger versus more mature plants) and if so, was this also
quantifiable based on phytoliths present in ceramic residues? In other words, would it be
possible to determine at which stage of growth (green/ripe or young/mature) maize was

utilized based on phytolith size and density found in ceramic residues?

Principles of Phytolith Production

There are a number of factors which control and account for the production of
phytoliths in plants. Such factors include climate of growth environment, soil
composition, water content of the soil, plant age, and plant taxonomic affinity. Soluble
silica present in groundwater is absorbed by plants and carried into plant tissue. Some of
this silica is eventually deposited as a solid in cell walls, cell interiors, and intercellular
spaces. The mechanism for this deposition is not entirely understood but it does seem to
be, at least in part, both genetically and metabolically controlled. Evidence for a degree
of genetic control in phytolith deposition is demonstrated by evidence that indicates the
patterns of silica accumulation and deposition are similar in plant species of closely
related taxa (Pipemo 2006:5-6). Various environmental factors and growing conditions
do not appear to effect phytolith formation within a taxon. Phytolith formation has been
found to be consistent despite widely varying environmental factors (Pearsall 2000:359).

At the same time, non—producers taxa of phytoliths, such as the Amaranthacea and

64

Chenopodiaceae, do not produce phytoliths regardless of which areas in the world they
grow.

Metabolic control over silica deposition and uptake occurs with active transport
of monosilicic acid by the plant. In this process, the plant is expending energy to absorb
silica, which seems to indicate a designated function for said silica (Pipemo 20069).
This soluble silica may then be converted into solid deposits of silica, or silicon dioxide.
The process by which soluble silica is converted into solid deposits is also poorly
understood but it may be related to transpiration. Active transpiration may result in plant
cells becoming supersaturated with silica and as a result, silica accumulation and
precipitation occurs within the plant cells (Pipemo 2006:10). This is not to suggest that
phytoliths only occur in cells associated with transpiration. High levels of phytoliths can
be found in other plant cells including idioblasts (or short cells), pericarp, mesophyll, and
epidermis cells (Pearsall 2000:363; Pipemo 2006:10). Deposition of silica in non-
transpiration cells indicates other metabolic mechanisms or cellular pathways may be at
work. Currently these mechanisms are by and large poorly understood.

One of the most well studied plant groups in regards to phytolith production are
the monocotyledons (or monocots). This focus on monocots in phytolith studies is due in
part to the location of the true grass, or Poaceae, family within this monophyletic group.
Grasses have been the focus of intense phytolith studies because of their economic
importance both today and in the past. As a result of this intense study, plants in the
Poaceae family, of which maize is one, are known to produce abundant and distinctive

phytoliths.

65

For phytolith production in maize, there is recent evidence to suggest for some
genetic control (Pipemo 2006:11). It has been demonstrated that the same genes which
control the deposition of Iignin (a chemical compound which provides rigidity to plant
cell walls and aids in providing overall structural support to a plant), also control
phytolith production in maize. As a result of domestication, the genes which controlled
for heavy lignification of the cob were selected against and a softer cob was favored.

This genetic change also caused the formation of phytoliths of different types and

numbers in maize (Pipemo 200621 I).

Research Design

Since it is known that maize phytolith production can be genetically controlled
and is not effected by environmental conditions, systematic experiments were designed
involving green and mature maize to test if there was a difference in phytolith production
or count between green and mature maize. It is hypothesized that green maize has lower
phytolith production than mature maize due to its earlier stage of growth (Mulholland
personal communication 2006). If there are differences in maize phytolith
concentrations, this could indicate which stage of maize is being used in the
archaeological record. A significant issue concerning the use of maize is its initial
utilization in its green form. The initial use of green maize is a likely contributing factor
to its low macrobotanical visibility in earlier contexts due its probable immediate use.
Green maize is most likely not being processed and stored upon first introduction.

In addition to the issue of green versus mature maize use, the experiments were
designed to also assess the potential of determining the percentage of a plant used to

create a residue based upon phytolith density. The ability to quantify phytolith

66

concentrations based on the percentage of maize in a pot would allow researchers to
assess the ‘importance’ of maize in a prehistoric diet. This would alleviate problems of
subjective descriptors assigned by paleoethnobotanists based on macrobotanical remains
(Chilton 2002). As Chilton (2002) correctly points out, the number of kernels found in
some archaeological contexts could come from one or two cobs. The overall importance
of these experiments, if successfirl, would be to develop a baseline against which to

analyze archaeologically derived materials.

Experimental Residues: Methods and Materials

A series of controlled experimental residues were created to test maize phytolith
taphonomy. These residues utilized maize in various forms of processing and stages of
development. The various forms with the type of processing used are listed in Table 4-1.
Experimental residues were created in increasing 20% increments and the forms of maize
utilized were separately combined with ground venison. This made for a total of 25
residues created, processed, and analyzed by the researcher. All experimental residues,
except the hand ground maize flour, were carbonized in 100ml glass beakers in the
researcher’s kitchen cum laboratory. The hand ground maize flour residues were
carbonized in aluminum cans over a charcoal grill. The difference in burning procedures
was due to the season in which experiments were undertaken as all but the hand ground

maize residues were created during the winter months.

67

Table 4-1: Maize form and processing used in experimental residues

 

 

 

 

 

 

Maize Form Description of ProcessingTechnique
H d d maize flour Maize kernels removed fi'om the cob and ground using
an groun a replica mano and metate
. Maize kernels removed by hand from a dried cob
9”" :dh‘f’r'jrfiemglg’ hand (kernels split between this and lightly pounded,
rem" “0 y described below)
11‘. ightlly porindeéiddglsr: cob Split sample of maize kernels from dried whole kernels,
b32212 re 0V lightly pounded with a hammer
33:16:" kernels removed by Maize kernels removed by hand with a chefs knife
Maize cob left intact, but divided into segments using a
Whole green 00b chef’s knife

 

 

 

 

Processing of Each Maize Form
Hand ground maize flour

The hand ground maize flour was ground using a replica mano and metate. Dried
maize kernels from an ear of Iroquois White were removed by hand. Undergraduate
student volunteers from my Anthropology 320 class took turns hand grinding the maize
kemels (see Figure 4-1). Proportionate maize to venison (for example, 2g of maize to 8g
of venison) mixtures were measured out on an analytic balance and placed in a cleaned
aluminum can. A total of 10 mixtures (IO-100% maize) were created. To better combine
the maize-venison mixtures, 30ml of water was added to each can and stirred to combine.
The cans were then placed on a charcoal grill, and allowed to carbonize (see Figure 4-2).
In order to ensure total carbonization, the mixtures were allowed to burn for 5 hours.
Once the experimental residues cooled, they were removed and placed in separate labeled
plastic sample bags. A sample of non-carbonized maize flour was taken as well in order

to assess phytolith taphonomy prior to burning.

68

Figure 4-1: Student volunteers hand grinding maize (photo taken by author)

 

Lighfly-pgunded dried maize kernels
Dried maize kernels were removed by hand and placed on a baking a sheet. A

hammer was used to lightly pound the kernels in order to crack them. Some kernels did
fracture into smaller pieces, however most only cracked slightly while a few did remain
intact. Kernels were then weighed out on an analytic balance and combined with
proportionate measures of venison (for example 2g maize kernels to 8g venison). A total
of 5 samples (20%, 40%, 60%, 80%, 100%) were created and placed separately in 100ml
glass beakers. To better combine the mixtures, 30ml of water was added to each. The
samples in beakers were then placed in a 232°C (450°F) oven and allowed to carbonize
for 5 hours in order to ensure total carbonization. Once samples had cooled, they were
removed and placed in separate labeled plastic sample bags. A sample of non-carbonized
lightly-pounded kernels was taken as well in order to assess phytolith taphonomy prior to
burning.
Whole dried maize kernels

Half of the dried maize kernels from the lightly-pounded dried maize kernel
experiment were reserved and utilized for the whole dried maize kernel experiment.
These were set aside prior to processing the kernels for the lightly-pounded maize
residues. Kernels were weighed out on an analytic balance and combined with
proportionate measures of venison. A total of 5 samples (20-100%) were created and
placed separately in 100ml glass beakers. Thirty millimeters of water was added to each
to better combine the mixtures. The samples in beakers were then placed in a 232°C
(450°F) oven and allowed to carbonize for 5 hours in order to ensure total carbonization.

Once samples had cooled, they were removed and placed in separate labeled plastic

70

sample bags. A sample of non—carbonized whole dried kernels was taken as well in order
to assess phytolith taphonomy prior to burning.
Green maize kernels removed by hang

Green cars of maize were bought at a local Meijer grocery store. Kernels were
removed by hand from the cob using a chef's knife in the researcher’s kitchen. Kernels
were weighed out on an analytic balance and combined with proportionate measures of
venison. A total of 5 samples (20-100%) were created and placed separately in 100ml
glass beakers. Thirty millimeters of water was added to each to better combine the
mixtures. The samples in beakers were then placed in a 232°C (450°F) oven and allowed
to carbonize for 5 hours in order to ensure total carbonization. Once samples had cooled,
they were removed and placed in separate labeled plastic sample bags. A sample of non-
carbonized green kernels was taken as well in order to assess phytolith taphonomy prior
to burning.
Whole cob green maize

Green cars of maize bought in the same trip as that used in the green kernels
removed by hand were used for this set of residues. Green maize cobs were cut into
pieces using a chef‘s knife in the researcher’s kitchen. The green cob pieces were
weighed out on an analytic balance and combined with proportionate measures of
venison. A total of 5 samples (20-100%) were created and placed separately in 100ml
glass beakers. To better combine the mixtures, 50ml of water was added to each sample.
The amount of water was increased in this set of experiments to ensure total coverage of
the cob pieces. The samples contained in beakers were then placed in a 232°C (450°F)

oven and allowed to carbonize for 5 hours in order to ensure total carbonization. Once

71

samples had cooled, they were removed and placed in separate labeled plastic sample
bags. A sample of non—carbonized green maize cob was taken as well in order to assess

phytolith taphonomy prior to burning.

Laboratory Processing Procedures

Experimental residues were processed in the Pollen Laboratory of Dr. Catherine
Yansa, Department of Geography, Michigan State University. A sample of each
experimental residue was taken using a separate, clean disposable scalpel for each residue
and weighed on an analytic balance. The entire residue was not processed for phytoliths
in order to replicate the procedure for sampling archaeological ceramic residues. Taking
the entire archaeological residue for phytolith processing would be avoided if possible
and so I felt the same circumstances should be replicated for the experimental residues.
The weight of the samples was determined by first weighing the test tube each sample
would be placed in, then adding the sample and weighing the tube again. Subtracting the
original tube weight from the combined sample and tube weight provided the weight of
the sample.

After each sample had been weighed, processing procedures for phytolith
extraction followed the protocol developed by Rapp and Mulholland at the Archaeometry
Laboratory, University of Minnesota—Duluth. A complete description of the processing
protocol can be found in Appendix A. Schulze solution (a 3:1 mixture of nitric acid to
dissolved potassium carbonate) was added to each test tube in the quantity of 10ml. Test
tubes containing samples were then placed in a hot water bath (heated to 90°C) and
allowed to sit for 24 hours to ensure complete dissolution of carbon in the sample. If any

carbon remained after 24 hours, the samples were centrifirged, supernatant decanted and

72

10ml of nitric acid added to the samples. Samples were then placed in the hot water bath
for another 24 hours. Extra processing time was unusual for the experimental samples.
If extra processing time was needed only nitric acid was used in order to provide a
different caustic environment.

Once the carbon in the samples had dissolved, water rinses were performed to
remove the chemicals from the samples. Samples were centrifuged for 15 minutes at
3000 rpm, the supernatant decanted, and samples agitated with a stirring rod. Distilled
water was used to rinse the stirring rod into the test tube to ensure phytoliths did not
adhere to the stirring rod. Extra water was added to the tube as needed to ensure similar
water levels in all tubes for balance purposes within the centrifuge. Once each sample
had gone through this procedure, samples were centrifuged again (for 15 minutes at 3000
rpm) and this process was repeated until the liquid in the sarnples was clear. The number
of water rinses ranged from 4 to 8, depending on how large (i.e. the original weight) the
original sample was.

After water rinses were completed, an ethanol rinse was performed. The
supernatant was decanted and samples were transferred to labeled 1dr storage vials using
a disposable pipette. A final rinse of the test tube and pipette with ethanol was performed
to better ensure total collection of phytoliths in each sample. If too much ethanol ended
up in the storage vials, these were centrifuged for 15 minutes at 3000 rpm (to ensure
sample material remained at the bottom of the vial) and the extra ethanol was removed
with a clean disposable pipette.

Microscope slides were made by placing three drops of the processed material

suspended in ethanol on a 3x1 inch slide and placing this on a slide warmer to allow the

73

ethanol to evaporate. Once the ethanol had evaporated, 4-6 drops of Permount was added
to mount slides for analysis. Slides were scanned using a Leica DM2500 microscope

with dark field and phase contrast capabilities at 40x magnification.

Experimental Residue Results

A summary of the experimental residue results are presented in Table 4—2. The
results were initially surprising and unexpected. The hand ground maize flour residues
were the first set experimental residues processed and analyzed by the researcher. It was
due to these residues that the researcher discovered the presence of starch remaining in
the residues after processing. Further discussion with other phytolith researchers, chiefly
Dr. Deborah Pearsall of the University of Missouri and Dr. Matthew Boyd of Lakehead
University, ON, proved that this was a fairly common occurrence. According to Dr.
Boyd (personal communication 2009), while there is very likely some starch attrition due
to the acidic nature of the Schulze processing procedure, starch is also left behind in the
residue. As a result, starch identification was added as another microbotanical
component to the residue analyses of both the experimental and archaeological samples.

The residue results presented in Table 4-2 portray some interesting trends. The
majority of phytolith occurrence found in a residue occurs with the use of whole green
cob maize. While at first surprising, it makes sense when one thinks about where and
how phytoliths form in maize. Phytoliths are created through silica deposition both in
between and within plant cells. This deposition can vary between species and also
between various plant parts (roots, leaves, stem, and inflorescence). In the case of maize,
most diagnostic phytoliths are found in the leaves, tassels, and cob and glumes.

Diagnostic phytoliths are not found in the kernel itself. This means then that the maize

74

kernel pericarp will only have basic body cell phytoliths that are non-diagnostic. Well, so
what? Why does that matter?

Table 4-2: Microbotanical results of experimental residues

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

'/o Maize in .
Maize Form Experimental Phytollth Starch Count
Residue Mix Count
Whole Green Cob 20% 73 0
40% 156 2
60% 50 7
80% l 1 0
100% 63 5
Whole Green
.Kemel 20% O l
40% 1 20
60% 0 O
80% 0 5
100% 0 0
Li htl Pounded
[gried Kernel 20% O 84
40% 0 44
60% 0 25
30% o 22
100% 1 1 17
Whole Dried
_ Kernel 20% l 56
40% 0 13
60% 0 18
80% I 53
100% 0 53
Hand Ground
Maize Flour 20% O 53
40% 0 24
60% 0 11
80% 0 27
100% 0 '9

 

 

75

of which form of maize is utilized first. In the experimental residues analyzed, all but
one form required removal of the maize kernel from the cob. Removal of kernels item
the cob means any diagnostics that would have shown up in these removed forms would
have been incorporated accidentally from any maize glumes that remained attached to the
kernels after removal. These ‘accidental.’ inclusions can be seen in the single maize
phytolith counts from the whole green kernel, lightly pounded dried kernel, and whole
dried kernel residues. As the phytolith counts in Table 4-2 demonstrate, significant
phytolith counts occur with whole green cob. Therefore, any significant phytolith
identification in residues is most likely from the use of whole green maize on the cob.

The flip side to the phytolith count data is the starch count data. As Table 4-2
demonstrates, maize starch is much more likely to occur in residues when maize kernels,
whether dried or green, are removed from the cob. Dried forms of maize (lightly
pounded whole kernel, whole dried kernel, and maize flour) will also have a greater
abundance of starch than the green forms. Once again, this is not as surprising as it might
initially seem. Diagnostic forms of starch come from long term storage organs of plants;
in the case of maize, this long term storage is in the kernels. These storage organs have a
seasonal build-up of starch; starch will be more abundant in the fall and winter than in the
Spring and summer (Messner 2008). Dried forms of maize represent the mature plant and
therefore starch in maize kernels will be at its peak. Lower starch production in the green
form of maize is to be expected since it would have been harvested early and starch

production would not have been at its peak. The experimental starch counts for the green

76

forms reflect this expectation; whole green kernel and whole green cob have minimal
starch counts.
Discussion: Phytolith and Starch Taphonomy

Location of diagnostic phytolith and starch production leads directly to the issue
of processing. Which human behaviors will most likely result in the incorporation of
either maize phytoliths, maize starch, or both in a residue? The results of the
experimental samples indicate that diagnostic maize phytoliths (Figure 4-3) found in
ceramic residues are more likely to be a result of the use of green maize when in its
whole form. This is not to suggest, however, that phytoliths from the cob and glume
could not be accidentally included via hand removal in the other forms of processed
maize. Based on the low to zero counts of the dried forms, however, it is most likely the
green form of maize that is utilized when maize phytoliths are found in early dated

archaeological ceramic food residues.

Figure 4-3: Diagnostic maize rondel

 

This likelihood is further corroborated when combined with the starch data.

Starch was present in abundance with the experimental residues created from the dried
77

forms of maize (Figure 4-4). The presence of only maize starch in a sample more likely
indicates a dried form of maize utilization while the presence of starch and phytoliths
together, is more likely indicative of a green form of maize utilization. Further
processing information can potentially come from damage to the starch granule itself
(Figure 4-5). Current evidence suggests there is specific damage to maize starch if it is
ground or pounded in the preparation of maize for flour (Pearsall conference paper;
Pipemo 2006; others). Identification of maize starch that has been ground or pounded
could further indicate the use of maize flour. While somewhat crude at a scalar level, this
data does allow us to identify how people were potentially utilizing maize, and therefore
make statements on behavior, based on the presence or absence of diagnostic maize

phytoliths and starch in residues (see Table 4-3).

Figure 4-4: Undamaged diagnostic maize starch

 

78

Figure 4—5: Maize starch with evidence for grinding

\. A
w’tk"

[r1 T ITTT FT]

0 25 50” erY‘. '
e . A ._ id,

 

Table 4-3: Probable utilization of maize based on presence/absence of diagnostics

 

Probable behavior Residue Evidence

 

Diagnostic cob phytoliths OR

Utilization of maize ln green form diagnostic cob phytoliths and starch

 

Utilization of maize in dried form Diagnostic starch only

 

Diagnostic starch only; some starch
Utilization of maize in flour form may show grinding or pounding
damage

 

 

 

 

Conclusion

While it is entirely probable that a combination of green and dried forms could

result in the presence of diagnostic phytoliths and starch in the same residue, this is

79

unlikely until maize use has become more ubiquitous. Current studies utilizing ceramic
residues to identify earlier dates for maize, dates a few hundred years before ubiquitous
maize use, consistently identify maize phytoliths with limited to no identification of
maize starch (c.f. Hart et al. 2004; Thompson et al. 2005). I propose that in these earlier
contexts, it is more likely that maize is being utilized in its green form since it is a new
and unknown, or at least not well known, plant. It is not being dried and stored for later
use or ground into flour. Instead, it could very well be boiled whole. Such processing
could potentially explain the presence of maize diagnostics in residues from earlier
contexts prior to the adoption of intensive maize agriculture. Intensive maize agriculture
is associated with large scale drying and storage of surpluses and is corroborated by the
presence of maize macrobotanicals in conjunction with storage pits. It is far more likely
that in these recently identified earlier contexts, some portion of the whole cob was being
cooked rather than an accidental incorporation of glume material coinciding with the
removal of dried maize kernels from the cob.

Returning to the issue of quantification of maize phytolith density based on the
percentage of maize used to create a residue or quantification of maize phytolith density
based on the use of green or mature maize, at this point quantification of either of these
scenarios has found to be untenable. The presence of diagnostic phytoliths primarily
occurring in only the whole green cob set of experimental residues made it pointless to
attempt phytolith quantification of green versus mature maize.

Quantification of maize phytoliths based on the percentage of maize used to
create a residue is also moot for the same reason. While it may have been possible to

quantify maize phytoliths to percentage of maize in the pot with the whole green cob

80

form, there appear to be other potential taphonomic factors not currently accounted for in
this study. One would expect a general trend of more diagnostic maize phytoliths
coinciding with a greater percentage of maize utilized in residue creation, however this
does not appear to be the case. If quantification based on percentage of maize cooked in
a pot had been successful, there is also the issue of how applicable this quantification
would have been to archaeological residues. Are archaeological residues adhering to
ceramics the result of one episode of accidental burning or is it a result of repeated use of
the pot and the residue represents an accretion of burned and charred food? If a residue is
created due to accretion, repeated cooking of a smaller percentage of maize over time
could mimic a cooking episode of a higher percentage of maize.

Since quantification of maize phytolith density was found to be flawed, the
possibility of quantifying maize starch density still remains. The processing procedures
utilized in this study however make such an endeavor unwise. It is assumed that some
starch attrition occurred with the Schulze solution processing. There are also a number of
as yet unknown taphonomic issues with starch production and incorporation into ceramic
residues (Barton and Matthews 2006). Starch is both a highly durable but also fiagile
compound and is susceptible to dissolution and gelatinization from the cooking process
(Barton and Matthews 2006; Messner 2008). A firmer understanding of these issues is
required before attempting any quantification of starch based on the percentage of a plant
used to create a residue. If and/or when these issues are clarified, the reuse of a pot
archaeologically, and how reuse effects residue taphonomy will likely remain antissue.

Similar taphonomic issues with ceramic residues were found by Hart et al.

(2007, 2009) in regards to the use of stable isotopes to identify maize in ceramic residues

81

adhering to ceramic vessel walls. In the Hart et al. studies, controlled cooking
experiments utilized various forms of maize individually mixed with various individual
C3 components (wild rice, chenopod, and deer). These residues were created to test the
threshold of 6‘3 C and 8 '4C isotopes for maize use in ceramic residues that had been
proposed by Morton and Schwarz (2004). The Hart et al. studies found that maize was
overwhelming masked by C3 producers and that maize might not be identifiable via the
stable isotope analysis of residues until maize represents 80—90% of a mix. From the
experimental work undertaken for this dissertation, in conjunction with the experiments
performed by Hart et al. (2007, 2009), it is clear then that quantification of these types of
variables through a number of different proxies may be difficult for multiple reasons and
may not be feasible beyond presence/absence identification of maize.

The application of these results to the archaeological data are discussed in
chapter six. The next chapter provides background on the sites used in this research,
especially the existing subsistence data. The data provided by the microbotanical
analyses of the archaeological residues and how they add to or complement the existing

data is also presented in the following chapter.

82

CHAPTER 5

Archaeological Residue Analysis

W

A total of 79 sherds from various archaeological sites in the Saginaw Drainage
were sampled. Appropriate collections were identified from site inventories at Michigan
State University (MSU), University of Michigan Museum of Anthropology (UMMA), the
Michigan Historical Center (MHC), and the Saginaw Archaeological Commission (SAC)
at the Castle Museum. Once sites dating to the desired time periods (Middle to Late
Woodland; ZOOBC—AD 500 and AD 600-AD 1400 respectively) were identified (see
Table 5-1 for a list of sites), I went through each collection to determine the number
ceramic sherds with residues were present. After this information had been collected
from each suitable site in the study area, Ideterrnined which residues to sample based on
time period and date range, and obtained formal permission for destructive sampling
from each institution to do so.

Sampling of each sherd took place in the Pollen Laboratory of Dr. Catherine
Yansa, Department of Geography, Michigan State University. Prior to sampling, the
ceramics were photographed to record where the residue was located on the sherd.
Residues were sampled by taking a clean, disposable scalpel and scraping it against the
sherd to remove a portion of the residue. Since a subsample of these sherds were going
to be radiocarbon dated, all scalpels were labeled with the site name and ceramic ID and
put back in their original packaging for storage. This was done to ensure that the same
scalpel would be used if it was decided to sample that sherd for ”c dating. Due to the

possibility of radiocarbon dating and to ensure the possibility of future studies, a

83

minimum amount of residue was sampled from the sherd. The amount of residue taken
varied based on how much residue was adhering to the ceramic but ranged from .00010g
to .0290g. Sample weights were calculated indirectly by weighing the piece of weigh
paper used, scraping off a portion of the residue onto the weigh paper, weighing the
combined weigh paper and sample, and then subtracting the weigh paper weight from the
weigh paper plus residue weight. A Cole Partner Symmetry PA analytic digital balance
was used to determine the weight of all samples. After the sample was weighed, it was
immediately funneled into a labeled centrifuge tube by creating a funnel from the weigh
paper used in determining the sample weight.

Archaeological residue samples were processed in sets of 16; this reflects the
capacity of the centrifuge used in the laboratory. The same processing techniques used
for the experimental residues were used to process the archaeological residues (see
Appendix A). Depending on the size of the sample, some archaeological residues
required a second day of processing in the Schulze solution for all of the carbon to
dissipate. The same procedure in making slides for the experimental residues was also
followed for the archaeological residue samNes.

Slides were scanned for the presence of both phytoliths and starches since it was
determined via the experimental residue results that starches remained in samples, albeit
probably with some attrition. A comparative phytolith collection was obtained by
photographing diagnostic forms from samples at Dr. Deborah Pearsall’s laboratory at
University of Missouri as well as by creating slides of comparative material from the
laboratory of Dr. Susan Mulholland at University of Minnesota—Duluth. Due to a lack

of comparative starch samples in my possession, the dissertation of Timothy Messner

84

(2008), “Woodland Period People and Plant Interactions: New Insights from Starch Grain
Analysis,” proved invaluable. Messner (2008) describes and photographs starch grains
for both medicinal and subsistence plants in the Delaware River Watershed (DRW).
Many of the plants described overlap in both their range and use prehistorically for the
DRW and the Saginaw Drainage of Michigan. Messner also provided further assistance

with starch identification via email.

Table 5-1: Middle and Late Woodland sites examined for residues; x denotes presence

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Site Time Period Residue Present
Liberty Bridge (ZOBY77) Middle Woodland x
Dutch Creek (20BY200) Middle Woodland
Miller (208A7) Middle Woodland
Satchell (20$A33) Middle Woodland
Fletcher (ZOBY28) Middle and Late Woodland x
Marquette Viaduct (20BY28) Middle and Late Woodland
Kantzler (20BY30) Middle and Late Woodland x
Schultz (ZOSAZ) Middle and Late Woodland x
208A1276 Middle and Late Woodland x
208A1251 Middle and Late Woodland x
Bndg(e2%()sr;:t2)3;nshrp Middle and Late Woodland
East Oxbow (20MD490) Middle and Late Woodland
Stroebel (2OSA14) Middle and Late Woodland
Cook (208A3 1) Middle and Late Woodland
Cassasa (208A1021) Late Woodland
Slavik (ZOGR221) Late Woodland
Tobico (20BY192) Late Woodland x
Weber 1 (203A581) Late Woodland x
Simons 249 (ZOSA665) Late Woodland x
Trombley House (2OBY70) Late Woodland
Solms (ZOSASD Late Woodland x
Mahoney (203A193) Late Woodland
__ Stadelmeyer (208A195) Late Woodland x
ZOBY210 Late Woodland x
Foster (ZOSA74) Late Woodland x
Butterfield (208Y29) Late Woodland
20BY192 Late Woodland x
. Late Woodland and Late
_* Clunie (208A722) Prehistoric x

 

85

 

Site and Ceramic Descriptions

Locations of the sites in the following discussion are presented in Figure 5-1.
These sites reflect those from Table 5-1 which were found to have ceramics with residue.
It should be noted that for the Saginaw Drainage, the Middle Woodland period coincides
with the presence of Hopewell ceremonial traits. Hopewell expression throughout the
Midwest, Great Lakes, and Southeast varies but is characterized by significant mound
construction (examples include sub-conical, elongate, and geometric earthwork
enclosures), exotic raw materials (obsidian, Gulf coast shell, copper, minerals such as
galena and mica), and elaborate burial customs including log-lined tombs, inclusion of
exotic artifacts, burial ‘houses’, and post-mortuary processing (Maxwell 1952; Morgan
1952; Seeman 1992; Wymer 1987). In the Saginaw Drainage, Hopewell expression
appears to be a variant of Illinois Hopewell (Kingsley 1999; Brashler et al. 2006). This is

not to suggest, however, that contact with other regional groups did not occur.

Liberty BriCLge (ZOBYTm Middle and Late Woodland

Site 20BY77 is a Woodland period occupation of the Liberty Bridge locale.
The site was located as a result of excavations conducted in association with the Third
Street Bridge right of way project in Bay City, MI (Egan 1993; Lovis 1993a). During
excavations conducted by Lovis in 1984, a total 1 17m2 of the site was excavated (Lovis
1993a). This includes a portion of site ZOBY79 (Block 1) which was found to be
identical to ZOBY77 in regards to depositional sequence and cultural material.
Delineation of site boundaries between 20BY77 and 20BY79 was placed at the railroad
tracks which cut through the area under investigation; it is likely that the eastern portion

of ZOBY79 represented in Block 1 is a part of ZOBY77 based on the sequence of soil

86

l. l tl r .1

Figure 5-1: Site Locations

 

N I

W‘ E

0 5 10 20
-::_Kilometers

" “‘ij

, Legend
:3 --— Rivers

l 7 Counties

1 ,

8 1 River

 

 

 

 

 

 

 

 

 

Sites: I, 2: Stadelmeyer (20$Al95) & Solms (208A57); 3: Clunie (20$A722);

4, S: 2OSA1276 & 2OSA125 I; 6: Schultz (208A2); 7: Foster (20SA74); 8: 208A665;
9: Kantzler (20BY30); 10: Liberty Bridge (20BY77); ll: Fletcher (20BY28);

12, 13: Tobico Marsh sites (20BY192 & 208Y210)

87

deposition and cultural material recovered (Lovis 1993a; Robertson 1993). Site ZOBY77
is located within Bay City on the west bank of the Saginaw River. The site itself sits on a
ridgejust above modern lake levels at approximately 183m asl (Monaghan 1993).

Activities conducted at ZOBY7 7 appear to include hide working, hunting and
butchering activities and associated lithic tool manufacture for these activities and
ceramics present at the site increase in density over the Woodland period (Lovis and
Davis 1993; Robertson 1993). Features present at 20BY77 do little to illuminate site
functionality since none of the eight cultural features identified were hearths or roasting
pits despite the presence of fire-cracked rock throughout the site (Lovis 1993b).
According to Lovis (1993b:194), “All features appeared midden filled and empty,
suggesting they were used for short-term, task-specific activities and then emptied.” Site
ZOBY77 appears to represent a short-term residential occupation given the suggested use
for pit features, presence of ceramics, and extent of lithic manufacture and tool use.

The Middle Woodland component of 20BY77 appears to be a summer
occupation, although this determination is tenuous. The faunal evidence for warm
weather seasonality is based on a young white-tailed deer (age approximately six
months), painted turtle, and freshwater mussels. Other faunal evidence for summer
occupation is tentative; painted turtle are more easily caught in the summer months and
warmer water facilitates mussel collection (Franz 1993). Despite its proximity to the
Saginaw River, there are a limited number of fish species represented at 20BY77, in
contrast with other Middle Woodland occupations in the vicinity where fish remains

dominate the faunal assemblage.

88

The Late Woodland occupation of 20BY77 is more difficult to identify due to
its sparse nature. No macrobotanical remains from the Late Woodland occupation at
20BY77 were identified. However at nearby site ZOBY79, both chenopod and sumac
were recovered in the Late Woodland occupation (Cleland 1993). F aunal remains are
also poorly represented in the Late Woodland component and much of what was
recovered was too fragmentary to be identified (Franz 1993).

The botanical remains recovered from 20BY77 do not clarify site seasonality
much either. A list of macrobotanical remains is found in Table 5-2. Chenopod and
nutshell fiom hickory and walnut were recovered. Chenopod seeds are commonly stored
but their greens are available in the spring while hickory and walnut are available in the
fall (Cleland 1993).

A total of three sherds with residue from Liberty Bridge (20BY77) were
sampled for this project. The sherds sampled date to the late Middle Woodland (Green
Point ware) and early Late Woodland periods (Appendix B, Figures Bl-B3). Phytolith
and starch results are found in Table 5-3. A radiocarbon sample from a sherd stylistically
typed as Green Point ware was sent to Beta Analytic for dating. Unfortunately this was
determined to be undateable due to insufficient material for radiocarbon analysis.

Phytolith and Starch Results

For the Middle Woodland occupation, maize starch was the only identifiable
microbotanical. This new data does little to clarify site seasonality. During the Late
Woodland, the combined presence of maize, wild rice, and acorn could indicate a late
summer to fall occupation since this is their peak harvest time. However all of these

plants are storable and could potentially be used year round. These data can be

89

considered consistent with a short-term residential occupation of the site since they could

have been temporarily stored in the pit features discussed above.

Table 5-2: Macrobotanical Data for Liberty Bridge; x denotes presence

 

 

 

 

 

 

 

 

Scientific Name Common Name Middle Woodland Late Woodland
C henopodium sp. Chenopod x
Cary/0% Hickory x
Juglans Sp. Walnut x

 

Table 5-3: Phytolith and Starch Data for Liberty Bridge; x denotes presence

 

 

 

 

 

 

Middle Woodland Late Woodland
Scientific Name Common Name
Phytolith Starch Phytolith Starch
Zea mqs Maize x x x
Zizam'a aguatica Wild Rice x
Quercus Sp. Oak (Acorn) x

 

 

 

 

 

 

Fletcher (20BY28): Middle grid Late Woodland

The Fletcher site (20BY28) is located in Bay City, M1, on the west bank of the

Saginaw River at an elevation of approximately 178m asl (Lovis I985). Periodic

professional excavation of the Fletcher site occurred from 1967 to 1987, while previous

surface collection by amateur archaeologists had also occurred (Brashler 1973; Egan

1993; Lovis 1985). The initial 1967 excavation, performed by Dr. Moreau Maxwell

(MSU), was undertaken after the US. Army Corp of Engineers came upon human

remains and artifacts during excavation for a sediment retention basin in the northern

 

 

portion of the site (Egan 1993). Subsequent excavations of Fletcher occurred in '1968
(also by Dr. Maxwell), in 1970 by Dr. James Brown (then of MSU), and lastly in 1988 by

Dr. William Lovis (MSU) when construction threatened the site.

90

The Fletcher site represents continuous occupation from the Terminal Archaic
to the Late Woodland and into the historic period as late as the 19th century. Major
occupation of Fletcher occurred during the late Middle Woodland and Late Woodland
periods based on lithic and ceramic assemblage concentrations (Lovis et al. 1996). One
problem encountered during excavations at Fletcher was the erosion, redeposition of
material, and disturbed areas of the site, especially those areas relating to the Middle and
Late Woodland components, due to fluctuating lake levels and flooding (Lovis et al.
1996).

Subsistence remains found at Fletcher indicate a mixed-subsistence economy.
There is little Early Woodland subsistence data available but Middle and Late Woodland
subsistence data is abundant. During the Middle Woodland, a variety of both terrestrial
and aquatic resources are present. Faunal remains present include deer, bear, wapiti,
beaver, otter, muskrat, waterfowl, and fish (sturgeon, channel catfish, northern pike,
drum, perch/walleye). Floral remains identified by Egan (1993) include nuts (hickory,
black walnut, acorn), seeds (chenopod, hawthorn, huckleberry), and both aquatic and
terrestrial tubers (see Table 5-4). Taken together, the faunal and floral remains indicate a
spring-summer occupation. However, due to post-depositional disturbance, the intensity
and duration of the Middle Woodland occupation of Fletcher is unclear (Egan 1993:191).

According to Lovis (1985), the Late Woodland occupation of Fletcher
represents peak occupation due to the presence of a full range of lithic reduction activities
as well as division of space across the site. Habitation areas and specialized activities
marked by hearths and storage pits are clearly separated from each other and cluster in

specific areas of the site (Lovis 1985). There is also significant and overlapping midden

91

accumulation during the Late Woodland occupation which supports intensive site
occupation at this point in time (Lovis 1985). F aunal remains identified to the Late
Woodland occupation suggest spring to autumn seasonality; floral remains were not
identified.

Given the extent of resources exploited both during the Middle and Late
Woodland occupations of Fletcher, site function has been interpreted as a long-term
collective resource extraction site with more intensive occupation occurring during the
Late Woodland (Egan 1993). Faunal and floral remains identified indicate multiple
seasonality of use. Extensive pit features excavated at Fletcher support this
interpretation; most pit features appear to have been used for storage of items that were
later removed (Lovis 1985). The presence of spatially segregated patterns of postrnolds
also indicates designated areas for residential activities separate from that of storage

features and various processing activities (Lovis 1985).

Table 5-4: Macrobotanical Data for Fletcher; x denotes presence

 

 

 

 

 

 

 

 

. . . Late

Scientific Name Common Name Middle Woodland Woodland
Quercus sp. Oak (Acorn) X
Carya Sp. Hickory X
Juglans nggra Black Walnut x
Chenopodium sp. Chenopod X
Crataegus sp. Hawthorn x
Vaccinium sp. Huckleberrjy X

Aquatic and/or

NA Terrestrial Tubers x

 

 

 

 

 

 

The Fletcher site material was used in a pilot study I conducted in order to
determine the feasibility of deriving phytoliths from ceramic residues. Three sherds with

residue were identified and sampled in this analysis conducted under the supervision of

92

Dr. Susan Mulholland (University of Minnesota—Duluth). Unfortunately these rims
were not photographed prior to their return to the MSU Museum Repository. Only
rimsherds were assessed for the presence of residue and as a result, it is not possible to
state which percentage of sherds was sampled from the total ceramic assemblage. Based
on observations of residues on the rims, however, it appears that residues are rare,
perhaps due to post-excavation processing for museum storage.

Phytolith a_n_d Sgch Results

While phytoliths were identified in all three residues, none of them were
diagnostic. They were all identified to be a generic wild grass rondel form. Starch was
also not identified in any of the three residues as well. This is surprising given the
identification of tubers in the macrobotanical remains from Fletcher. Currently, these

results do not add to interpretation of the site.

Kantzler (20BY30); Middle and LatLe Woodlflig

The Kantzler site was excavated by the Saginaw Drainage Chapter of the

 

Michigan Archaeological Society in 1965 and by the University of Michigan Museum of
Anthropology in 1966. Kantzler is located within Bay City, M1 on an Algoma terrace on
the western bank of the Saginaw River, at an elevation of approximately 183m. Due to
its location in the estuarine area of the Bay City moraine, and as a former floodplain for
the Saginaw River, flooding has been a feature of the site location throughout its history
(Crumley 1973).

The earliest occupation of Kantzler dates to the Early Woodland, where site
occupants utilized the area for fishing. Most of the fish species recovered, such as

walleye, catfish, gar, and drum, are present in the Saginaw River year round (Crumley

93

1973). However the presence of young deer as well as whistling swan, ring—necked duck,
and passenger pigeon indicate a spring-summer occupation of the site. Only one ceramic
sherd dating to the Early Woodland was recovered during excavations. Upon
examination of the collection in the UMMA, no residue was present on the sherd.

The Middle Woodland occupation of Kantzler can be divided into Early Middle
Woodland and Hopewell (middle to late Middle Woodland) occupations based on the
ceramics present (Crumley 1973). The ceramics sampled for this project, including both
Tittabawassee and Greenpoint wares, correspond to the Hopewell-related portion of the
Middle Woodland occupation. Evidence for occupation during the entirety of the Middle
Woodland suggests the site may have been in use for all economic seasons. During the
early Middle Woodland, fish species captured indicate spring and fall spawning fish (cod
and trout respectively), as well as those available year round. The mammal species
recovered at the site support use of the site for all economic seasons with the presence of
beaver, which is usually hunted during the winter (Crumley 1973).

Occupation during all economic seasons of Kantzler is surmised for the
Hopewell-related occupation period as well. The presence of beaver, elk, and muskrat
indicate winter hunting while the remains of young deer and acorns are indicative of
summer and fall resource exploitation (Crumley 1973). Macrobotanical remains were
sparse; two acorns and a modern grape seed were recovered (Table 5-5). The grape seed
reflects intrusion from the historic period occupation of the site area. The presence of
acorn could also be indicative of fall occupation; however these nuts were oftenstored as

well.

94

The Late Woodland occupation at Kantzler is difficult to tease out since
throughout this stratum it is mixed with historic material (Crumley 1973). Ceramics
present in the Late Woodland stratum appear to have stylistic indicators which may show
affmities with Point Peninsula wares from southern Ontario as well as wares commonly
found in Ohio (Crumley 1973). There is a known historic period Algonquin camp at
Kantzler sometime in the late 17th century and this material is mixed with more modern
historic material (Crumley 1973).

A total of 14 sherds with residue were identified from the Kantzler collection of
materials located at UMMA. Four of these sherds (29%) were sampled for this project
(Appendix B, Figures B4-B7). All three sherds date to the Middle Woodland period and
are either Greenpoint or Tittabawasse wares.

thtolithgnd Starch Results

Results of the phytolith and starch analysis are presented in Table 5-6. Maize
and wild rice were both identified from Middle Woodland sherds. The possibility of
maize during the Late Woodland is from the presence of a phytolith which could be
maize diagnostic. The identification as definitively maize is not possible since the
phytolith was not able to be rotated during analysis to view all sides. The identification
of maize and wild rice during the Middle Woodland occupation of Kantzler also supports
an interpretation for use during all economic seasons. These resources are available in

the late smnmer to early fall and can be processed for long-term storage.

Table 5-5: Macrobotanical Data for Kantzler; x denotes presence

 

Scientific Name Common Name Middle Woodland Late Woodland
Quercus Sp. Oak (Acorn) X

 

 

 

 

 

 

 

95

Table 5-6: Phytolith and Starch Data for Kantzler; x denotes presence

 

Middle Woodland Late Woodland

 

Scientific Name Common Name Phytolith Starch Phytolith Starch

 

Zea mays Maize x possible

 

 

 

 

 

 

 

Zizam'a agatica Wild Rice x

 

 

208A1276: Middle and Late Woodla_n_d

Site 2OSA1276 is located within the Shiawassee National Wildlife Refuge, in
Saginaw County, MI. Site excavations have been conducted by Mr. Jeffrey Sommer of
the Historical Society of Saginaw County. The site is 250m north of the Shiawassee
River and the main extent of the site is bordered on the west by an old drainage or
channel which runs between the Shiawassee and Titabawassee rivers (Sommer 2002,
2003b, 2005). Based on the ceramic and projectile types recovered, 2OSA1276 dates to
the Middle and Late Woodland periods. Floral remains recovered from flotation and
encountered during excavation include hickory and walnut (Sommer 2002, 2003b, 2005).
Faunal remains include deer, fish (catfish, sturgeon, gar), muskrat, and beaver (Sommer
2002, 2003b, 2005). Taken together the faunal and floral remains seem to indicate a fall
to spring occupation. Sturgeon is a spring spawning anadromous fish while muskrat,
beaver and nuts (hickory and walnut) are available during the fall and winter months.
Feature descriptions indicate burning activities from the presence of ash, reddened soil,
calcined bone, and charcoal (Sommer 2002, 2003b, 2005). Storage features have also
been identified and it is likely that 2OSA1276 represents a winter camp (Sommer 2002,

2003b,2005)

96

Phytolith and Starch Results
Results of the residue analyses for the sherds from 208A 1276 produced

minimal diagnostic phytolith and starch bodies. The only diagnostic forms were found in
two sherds which date to the Middle Woodland (see Table 5-7). The other Middle
Woodland sherd examined did not produce diagnostic forms and the Late Woodland
sherd examined had no phytoliths or starches present in the residue. Two sherds from
this site were AMS dated. One sherd was not stylistically identifiable while the other was
identified as Tittabawassee Ware, which dates to the early Middle Woodland (Appendix
B, Figures B8-BIB). Both radiocarbon dates derived from these samples are early in the
Middle Woodland period; the unidentifiable sherd dated to Cal BC SO-Cal AD 90 (2
sigma) while the Tittabawassee Ware dated to Cal AD 10-140 (2 sigma; Sommer 2003a).
The plants identified through microbotanical analysis are available in late
summer and early fall. The best time to harvest wild rice is generally early September to
October (Vennum 1988). Acorns become available during the fall, and the presence of
maize starch (Figure 5-2), which likely indicates mature maize, would also be available
late summer-early fall. The plants identified also represent resources which are
conducive to long-term storage, are easily collected, and can occur in abundance. This
data supports the interpretation that 205A1276 is likely a fall to early spring occupation

and represents a winter camp.

97

Table 5-7: Phytolith and Starch data for 20$A1276; x denotes presence

 

 

 

 

 

 

 

 

 

 

Scientific Name Common Middle Woodland Late Woodland
N’me Phytolith Starch Phytolith Starch
Zea mays Maize x
Quercus Sp. Oak (Acorn) x
Zizania aquatica Wild Rice x

 

 

 

 

 

 

 

 

Figure 5-2: Maize starch with possible evidence for grinding

 

20$A1251: Middle and Bale Woodland

Site 208A 1251 is located along the Shiawassee River, within the Shiawassee
National Wildlife Refuge, near Saginaw, MI. Excavations have been undertaken by Mr.
Jeffrey Sommer, of the Historical Society of Saginaw County. An AMS date derived
from residue adhering to the inner rim of a Green Point Incised vessel produced a date
range of 40 BC-AD 120 (Sommer 2004). In addition to the dated Middle Woodland
component of208A1251, diagnostic artifacts provide evidence of Late Archaic, Early
Woodland, and Late Woodland occupations (Sommer 2002, 2003, 2005). The faunal

assemblage consists of bird, turtle, drum, sturgeon, catfish/bullhead, other fish, muskrat,

98

beaver, raccoon, deer, probable elk, bear, bobcat, fox, and other mammal (Sommer 2002,
2003b, 2005). Macrobotanicals have yet to be identified. Based on the artifact
assemblage consisting of ceramic sherds, bifaces and debitage, and fire-cracked rock,
Sommer interprets ZOSA1251 as a site where lithic tool manufacture, on-site butchery,
ceramic manufacture, and food preparation took place (Sommer 2002, 2003, 2005, 2007).

Phytolith and Starch Results

The results of the phytolith and starch study from the sherd examined from this
site produced no diagnostic phytolith or starch forms (Appendix B, Figure B 14). There
was also very little in the way of unidentifiable forms. It appears that whatever this
vessel may have been used for, its contents included little plant material. There are a
number of explanations for why little plant material was found in these vessel. These
include post-excavation cleaning of the artifact, the minimal use of plants cooked in this

vessel, and the use of the vessel for cooking meat and/or fish.

Schultz IZOSAZ): Middle and Late Woodland

The Schultz site is located within the city of Saginaw, at the confluence of the
Tittabawassee and Shiawassee rivers where they converge to form the Saginaw River.
The site’s proximity to upland, riverine, and wetland resources made it an ideal
settlement location. Systematic excavation of Schultz was undertaken by UMMA from
1959 to 1965. The most intensive period of excavation occurred from 1962-1965.
During this period UMMA crews identified a number of significant cultural features
including a stratified Early through Late Woodland midden, three burial mounds and a
circular stockade. hnportantly, throughout the site, trenches were excavated during the

UMMA excavations to aid in understanding the site‘s complicated stratigraphic record;

99

the result of a number of flooding and erosional episodes. These periodic flooding
episodes played a significant role in the duration and intensity of occupation at Schultz.
Excavations at Schultz were also undertaken in 1999 by Lovis on a smaller scale. The
goal of these excavations was the recovery of more fine—grained subsistence data through
the use of systematic flotation samples, a method not utilized during the original UMMA
excavations.

Occupational histogy:

During the Early Woodland, initial occupation of the area occurred after the
levee first forms. During site occupation the area around Green Point is initially swampy,
at c. 600BC, but becomes less so by 400BC. However, the general area around the site
does remain marshy (Fitting 1972). Floral remains for this time period consist of squash,
hickory, black walnut, butternut, and walnut as well as one aster seed (Allison 1972;
Egan I993; Ozker 1982). Faunal remains are more abundant than floral remains and
include deer, muskrat, beaver, bear, otter and a variety of fish (sturgeon, drum, walleye,
channel catfish, bowfin). Human occupation during the Early Woodland at Schultz is
extensive but of short duration and the site is abandoned after a flooding episode (Fitting
1972)

The early Middle Woodland period at Schultz demonstrates an increase in
flooding at Green Point around 2000 years ago (Speth 1972). Occupation of the site at
this time occurs along the levee and on the edge of a filled-in channel. Season of
occupation varied, but warmer weather species increase. A variety of floral remains have
been identified (Allison 1972; Egan 1993; Lovis et a1. 2001) and are listed in Table 5-8.

Terrestrial faunal remains during this time period are similar to that of the Early

100

Woodland but new species include elk, wolf, woodchuck, and mink. Fish species
exploited are also similar to the Early Woodland but more extensive fishing behaviors are
indicated by the presence of gar, perch, bass (large and smallmouth), sucker, and trout.
The circular stockade at Schultz is built towards the end of this occupation and artifact
densities and types suggest the presence of specialized activity areas (Fitting 1972).

During the late Middle Woodland, the rate of flooding which occurs at Schultz
decreases and, as a result, the overall area around the site becomes less swampy.
Occupation during this time period is on a stable levee which has a seasonal backswamp.
Towards the end of this occupation, environmental data provided by Brose's (1972)
mollusk analysis suggests a drier and possibly cooler climate. F aunal (aquatic and
terrestrial) and floral exploitation during the late Middle Woodland is similar to that of
the early Middle Woodland, however there is an increase in the potential reliance of
Schultz occupants on aquatic species compared to terrestrial fauna (Fitting 1972; Egan
1993; Lovis et al. 2001). An increase in reliance on aquatic species is an overall trend
from the Early through Late Woodland periods at Schultz.

Late Woodland occupation of Schultz declines in intensity and Fitting (1972:
264) characterizes these occupations as “. . .sporadic [and] seems to have taken place
many times between AD 600-AD 1300.” An increase in flooding episodes appears to
occur between AD 700 and AD 900 which caused the area to become marshy again. It is
perhaps due to this increased flooding that Late Woodland occupation at Schultz appears
to be so sporadic. However by AD 1100, the site is drier once again (Fitting 1972:264).
Overall, seasonality of these occupations is varied although the importance of aquatic

species, especially various types of fish, indicates warm weather.

101

A total of 42 sherds with residue from the Schultz (2OSA2) site were sampled
for this project. This reflects approximately a 10% sample of the number of sherds (total
count 421) with residues identified in the UMMA collection. The sherds sampled from
Schultz for this project primarily date to the Middle Woodland but also include some
transitional Middle to Late Woodland ceramics as well as one possible Early to Middle
Woodland sherd (Appendix B, Figures BIS-BS3).

Phflolith and Mh results:

The results of my phytolith and starch analyses confirm the macrobotanical
evidence, as well as add to it. Table 58 presents the macrobotanical data for the Middle
and Late Woodland periods at Schultz while Table 5-9 presents the phytolith and starch
data for the same time periods. Further evidence for the use of wild rice (Figure 5-3) and
aquatic tubers during the Middle Woodland is identified. Egan (Lovis et al. 2001) had
identified the presence of tubers at Schultz in the small-scale re-excavations undertaken
by Lovis in 1991 but was unable to identify them to the genus or species level. The
starch analysis performed here was able to identify a specific species of tuber (duck
potato) exploited prehistorically and also demonstrated evidence for the presence of
maize (Figure 5-4). The microbotanical data also supports the interpretation that the
Middle Woodland occupation at Schultz occurred during all economic seasons. The
plants identified in this analysis are all available in late summer to early fall and represent

resources that can be processed for long-term storage.

102

Table 5-8: Macrobotanical data for Schultz; x denotes presence

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Scientific Name Common Name Middle Woodland Late Woodland)
Amphicarpa bracteata Hog-peanut x j
Celtis occidentaIiS Hackberry x j
Chenopodium Sp. Chenopodium x x l
Gaylussacia baccata Huckleberry x l
Polygonum Sp. Knotweed x :l
Sambucus canadensis Elderberry x l.
Viburnum sp. Nannyberry x j
Iva annua Sumpweed x
Zizania aquatica Wild Rice x
Carya Sp. Hickory x
CoryIuS americana Hazelnut x x
Juglans Sp. Walnut x
Juglans cinerea Butternut x x
Juglans m'gra Black Walnut x x
QuercuS Sp. Oak (Acorn) x
Juglandaceae Walnut Family x x
C ucurbita pepo Squash x
Zea mays Maize x J
Table 5-9: Phytolith and Starch data for Schultz; x represents presence
Scientific Name Common Name Middle Woodland Late Woodland
Phytolith Starch Phytolith Starch
H Zea mayS Maize x x
Zizania aquatica Wild Rice x fl
Quercus sp. Oak (Acorn) X x a
LSagitaria latifolia I Duck Potato “i —— "x U“ —E—_— __.

 

 

 

 

 

 

 

 

 

 

103

Figure 5-3: Diagnostic wild rice sheath phytolith from Schultz

 

 

Figure 5-4: Maize starch, polarized and bright field views

 

The vast majority of radiocarbon dates obtained in this study come from the
Schultz site assemblage. This was primarily due to a number of factors including the
number of sherds where maize was identified in Middle Woodland wares, the presence of
sufficient residue on sherds for both AMS sampling and phytolith and starch analysis,

and the sherds from Schultz comprising most of the samples analyzed. The dates derived

104

from these samples are all early and in the range of Cal BC 300-AD 130. Table 5-10
provides specific dates and corresponding stylistic identification for sherds sampled for
AMS dating.

Table 5-10: Radiocarbon samples from Schultz

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Lab Number Stylistic Time Period md'm'b?“ ”‘2'“ (2 3'3“"
calrbratrorg)
Beta-261449 Middle Woodland, possible Early Cal BC 170_ AD 50
Woodland

Beta-261454 Green Point, Middle Woodland Cal BC l60-AD 60

Beta-261455 Middle Woodland Cal BC 90-AD 80
Possible Tittabawassee, early .

Beta-261456 Middle Woodland Cal BC 350-300, Cal BC 210-40

Beta-261459 Green Point, Middle Woodland Cal BC 190-AD 10

Beta-261460 Woodland Cal BC IOO-AD 120
Tittabawassee, early Middle

Beta-261463 Woodland Cal BC 170-AD 30

Beta-262038 early to late Middle Woodland Cal BC 30-AD 130

Beta-262039 late Middle Woodland Cal AD 10-210

Beta-262040 Green Point, Middle Woodland Cal BC 90-AD 80

 

 

Tobico Matrsh sites QOBY19gand 20BY210L Late Woofla_n_d

Of the sites identified within Tobico Marsh by James Payne in the 19908, sites
ZOBY192 and 20BY210 both had large enough sherds (i.e. those larger than a fingernail)
and sherds with residues. One sherd with residue was sampled from each site. The sherd
from 20BY210 (Appendix B, Figure B54) was temporally unidentifiable since the
exterior was too exfoliated. The sherd from ZOBY192 (not photographed due to fragility)
is thought to date post-1200AD due to its fine paste and the visibility of a fabric
impressed exterior when viewed under direct light.

Tobico Marsh itself is located on the southwestern side of Saginaw Bay in Bay

County, Michigan. The marsh contains a number of low parallel ridges, upon which a

105

number of sites within the marsh, including 2OBY192 and 20BY210, are located (Payne
1994).

Site 20BY210 (Tobico #21/Sharon's Scourge) was discovered during a shovel
test survey in 1990. Finds from the shovel test pits and subsequent excavations indicate
that 20BY210 probably represents a series of early and late Late Woodland occupations
(Payne 1994). Diagnostic ceramic decoration, in conjunction with diagnostic lithics
(including a Levanna-like biface), indicate Late Woodland occupation. Floral remains
recovered from 208Y210 include acorn fragments, a possible maize kernel, and an
unidentified tuber (Sommer 1994). Faunal remains found during excavation indicate fish
and mammal exploitation. Fish species, identified by Sommer (1994), include bowfin,
gar, perch, walleye, bullhead, drum, and sucker. Mammal remains include white-tailed
deer, dog or coyote, muskrat, fox, raccoon and squirrel. Migratory waterfowl included
duck, and a small number of amphibians, such as toad and turtle, were also identified
(Sommer 1994).

The variety in size of fish species found at ZOBY210 indicates the likely use of
netting or trapping. Many of the perch recovered are smaller in size which would make
spearing or line fishing inefficient capture techniques. The wide variety of species
present at the site also indicates a non-discriminatory fishing technique; netting or
trapping best represents this (Sommer 1994:58). The presence of walleye is a good
seasonal indicator of early spring since walleye is a spring spawning species and can be
found in shallower waters at this time of year.

Site 20BY192 (Tobico #8/Chris' Crossroads) is located on a ridge

approximately 35-40m wide and was first identified during survey in 1988 (Payne 1994).

106

Cultural material recovered from ZOBY192 indicates a late prehistoric occupation.
Diagnostic artifacts include Madison points and shell-tempered pottery. What Payne
(1994) identifies as an acceptable l4C date from one of the pit features at the site,
provides a date of AD 1 400-163 3, which corresponds to the time period represented by
the Madison points and shell-tempered pottery. A date of AD 1220—1395 was derived
from acorn fragments on the site but Payne rejects it for being too old (Payne 1994).
Little floral material was recovered other than the acorn already discussed. F aunal
remains include a variety of fish and mammal species. White-tailed deer, dog, elk, and
muskrat are among the mammal species identified. Fish species include catfish and perch
(Sommer 1994). Shallow stratigraphy and potential intrusion of historic activity into
prehistoric deposits provide some problems in associating certain of the faunal remains
with a time period. Both the dog and elk could be from historic activities rather than
prehistoric (Sommer 1994).

Both sites 20BY210 and 20BY192, likely represent resource extraction camps.
The large number of fish and aquatic mammals identified at both sites, along with
Sommer’s supposition of netting and/or trapping for fish capture strongly suggest
specialized collective resource extraction (a la Egan 1993). This is also supported by the
presence of tubers which are another collective resource and could easily be extracted
during fishing activities.

PhfioliMth Results

The results of the phytolith and starch analyses are listed in Table 5-12. These
results add to the few macrobotanical remains recovered from 20BY210 (Table 5-1 1).

No diagnostic phytoliths or starches were identified in the residue derived from the sherd

107

found at 20BY192. The phytolith and starch results confirm the macrobotanical data.
Maize is positively identified in conjunction with a specific species of aquatic tuber, duck
potato. The identification of duck potato supports the interpretation of 20BY2 10 as a
resource extraction camp. The presence of maize is probably the result of maize being

carried with individuals as a field food source (Raymond and DeBoer 2006).

Table 5-1 1: Macrobotanical remains from 20BY210; x denotes presence

 

 

 

 

Scientific Name Common Name Late Woodland
Zea mays Maize possible
Quercus sp. Oak (Acorn) x
NA Unidentified tuber x

 

 

 

 

 

Table 5-12: Phytolith and Starch Data for 20BY210; x denotes presence

 

 

 

 

 

, , Late Woodland
Scientific Name Common Name Phytolith Starch

Zea mayS Maize x

Sagitarz'a latifolia Duck Potato x

 

 

 

 

 

Simons 249 (ZOSA665): Late Woodlfiang

Site ZOSA665 was discovered by Mr. Donald Simons in the 1970s and surface
collected by him throughout the 19703 and 805 (Beld 1993). A formal excavation was
undertaken in May 1990 by Alma College as a field school under the direction of Dr.
Scott Beld. 20$A665 is located in an agricultural field on the Flint River floodplain in
Taymouth Township (Beld 1993). Based on two l4C dates from the Alma excavations, as
well as pottery and lithic diagnostics, the site dates to the Late Woodland period (Beld
1993). Radiocarbon dates from two features excavated during the Alma field school

resulted in dates of cal. AD 1323-1490 (2-sigma) and cal. AD 434-1 146 (2-sigma; note
108

that the large standard deviation is due to a small sample size). These dates represent
occupation during the entire Late Woodland.

Site seasonality for ZOSA665 is hypothesized to occur during the summer due to
the recovery of many fish bones and scales (Beld 1993). Other faunal and floral remains
are not discussed due to ongoing analysis at the time of Beld’s (1993) article. Features
identified during site excavation represent a probable storage pit and shallow hearth (Beld
1993). Lithics from the site include bifaces, debitage, cores, and scrapers which indicate
some tool manufacture and probable butchering activity associated with fish processing.
Site 208A665 could represent a short term resource extraction camp.

In his report on 20$A665, Beld (1993) includes Simons’ surface collected
materials. The sherd sampr for residue analysis (labeled vessel 92; Appendix B, Figure
B55) was part of Simons’ surface collection and is discussed by Beld (1993) as part of his
Moccasin Bluff exterior modified lip section. Beld (1993) describes this vessel as similar
to material found at Moccasin .Bluff due to the finger grooves on the exterior and
presence of one vertical and one horizontal row of circular punctates. When this sherd
was assigned a type for this study, with the aid of Dr. William Lovis, it was not assigned
a specific type but it was placed in the Late Woodland or later due to its being well fired
and having a hard fabric. This corresponds time-wise to Beld’s observation that it is
similar to material present at Moccasin Bluff.

Phytolith aid Sfirch Results

Phytolith and starch results are presented in Table 5-13. The data provided
from these analyses support a summer occupation but could also indicate early fall

through the presence of acorn. Maize can be harvested green during the summer.

109

However, mature maize is harvested in the fall so the presence of maize does not add

much to detail to occupation seasonality or firnction.

Table 5-13: Phytolith and Starch Data for 208A665; x denotes presence

 

 

 

 

 

 

 

. . Middle Woodland Late Woodland
sc'e'mf" Name comm" Name Phytolith Starch Phytolith Starch
Zea mqs Maize NA NA x x
Quercus sp. Oak (Acorg) NA NA x

 

 

 

 

 

Solm_sLZOSA57): Igate Woodlafl

The Solms site is primarily known from surface collections although salvage

excavations were undertaken by Fel Brunett in 1978. The report on file with the Office

of the State Archaeologist provides little detail on the excavation and materials recovered

other than a list of artifacts. The only macrobotanical data known for Solms is in the

recovery of squash seeds from a late Middle Woodland stratum. Floral data derived from

phytolith and starch analysis of Late Woodland material (see Table 5-14) denote the

presence of maize, aquatic tubers (Figure 5-5), nuts, and wild rice. A total of seven

sherds, from subsurface contexts, were sampled for residue analysis (Appendix B,

Figures B56-B62).

Table 5-14: Phytolith and Starch results for Solms; x denotes presence

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Scientific Name Common Name Middle Woodland Late Woodlanfl
Phytolithl Starch Phytolithl Starch
Zea mayS Maize NA F NA 1 x l x 1‘
Jzania aquatica Wild Rice NA NA x 1
Quercus Sp. Oak (Acorn) NA NA 1 x
:Nuphar Iutea Yellow Pond Lily 1 NA NA l x

 

 

 

 

 

110

Figure 5—5: Yellow Pond Lily

Sgdelmever (ZOSA195): Late WoodLand

Stadelmeyer is located in the southwestern portion of Saginaw, M1, on the north
bank of the Tittabawassee River approximately 5.8km from the confluence of the
Tittabawassee and Shiawassee Rivers (Bigony 1970). The site is located on a sandy
ridge at an elevation of approximately l81masl. The prehistoric component of the
Stadelmeyer site was initially found in the 1960s by amateur archaeologists associated
with the Saginaw Valley Chapter of the Michigan Archaeological Society. The Saginaw
Drainage Chapter undertook a number of excavations until 1967 when the UMMA
conducted their own fieldwork (Bigony 1970).

Based on the pottery recovered from the UMMA excavations, as well as the
inclusion of the collections of two local amateur archaeologists, Messrs. Arthur Graves
and Cleve Pomranky, the site was determined to date to two periods of the Late
Woodland (Bigony 1970). The earlier Late Woodland occupation dates to AD'800-1100

based on the presence of Riviere au Vase ware. The later phase dates to AD 1200-1400

111

due to Upper Mississippian influence pottery. These Upper Mississippian influenced
sherds are well-fired and use shell or grit temper (Bigony 1970).

Season of occupation for the Late Woodland component of Stadelmeyer seems
to represent a late fall to early spring period of habitation. Fish remains, pike, perch,
sturgeon and bowfin, are indicative of those caught in the early spring, based on their
spawning season, while walnut and butternut hull remains are suggestive of fall. Other
identifiable macrobotanical remains (see Table 5-15) were not recovered (Bigony 1970).
Features at the site support this interpretation through the identification of hearths and
storage pits, which would have been necessary for a winter occupation, and what Bi gony
calls a living floor. The living floor is described as a shallow lens of darker soil than the
surrounding matrix which contained pottery, bone, clay daub, and charcoal (Bigony
1970:121).

A total of 14 sherds with residues were identified in the Stadelmeyer collection
at the UMMA. Of these, six sherds (43%) were sampled for this project. From these six
sherds (Appendix B, Figures B63-B68), three dated to the later Late Woodland period
(post-1200AD), one was a generic Woodland ceramic, and two were a stylistic toss-up
between the Middle and Late Woodland periods. The latter had paste which looked
Middle Woodland but the rims had characteristics more consistent with the Late
Woodland.

Hiltolithpnd 8% results

The data derived from the phytolith and starch studies for Stadelmeyer
significantly increased the number of known plant taxa. Macrobotanical data had only

identified acorn, walnut and butternut as plant species present (see Table 5-15). Plant

112

taxa identified through the residue analysis performed in this study (see Table 5-16) now

include maize and aquatic tubers. This new data also supports the interpretation of a fall

to early spring occupation at Stadelmeyer.

Table 5-15: Macrobotanical Data for Stadelmeyer; x denotes presence

 

Scientific Name

 

 

 

 

 

 

 

 

Common Name Middle Woodland Late Woodland
Jrgglans sp. Walnut/Buttemut NA x
Jgglans cinerea Butternut NA x
Juglans nggra Walnut NA x
Quercuspp. Oak (Acorn) NA x

 

 

 

Table 5-16: Phytolith and Starch data for Stadelmeyer; x denotes presence

 

 

 

 

 

 

 

 

 

 

Scientific Name Common Name Middle Woodland 1 Late Woodland
Phytolithf Starch [Phytolithl Starch—l
Zea mays Maize ) NA NA x L x l
Quercus Sp. Oak (Acorn) 1 NA 1 NA ) x W
Sagitaria Iatifolia Duck Potato i NA [ NA l x 1
Nuphar [urea Yellow Pond Lily j NA T NA F x W

 

Foster (20$A74): Lane Woodlgng

Formal excavations of Foster were undertaken by the WMA in the summer of

1967. Foster is located in Taymouth Township of Saginaw County, M1 on the

northeastern side of the Flint River (Bigony 1970). Foster is dated to the Late Woodland

period on the basis of diagnostic ceramic and lithic artifacts. The small ceramic

assemblage present is described as, “.. .suggestive of Mackinac ware which has been

dated to ca. AD7 00-1 100.. .” while the presence of shell-tempered pottery is indicative of

a post-AD1300 occupation (Bigony 19702208). The presence of stemmed and small

triangular points at Foster also speaks to a Late Woodland occupation since these are

113

typical of the period for the Saginaw Drainage. Bigony (1970) determines site
seasonality to be fall/winter on the basis of walnut and butternut shell and the
preponderance of mammal bone in the faunal assemblage. Fish bone, of drum and
catfish, in the faunal assemblage do not narrow down seasonality since they are available
year round. The identification of hearths and storage pits would also be consistent with a
fall/winter occupation and Foster is likely a winter camp.

Three of the four identified sherds (75%) with residue were sampled from the
UMMA collections of Foster. Two of these sherds date to the later Late Woodland while
one was unidentifiable due to its exfoliated exterior (Appendix B, Figures B69-B71).

Phytolithind Starch Results

Data from the phytolith and starch analyses are presented in Table 5-17.
Analysis of residues provides more detailed floral information than originally reported.
Both maize and yellow pond lily, add to the subsistence data for Foster. Bigony (1970)
proposes that Foster is a fall/winter occupation based on the presence of nuts such as
walnut and butternut. Given the identification of only maize starch, this likely indicates
use of a dried form of maize. Tubers such as yellow pond lily are available year round
but are best harvested in the late summer or early fall. Both maize and yellow pond lily
are storable resources and support Bigony’s (1970) suggestion of Foster as fall/winter

occupation.

Table 5-17: Phytolith and Starch Data for Foster; x denotes presence

 

 

 

 

 

 

 

 

, , Middle Woodland 1 Late Woodland T

scum“ Name “mm" Name Phytolith Starch Phytolith \ Starch
Zea mays Maize NA NA . l x

Ntflvhar Iutea Yellow Pond Lily NA NA 1 l x j

 

 

 

 

 

 

 

114

Cluniet20$A7221z Late Woodland

The Clunie site is currently undergoing excavation by Mr. Jeffrey Sommer of
the Historical Society of Saginaw County at the Castle Museum. There have been three
field seasons of excavation thus far, all of which have yielded a wealth of information.
Clunie is located on the banks of the Tittabawassee River in the Shiawassee National
Wildlife Refuge in Saginaw County, MI (Sommer 2005, 2006, 2007). Excavations take
place every summer under the direction of Sommer with volunteer archaeologists from
the local community.

Ceramic material at the site indicates the primary occupation as Late Prehistoric
(AD 1400-1650). This time period assignment is also supported by four radiocarbon
dates derived from charcoal samples at Clunie (Sommer 2005, 2006, 2007). However
other ceramic material also indicates an early Late Woodland component as well. One of
these Late Woodland sherds (Appendix B, Figure B72) was sampled for this study and
radiocarbon dated. The date derived from residue on this ceramic, however, dates the
sherd to the Late Prehistoric with a date of Cal AD 1440-1640 (2-sigma).

Sommer has encountered numerous prehistoric features at Clunie and has
utilized intensive flotation sampling as part of his excavation strategy (Sommer personal
communication). Analysis of flotation material in conjunction with excavation-
encountered floral and faunal remains has yielded a variety of species. Floral remains
include acorn (both nut and meat remains) as well as maize cob and individual kernels
(see Table 5-18). Faunal remains from the site are comprised of mussel, fish bone and
scales (including sturgeon and walleye), elk and deer, snapping turtle, beaver, muskrat,

porcupine, and various species of bird (Sommer 2005, 2006, 2007). The faunal remains

115

indicate extensive use of the available surrounding habitat. Walleye and sturgeon are
both spring spawning fish while beaver is most often hunted in the winter. The presence
of maize is indicative of late summer while acorn is harvested in the fall. The
combination of this seasonality data indicates Clunie may well have been occupied either
year round or nearly year round (Sommer 2005, 2006, 2007). Deer and elk are available
year round while acorn and maize are very commonly stored food resources, which

would have made year round occupation feasible.

Table 5-18: Macrobotanical Data for Clunie; x denotes presence

 

 

 

 

 

 

 

 

 

Scientific Name Common Name Late Woodland Late Prehistoric
Zea mgs Maize x
JOLuer-cus gp. Oak (Acorn) x
Phytolithgnd Stgrch Results

 

Phytolith and starch results for Clunie are given in Table 5-19. Residue
analysis did not add any new subsistence data but does confirm the presence of maize in
the Late Prehistoric occupation of Clunie. This new data is consistent with the

interpretation that Clunie was likely occupied during all economic seasons.

Table 5-19: Phytolith and Starch Data for Clunie; x denotes presence

 

Late Woodland Late Prehistoric

 

Selentlfic Name Common Name Phytolith Starch Phytolith Starch

 

 

 

 

 

 

 

Zea mqs Maize x

 

 

Summag of New Phytolith and Starch Subsistence Data

The phytolith and starch data presented in this chapter confirms previously

known subsistence behaviors, as well as providing significant new data that extends and

116

adds to our knowledge of subsistence for the Middle and Late Woodland period of the
Saginaw Drainage. A summary table of phytolith and starch data with the associated
radiocarbon dates for split residue samples is found in Table 5-20; the detailed results for
each residue analyzed can be found in Appendix C, and the full results for the AMS dates
can be found in Appendix D. Phytolith and starch data on Middle Woodland wild rice
and aquatic tuber use provides more detail than the existing macrobotanical data.
Evidence for the earlier use of maize is demonstrated through identification of maize
starch in conjunction with a number of AMS dates which place this use of maize
significantly earlier, by approximately 600 years, than a previous early date of AD 400
derived from a maize kernel at ZOSA1043 (Parker 1996). What this data means,

especially the evidence for such earlier use of maize, is discussed in the next chapter.

117

Table 5-20: Radiocarbon dates with plants identified from split residues

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Lab Radiocarbon Date Site Name and , ,
Number (2 sigma calibration) Number Plant Spec” Identified
Beta- . .
261462 Cal AD 1440-1640 Clunie (208A722) Maize
216373561 Cal AD 880-1020 Solms (208A57) Maize, Wild Rice
Beta- . . .
261452 Cal BC 50-AD 90 208A1276 Maize, Wild Rice
Beta-
261457 Cal AD 10-140 208A1276 Acorn
Beta- Possible sedge family
261449 Cal BC l70—AD 50 Schultz (2OSA2) species
2123; Cal BC 160-AD 60 Schultz (ZOSAZ) Possible wild rice
Beta- .
261455 Cal BC 90-AD 80 Schultz (208A2) Maize
Beta- Cal BC 350-300; Cal .
261456 BC 21040 Schultz (ZOSAZ) Maize
Beta- . .
261459 Cal BC 190-AD 10 Schultz (ZOSAZ) Wlld Rice
Beta- Possible maize starch
Cal BC 100-AD 120 Schultz (2OSA2) determined to be generic
261460
form
223:3 Cal BC 170-AD 30 Schultz (ZOSAZ) Maize, Wild Rice
21:63:33.8 Cal BC 30-AD 130 Schultz (208A2) Maize
Beta- Possible maize starch
Cal AD 10-210 Schultz (20$A2) determined to be generic
262039
form
Beta' c lBC 90 AD 80 s h ltz (203/312) M °
262040 a ' c H 3.126

 

 

 

 

118

 

CHAPTER 6

Saginaw Drainage Connections to the Great Lakes Region

Introduction

Results from the archaeological analysis indicate the earlier use of maize in the
Saginaw Drainage as well as confirming the accepted hypothesis of maize utilization by
c. AD 1000. The new data provided by the starch and phytolith analyses present some
interesting questions. The earlier use of maize during the Middle Woodland in the
Saginaw Drainage coincides with the expression of Hopewell-related traits both in
Michigan and throughout the Midwest. Is the earlier use of maize somehow connected to
the spread of Hopewell traits? What does the presence of maize starch, indicative of a
form of dried maize, prior to the presence of maize phytoliths, i.e. evidence for the use of
green maize, potentially mean? According to the behavioral table created from the
experimental residues in Chapter 4, the presence of maize starch is most likely indicative
of dried whole kernel or maize flour. The lack of maize phytoliths identified in residues
prior to the Late Woodland period indicates that green maize was not being utilized
before this point in time. If, as discussed in Chapter 4, part of the potential ‘invisibility’
of early maize use is its use in a green form, how can we explain the use of a dried form
prior to a green form of maize? A summary of those sites in the Saginaw Drainage that
were detected to have early maize is needed.

The identification of the presence of maize during the Middle Woodland (200
BC-AD 500) for sites in this study is limited to the following: 208A1276, Liberty Bridge
(20BY77), Kantzler (208Y30), and Schultz (ZOSAZ) (Figure 6-1). It is interesting that

all of these sites are thought to have either year-round economic activity (Kantzler and

119

Figure 64: Sites with maize dating to the Middle Woodland period (figure courtesy of
Jon Carroll)

 

N I
O 5 10 20
w E .2- Kilometers

s “ $1

. , Legend

I .
. ,3 —- Rivers
Tittabawasseee . Bay ‘2‘. w: Counties

River

 

 

 

 

 

 

 

 

Sites: l=208A1276; 2=Schultz (208A2); =Kantzler (ZOBY30); 4=Liberty Bridge
(20BY77)
120

Schultz) or represent smaller short term seasonal occupations, either warm weather
(spring/summer at Liberty Bridge) or cold weather (fall to early spring at 203A1276)
occupations. The occupations at Kantzler and Schultz also have significant evidence for
Hopewell ceremonial affiliations while Liberty Bridge and 208A1276 have more limited
manifestations of Hopewell material culture such as slate pendants (Liberty Bridge) and
engraved turtle carapace bowls (2OSA1276). The following section discusses potential

mechanisms for how maize may have been incorporated into these earlier dated residues.
Explanations for Maize in Residues

Contamination

One explanation for the identification of maize starch in residues is through
contamination. Many of the collections utilized in this study have been curated in
repositories for up to 40 years. Corn starch (i.e. maize starch) is a very common additive
to packing materials and laboratory gloves. There is the possibility for the transfer of
starch from these types of sources to artifact surfaces if collections are being rehoused,
reanalyzed, or repackaged.

While a possibility, I do not think this is the source of corn starch found in my
samples. Many samples were curated at the same institution. One such example are the
samples from the Schultz site which were all housed in the Great Lakes Range at the
University of Michigan Museum of Anthropology. Yet despite this common factor, maize
starch was not identified in all residues derived from samples at Schultz. It would be
expected that all, or at least most samples, would have maize starch present if i
contamination fi'om packaging materials had been a factor.

The other possibility is the transfer of corn starch from powdered gloves. Since
121

1 was aware of the potential for sample contamination with the use of powdered gloves,
these types of gloves were not used when performing my analyses. 1f powdered gloves
were used by museum staff at the institutions where the collections used in this study
were housed, I do not know. However, if contamination from this source was a factor, 1
would once again expect maize starch to be present on all or most samples analyzed. To

reiterate, this was not the case. Overall, I think contamination is an unlikely explanation.

Low-level Cultivfiion of Maize in the Saginaw Drainage

The explanation I expected (see Chapter 1) is that the early presence of maize
would be due to its cultivation as a supplement to both extant Eastern Agricultural
Complex (EAC) crops and wild plant foods already in regional economic use. However,
I do not think planting of maize can be explained only through the presence of maize
starch. The presence of only maize starch in these early residues does not support the
hypothesis that initial maize utilization in its green form accounts for the invisibility of
early maize use in terms of macrobotanical remains. The experimental data generated in
this study also reveals that the best indicator of green maize utilization is in the
identification of diagnostic cob phytoliths and not through the presence of starch. This
finding is supported by the recovery of maize cob phytoliths from other residue studies
(c.f Hart et al. 2003; Thompson et al. 2004) which indicate significantly earlier maize
use. The early evidence for maize in the Saginaw Drainage derived from this study only
identifies maize in its dried form (i.e. maize starch).

As determined from the experimental residues, maize starch is an overwhelming
indicator for a dried form of maize. This could mean that at this earlier point in time,

maize was under cultivation in the Saginaw Drainage. However, if maize were being

122

cultivated and dried for storage during the Middle Woodland, one would also expect the
recovery of macrobotanicals or cob fragments as well as tools associated with cultivation
such as hoes. Recovery of maize macrobotanicals has not been the case despite the
widespread and systematic use of flotation in Saginaw Drainage research for the past
thirty years; evidence for tools associated with cultivation is also lacking.

Additional evidence which argues against the early cultivation of maize in the
Saginaw Drainage is the increase in flooding documented during the early Middle
Woodland (Monaghan and Lovis 2005; Speth 1972). This increase in flooding is a result
of higher lake levels in Lake Huron which would have had a dramatic effect on river flow
and would have inundated the flood plain (Monaghan and Lovis 2005). Additionally,
GIS modeling currently undertaken by Cheruvelil (personal communication 2010)
corroborates extensive flooding during the early Middle Woodland due to the higher lake
levels in Lake Huron.

Evidence for wetter conditions is also demonstrated in environmental
reconstruction via pollen analysis undertaken by Hupy and Yansa (2009) in the ecotone
transition between deciduous and coniferous forest. Their analysis demonstrates an
increase in mesic tree species such as beech, maple, elm, and basswood in the deciduous
ecotone transition at c. 250 AD (Hupy and Yansa 2009:218). The residues with maize
AMS dated for this study all date to the early part of the Middle Woodland (BC 200-AD
300) and the potential for maize cultivation would have been affected by these wetter
conditions.

While flood plain cultivation was common during the Late Woodland and Late

Prehistoric periods, it would probably not have been viable during the early Middle

123

Woodland. Increased early Middle Woodland flooding episodes would most likely have
inhibited sufficient drainage of the flood plains, especially due to the already low-lying
topography of the area. A reexamination of Figure 6-1 demonstrates that the four sites
with early maize are at most 200m asl. All four sites would have been prone to flooding
during an episode of increased flooding due to higher lake levels in Lake Huron. The
area these sites are located in were prone to floods throughout the historic period and up
to today (Krakker 1983). It is more than likely that these areas near site locations would
have remained marshy and too wet to allow for successful maize cultivation (c.f. Lovis
and Bogdan-Lovis 2004 on the effects of copious precipitation during the planting season
and Chapter 2 of this work on the requirements for successful maize cultivation).

The use of ridged fields in gardens could have aided in the successful
cultivation of maize during flood periods (Gallagher et al. 1985; Gartner 1999; Sasso
2003), and while there are garden beds identified in both Saginaw and Shiawassee
Counties (Hinsdale 1931), their identification is farther downstream on the Shiawassee
River. A review of Figures 5-1 and 6-1 (maps of site locations) demonstrates that the
sites used in this study, are all located farther upstream along the Saginaw River and at
the convergence of the Shiawassee, Tittabawassee, and Cass Rivers. According to
Hinsdale’s (1931) maps, garden beds have not been identified in this area. Ultimately the
environmental conditions for this part of the Middle Woodland period were not

conducive to maize cultivation, even for low-levels of production.

mm: of Maize in Dried Form(§)
Another explanation for the presence of maize is its exchange between

populations. It is well known that hunter-gatherer populations participate in gift

124

exchange to reinforce social ties (Cashdan 1985; Hegrnon 1991; Holman and Kingsley
1996; Kelly 1995; O’Shea 1981 and many others). It would be reasonable to assume that
this type of behavior also applies to low-level food producers. Utilizing a mixed-
subsistence economy still requires the maintenance of social ties to act as a buffer if one
resource group (wild or cultivated) should fail. This would still require maintenance of
social ties via gifl exchange; maize, as an exotic food source, could be a potential type of
gift. The gift aspect of exchange does not have to be included in this scenario as well.
Sharing of food resources is a daily occurrence among hunter-gatherers and is a behavior
which could be extrapolated to low—level food production (Keeley 1995; Kelly 1995;
Knaufi 2010; Lee 2003).

A study undertaken by Fie (2006) to determine which type of exchange best
characterizes Middle Woodland (i.e. HOpewell) exchange networks also supports the idea
of subsistence based exchange. Fie’s (2006) study looks at the exchange of everyday,
non-exotic items in the Lower Illinois River Valley. Fie’s analysis of everyday item
exchange during the Hopewell is important since so much emphasis is placed upon the
exchange of exotic items during this period. F ie tests four models using ceramic
composition and stylistic data, to determine which best characterize Middle Woodland
interaction. For the Lower Illinois Valley, F ie finds that the most congruent model is
based on socially motivated exchange for subsistence-maintenance materials (F ie
2006:445). Exchange of lithic raw material and food stuffs were important aspects of
Middle Woodland exchange; Hopewell ceremonial items were not the sole focus. A
subsistence-maintenance materials model supports the hypothesis of maize being utilized

in subsistence exchange form. In fact, maize identified at Schultz, which has a

125

significant Hopewell ceremonial component associated with a stockade, does not occur in
these ceremonial contexts. Maize identified in residues from Schultz occurs in areas
outside the stockade and are not affiliated with Hopewell-related ritual. 1f maize is being
utilized as an exchange item, this now brings up the question of transport.

Maize in its dried form (whole kernel or flour) is easily transportable. It would
not be unduly burdensome to transport maize longer distances in the form of kernels or
flour. Transportation prehistorically would have most likely occurred in skin bags and
baskets or birch bark containers; transportation of maize in ceramic jars would have been
cumbersome and the ceramics prone to breakage, especially if transport was occurring
over long distances. However, this is not to assume that transport of materials via ceramic
jars never occurred.

If we suppose that maize is being exchanged in to the Saginaw Drainage, the
next question to ask is, where is it coming from? The assumption made here is that any
exchange of maize to the Saginaw Drainage is through down-the-line exchange and not
direct exchange or trade between maize producing populations and non-maize cultivators.
Based on artifact typologies and lithic raw material sources, the likely areas of interaction
between the Saginaw Drainage and Great Lakes/Midwest region, either individually or

collectively, are with Illinois, Indiana, Ohio, and southern Ontario.

Evidence for Middle Woodland Maize

Southern Ontario

 

Early evidence for maize in southern Ontario dates to 0. AD 230/260-670/610
(Z-Sigma; Crawford and Smith 2003) from the Grand Banks site located in eastern

southern Ontario. The Middle Woodland component of southern Ontario is characterized

126

by the Couture (western southern Ontario), Saugeen (central western Ontario), and Point
Peninsula (eastern Ontario) Complexes. For the purposes of this discussion, the most
relevant, geographically speaking, Middle Woodland Ontario cultural manifestations are
the Couture and Saugeen complexes. The Grand Banks site is a Middle to Late
Woodland site and is best characterized by Point Peninsula and subsequently Princess
Point material culture. The site is located on the isthmus between Lakes Ontario and Erie
and is approximately 257 km from the Finger Lakes region where Hart et al. (2003, 2004)
identified early maize in ceramic residues. The approximate distance to the Saginaw
Drainage from Grand Banks would be around 360 km. Early maize has yet to be

identified at Couture or Saugeen complex sites.

QhLo

Extensive paleoethnobotanical studies have been conducted on Middle and Late
Woodland Ohio sites. Out of those studies, early maize has been identified at the Harness
Mound site (Smith 1992). It was hypothesized early on (Braun 1987; Brose and Greber
[eds] 1979; Morgan 1952) that Hopewell manifestation (and even possibly the Early
Woodland Adena cultural complex) in Ohio was not possible without maize agriculture
(Hall 1976). As Eastern Agricultural Complex (EAC) cultigens became recognized as an
independent agricultural manifestation within ENA, a shift from maize agriculture as the
basis of Hopewell society to cultivation of EAC crops as the basis for Hopewell
subsistence occurred (Dancey and Pacheco 1997; Wymer 1987, 1993, 1994; Wymer and
Abrams 2003).

In part this shift in hypothesizing reliance on maize to EAC cultigens was due to

the paucity of maize macrobotanicals identified through the systematic use of flotation on

127

many Ohio Hopewell projects (Wymer 1993, 1994, 1996, 1997). Stable isotope studies
of human bone have also not identified the presence of maize in Ohio Hopewell diets
(Redmond 2007; Tankersley and Tench 2009; Yerkes 2005). Yerkes (2006) also notes a
lack of dental caries, which would be associated with a high carbohydrate, agricultural
diet, on Ohio Hopewell skeletal remains. The extant paleoethnobotanical and skeletal
data in Ohio overwhelmingly points to the use of EAC cultigens used in combination
with wild food resources (both faunal and floral) rather than reliance on maize

horticulture during the Middle Woodland (Cowan 2006; Wymer 1994; Yerkes 2006).

[Liliane

The area of Indiana archaeology pertinent to Michigan is located in northwest
Indiana. The cultural manifestation in this area is known as the Goodall Tradition and is
the local Hopewell variant (Brashler et al. 1999, 2006; Garland et al. 2006; Quimby 1941 ,
1943, 1944; Schurr 2009). The Goodall Tradition is located primarily in northwestern
Indiana as well as extending northwards into southwestern Michigan and, relative to
other Hopewell manifestations, is poorly understood. Studies by Schurr and Mangold
(Ketchum et al. 2009; Mangold and Schurr 2006) have attempted to identify Goodall
Tradition subsistence practices; however, very little floral and faunal data have been
identified. Chenopod and hazelnut was identified at the Bellinger (12SJ6) site while
bottle gourd was found at the Goodall (12LE9) site. Site locations on “dunal islands”
(Ketchum et al. 2009) and in the uplands of the Kankakee River seem to indicate a mixed
subsistence system of aquatic and upland terrestrial resources. Currently there is little
evidence to suggest plant cultivation, whether EAC or maize, for the Goodall Tradition.

A survey of the published literature also failed to identify any early evidence for maize

128

within other regions of Indiana.

Hum

Evidence for early maize within Illinois is present in both macrobotanical and
stable isotope data Maize macrobotanicals, two of which were AMS dated, were
recovered from the Holding site (11MS118), located in the American Bottom in Illinois.
These AMS dates were derived from a cob fragment and kernel recovered from two
different feature contexts at Holding (Fritz 1993). The respective dates (kernel and cob)
are 2017i50 BP (67 BC) and 2077i70 BP (127 BC) (Fritz 1993; Riley et al. 1994). For
the Illinois archaeological chronology, this is squarely in the Middle Woodland (200 BC-
AD 300) period. Other occurrences of Middle Woodland maize macrobotanicals in
Illinois have also been recovered from Peisker, Smiling Dan, IISA86, and Console (Asch
and Asch 1985;.lefferies and Butler 1982; Struever and Vrckery 1973). The maize
remains from these latter contexts are all dated via artifact association and does not
represent an absolute date.

Recent stable isotope studies (Rose 2008) of Middle Woodland to Mississippian
skeletal populations in west-central Illinois have identified maize consumption at some
early Late Woodland (AD 400) sites. Maize was identified in skeletal material at the
Knight and Joe Gay burial mounds. The evidence from the Joe Gay site is especially
compelling since the Late Woodland component of this site is presumed to be continuous
from the Middle Woodland component on the basis of ceramic style continuity (Rose
2008:424).

The Illinois sites with Middle Woodland maize present are all located between

the Illinois and Mississippi River Valleys in west-central Illinois. These are also sites

129

with substantial Hopewell components. Ceramics at Smiling Dan, for examme, are
characteristic of a range of Middle Woodland forms including Havana, Pike, Baehr, and
Hopewell vessels. Other evidence for Hopewell material culture is in the presence of
blades, obsidian, cut mica, and slate gorgets. Importantly, Smiling Dan also has evidence
for tools associated with plant cultivation in the form of manos (n=l 34), metates (n=2),
and hoes (n=24). Cultivation of EAC plants is known to occur at Smiling Dan; the
presence of Middle Woodland maize, however, also indicates the incorporation of maize

into the extant subsistence system.

Evidence for Interregional Influence in Michigan

If one accepts the premise that the early maize identified in Michigan is a result
of down-the-line exchange, there are only two geographic regions which could be the
likely sources for the exchange of maize based on the evidence presented above. These
areas are southern Ontario and Illinois. There is evidence from both regions for contact
between populations in both the presence of lithic raw material and projectile point styles.
From the southern Ontario/Northeast region Onondaga chert is a common exotic lithic
raw material found in the Saginaw Drainage. Other Ontario chert sources found to
varying degrees in Michigan also include Kettle Point and Collingwood. Meadowood
points are one of the most common projectile point types for the Late Archaic/Early
Woodland period in Michigan and these have a distribution throughout the Northeast and
into Ontario and Michigan. Other projectile point examples following a Northeast
distribution into Michigan include Genesee, Snook Kill, and Larnoka (Justice 1987).

Examples of Archaic period bifaces with Illinois affinities found in Michigan

are Turkey Tails and a variety of stemmed points falling into the Table Rock Cluster

130

(Justice 1987). In regards to lithic raw material originating from Illinois, Burlington
chert would be the most likely source. However there is very little Burlington chert
found in the Saginaw Drainage of Michigan. Sommer (2004) reports one flake of
Burlington from 2OSA1276 and Crumley describes a high quality non-local white chert at
Kantzler but does not identify it. Burlington is characterized as high quality and is white
in color; the chert described at Kantzler could be Burlington based on the description but
visual confirmation is required.

Projectile point continuity with Illinois continues into the Early and Middle
Woodland. The characteristic Hopewell Snyders point originates in Illinois (Justice
1987) and has a Michigan variant known as Norton Comer Notched, primarily found in
southwestern Michigan. Snyders points found outside of Illinois are predominately in
Indiana and Michigan but are less frequent in Ohio (Justice 1987). Interestingly, Snyders
points identified in southwestern Michigan appear to be related to the extension of the
Goodall Tradition into Michigan and are often made on Burlington or Wyandotte (an
Indiana lithic material) cherts (Brashler et al. 2006; Garland and DesJardins 2006).
Brashler’s excavations (1996-97, 1999) at the Prison Farm site near Grand Rapids, MI
identified Burlington chert as part of the lithic assemblage and Garland and DesJardins
(2006) characterize Burlington as an important exotic lithic in the St. Joseph and Portage
River drainages. To a lesser extent, Snyders points in Michigan are also found made on
Onondaga, Flint Ridge and Upper Mercer cherts which does at suggest ties of some kind
in terms of lithic raw material exchange to Ohio and southern Ontario. The southwestern
Michigan variant of Snyders, Norton Comer Notched points, are commonly made on

Bayport chert, which is ‘exotic’ to southwestern Michigan, but local to the Saginaw River

131

drainage (Brashler et a1. 2006; Garland and DesJardins 2006).

Based on the lithic information, both southern Ontario and Illinois seem to be
equally likely areas of interaction. However the ceramic typology of the Saginaw
Drainage provides a different set of information. Ceramics first appear in Michigan
during the Early Woodland period. Early Woodland Saginaw Drainage ceramics,
particularly Schultz Thick and Shiawassee Ware, are Illinois variants of Marion Thick
and Dane Incised respectively (Brashler 1973; Fischer 1972; Fitting 1970; Ozker 1982).
These Early Woodland Schultz Thick ceramics have also been identified at southwestern
Michigan sites such as Spoonville, Norton Mound, and Moccasin Bluff (Brashler 1973).

Evidence for ceramic affinities with Illinois continues into the Middle
Woodland (and Hopewell manifestations) of both southwestern Michigan and the
Saginaw Drainage. Tittabawassee and Green Point Wares are characterized by nodes,
punctates, dentate stamping, and rocker stamping (Brashler 1973). These two wares
correspond to the expression of Havana and Hopewell wares for the Saginaw Drainage.
Tittabawassee Ware has afi'mities with Havana ceramics from Illinois (Brashler 1973;
Fischer 1972). Research occurring on Hopewell manifestations in southwestern
Michigan has also identified Illinois Havana wares (Brashler et al. 2006; Kingsley et a1.
1999; Mangold and Schurr 2006). Illinois Havana wares have been identified at the
Moccasin Bluff, Hart, Norton Mounds, Jancarich, Sumnerville, Toft Lake, and Brooks
Mound sites (Brashler et al. 2006; Garland and DesJardins 2006; Kingsley et al. 1999).

In regards to southern Ontario ceramic styles, there is little evidence for
interaction between Saginaw Drainage and southern Ontario populations during the Early

and Middle Woodland periods (Brashler 1978; Fischer 1972). In general, there is very

132

little evidence for any sort of Hopewell cultural manifestation in either Saugeen or
Couture Middle Woodland complexes. The evidence for any kind of Hopewell influence
in southern Ontario is seen in a couple of caches of Hopewell bifaces and prismatic
blades made on Flint Ridge chert found at Couture Complex sites in southwestern
Ontario (Spence and Fox 1986; Spence et a1. 1990; Crawford and Smith 2003). Other
evidence for Hepewell influence that has been found is in ceramics with dentate and
rocker stamping which have been characterized as influenced by Ohio Hopewell and
some Hopewell burial goods present in Saugeen complex burials (Deller et al. 1986;
Spence et al. 1990; Wright 1967).

Apparent interaction for southern Ontario at this point in time is most likely
with points in Ohio instead of the Saginaw Drainage. Interaction between the Saginaw
Drainage and southern Ontario, as manifested via ceramic styles, does not appear until
the Late Woodland period which appears to coincide with a shift in population density
and focus (Krakker 1983). This particular phenomenon will be addressed in the

concluding chapter.

Summary

Given the evidence presented above, the best candidate for where this maize
may have originated is currently in Illinois. This is primarily based on the number of
early occurrences (i.e. Middle Woodland) for maize that have been identified in Illinois as
well as the evidence for cultural influence in the Saginaw River drainage and throughout
southwest Michigan. Cultural manifestations in the Saginaw Drainage demonstrate
continuous afiinities with Illinois-based material culture up through the Middle Woodland

period and HOpewell-related manifestation of the Saginaw Drainage.

133

In fact, it is probably through interaction between the occupants of the Saginaw
Drainage and populations in southwestern Michigan that Illinois stylistic and cultural
influence is manifested. Saginaw Drainage Hopewell-related manifestations have been
characterized as ‘doing Hopewell out of context’ (Fitting 1975; Kingsley 1999:152).
There is good evidence to suggest that certain Hopewell characteristics were adopted by a
local population through diffusion. H0pewell mortuary treatments for the Saginaw
Drainage are characterized as truncated and there is a higher proportion of decorated
ceramics in the ceramic assemblage for Saginaw Hopewell than is seen in other Hopewell
manifestations (Kingsley et al. 1999:] 52). The evidence for diffusion and/or
acculturation of Hopewell in the Saginaw Drainage is from southwestern Michigan,
which has evidence for connections to Illinois Hopewell.

There is ample evidence for interaction between southwestern Michigan and
Indiana with the Goodall Tradition. The Goodall Tradition has direct connections and
ties to Illinois Hopewell cultural manifestations (Garland and DesJardins 2006; Kingsley
et al. 1999; Mangold and Schurr 2006). Goodall appears to be directly related to Illinois
Hopewell in decorative ceramic motifs such as dentate stamping and rocker decoration.
Importantly, Burlington chert is also found on Goodall Tradition sites and in larger
percentages on these sites. Besides lithic raw material, bifacial blades found at Goodall
Tradition sites are more similar to Illinois Middle Woodland bifacial blade production,
than that seen in Ohio (Mangold and Schurr 2006).

As discussed above, there is also evidence for Illinois-produced Havana wares
at southwest Michigan sites. Whether the presence of these ceramics is due to direct

population migration from Illinois into this area of Michigan, or through diffusion and

134

down-the-line exchange with Goodall Tradition populations as ‘intermediaries’ is
currently a question up for debate. Previous research (Brashler et al. 1999; Kingsley et
al. 1999; Mangold and Schurr 2006) had argued for Hopewell migration into the area.
Based on more recent research (Brashler et al. 2006; Garland and DesJardins 2006) into
Hopewell manifestations in southwestern Michigan, however, it appears that diffusion is
the more likely scenario. The diffusion hypothesis is also supported by the presence of
some Ohio material, such as Flint Ridge chert and a vessel found at the Schilling site with
a herringbone decorative pattern that is reminiscent of Ohio Hopewell (Garland and
DesJardins 2006). The presence of other Hopewell manifestations, such as Ohio
Hopewell, would be unexpected if southwest Michigan Hopewell was derived from a
population migration into the area. The likely method of diffusion of Illinois Hopewell
into southwest Michigan and subsequently into the Saginaw Drainage is likely via river
routes.

Hinsdale’s (1931) map of river portages for Michigan and northern Indiana and
Ohio identify portages between the Kankakee to the St. Joseph River. From this point,
there are a number of paths (both overland and via waterways) from southwestern
Michigan to the Saginaw Drainage. The most prominent portage from western Michigan
to the Saginaw Drainage is between the Maple and Looking Glass Rivers to the
Shiawassee River (see Figure 6-2 for a reconstruction based on Hinsdale 1931 and
Howey 2007). Along the Grand River, at its confluence with the Maple River in Ionia
County, there is also an Early Woodland earthwork enclosure site, Arthursburg Hill
(201A37; Figure 6-2) which Beld (1993b) characterizes as a population aggregation site

for social integration purposes.

I35

Figure 6—2: River routes from southwest Michigan to the Saginaw Drainage

. 6/ .

 

 

    

- in?"
----- d" ’3‘,
f | \I’IJ‘ 'n..’I“_'.. 7‘ .
' '. ' ‘ h"
k *0 i.‘ "' J3 . ‘J§ )icn' ~.
‘. _ | - .—
j ‘g 1 _ .‘3‘7h ; _;fl 7
.5 — ‘ --‘_ .~..‘{5‘.\
l -.. ’3 . j. 1' . f‘ Irw‘ ‘1
. . r ” ----... . .. ._... -
‘-.-; ' tr”:- .3 —
‘ A D I.
I , '- ~ ‘5 . .-‘ff' or“;
—- Rivers .. .. .,,- ‘
I U c
g, A V . I
3 Lakes I ' -.'l ’4 ‘2‘. 1‘1‘ \
I F I _, .' . "‘IA 3

Meters Above Sea Level ' .- f .
LLJ Over 550 , ' ' " T '- ’13 ' . ‘ M ,5 'r ‘ Huron
- 451-500 - - . ‘ : ‘

' 401-450

. h
\ .~
I" .~ " " . _‘..;
.. *- 7‘ ' -: ake
. .' '1 -. >| .. 7'.“-
I I '
. . , i a. L
I ' _ , .
1 r
I d

    

 

 

 

 

 

[.1 351-400 Lake I":
Lfil 301-350 ‘ Michigan r. 1‘
|__1 251-300 .
[3 201-250 " ,_
fl 0-200 ,3.»
. ‘ :25: if:
0 40 so '. -
“ ,_ ‘

 

 

 

 

Kilometers l

Arthursburg Hill is considered to have served some type of population
integration function not only because of the presence of the earthwork enclosure, but also

due to the presence of numerous types of lithic material, both local and non-local;

136

examples include Bayport, Norwood, Onondaga, Flint Ridge, and Wyandotte/Indiana
Homstone (Beld 1993b; Garland and Beld I999). Areas of origin for these raw material
types include the Saginaw Drainage, Northwest Michigan, Ontario, Ohio and Indiana.
Arthursburg Hill also has evidence for both domestic and ritual activities conducted at the
site (Garland and Beld 1999). Evidence for domestic activities such as stone tool
manufacture, cooking, and plant and animal processing are present while evidence for its
integrative function is due to its location on a major river confluence and presence of
exotic materials (lithics, c0pper, slate pendant). The presence of black bear, beaver, turtle
and bobcat in the faunal assemblage could also speak to the possible ritual function of the
site. All of these are common figures in Great Lakes Native American mythology and
items such as fur, teeth, claws, and turtle carapaces for rattles often play a role in ritual
ceremonies (Clifton 1978; Densmore 1929; Howey and O’Shea 2006; Lovis 2001).
Exchange of exotic materials, including maize, could have been conducted at social
integration sites such as Arthursburg Hill.

The presence of Illinois influenced material culture in the Saginaw Drainage, as
well as the presence of Bayport chert, the most common lithic raw material utilized in the
Saginaw Drainage, found on Middle Woodland southwest Michigan sites, argues for
some kind of interaction between the two areas. Social integration sites located between
the Saginaw Drainage and southwest Michigan, such as Arthursburg Hill, could have
acted as point for exchange between these two areas. How this data ties in with other
evidence for down-the-line exchange of maize, as well as a discussion of why there is this
shift to maize cultivation during the Late Woodland period, documented both macro- and

microbotanically, are provided in the next (and last) chapter.

137

CHAPTER 7

Conclusion

Based on the existing evidence for early maize, ceramic decoration, and lithic
raw material and projectile point styles, the current best explanation for the presence of
maize in the Saginaw Drainage at c. 0 AD is due to down-the-line exchange. The most
likely area from where this maize originates, based on the information presented, is
within Illinois. This is not the first microbotanical analysis which has concluded the
earlier presence of maize in an area to derive from down-the-line exchange. Phytolith
and starch studies conducted by Boyd and Surette (2010) and Zarillo and Kooyman
(2006) make arguments for down-the-line exchange of maize as well.

The Zarillo and Kooyman (2006) study examined residues derived from a
grinding slab and grinding stone found on a site located in Calgary, Alberta on the
Northern and Canadian Plains. Due to the informal nature of these tools, a use wear
analysis was performed prior to sampling the artifacts for starch grain analysis. The use
wear analysis confirmed these artifacts as being utilized for plant processing (Zarillo and
Kooyman 20062483). The results of this analysis identified a number of plant species
including choke cherry, saskatoons, and maize. Zarillo and Kooyman explain the
presence of maize through trade, due to environmental limiting factors in the Northern
Plains. Based on ethnographic and early historic accounts, Zarillo and Kooyman
(2006:490) identify the Middle Missouri region as the origin for trade items into the
Canadian Plains. Throughout prehistory, especially c. AD 1000 and later, there was
contact and influence between populations within the Missouri River valley and farther

east toward the Mississippi and Illinois Rivers which does exert some influence on more

138

peripheral regions and populations (Brown and Sasso 2001; Mainfort 2001; Theler and
Bozshardt 2006).

Boyd and Surette’s (2009) study of a series of Northern Plains and Boreal
Forest Middle Woodland Laurel Tradition (300 BC-AD l 100) sites also demonstrates the
earlier presence of maize. Out of their sample, ten Laurel sherds derived from ten
different sites produced evidence for early maize (Boyd and Surette 2009: 124).
Microfossils recovered include only maize starch, only maize phytoliths, and both maize
phytoliths and starches. Boyd and Surette (2009) argue that the presence of this earlier
maize is most likely due to down-the-line exchange and indirect influence from upper
Mississippi River Valley (i.e. Illinois) Hopewell groups. The timing for the presence of
maize at these Boreal Forest and Northern Plains sites coincides with the earliest
evidence for maize at the Holding site in Illinois (Boyd and Surette 2009:129).
Importantly, Boyd and Surette view this down-the-line exchange of maize as a primarily
social occurrence (i.e. gift exchange) rather than subsistence based.

Both studies demonstrate the presence of maize in unexpected contexts. The

Northern Plains and Boreal Forest are not conducive to maize cultivation, although
microclimate areas such as the Lake of the Woods in northwestern Ontario may have
allowed for low-level cultivation of maize. In both cases, however, down-the-line
exchange would have played some role in the presence of maize prior to the ubiquity of
maize cultivation in the Midwest.

There are, however, some questions raised in using down-the-line exchange as
an explanatory mechanism for the presence of early maize in the Saginaw Drainage.

First and foremost is the question “Is there evidence for maize cultivation along the

139

Illinois, Kankakee, St. Joseph, Maple, and Looking Glass River valleys?” Presently the
answer is no. Early maize macrobotanical remains have not been identified in these
areas. Most subsistence information seems to indicate wild resource utilization with
varying degrees of Eastern Agricultural Complex (EAC) cultigen incorporation into the
diet. However, taphonomy and differential recovery issues also play a role in what has
been recovered so far fi'om these other areas. To test this hypothesis of down-the-line
exchange, the sampling of residues from artifact surfaces (either ceramic or lithic) is
required. The next logical step is to test materials from southwestern Michigan and
northwestern Indiana for the presence of maize.

As a countermeasure, materials from southeastern Michigan, southeastern
Ontario, and northeastern Ohio also need to be analyzed. While the exchange of maize
from southern Ontario was discounted due to the lack of ceramic style affinity, the
exchange of lithic raw material does indicate contact. While 1 do believe that there is
likely a secondary mechanism Operating for the exchange of lithic raw material from this
area into the Saginaw Drainage, that of access proximity and efficiency of transport, this
region around Lake Erie cannot be discounted in regards to future testing of residues.

In fact, due to the apparent restructuring of both subsistence and settlement
patterns during the Late Woodland, in both the Saginaw Drainage and southeastern
Michigan, which clearly indicate affinities between the two areas, analysis of materials
from this region is critical. A major question in the past thirty to forty years of research
around the western shore of Lake Eric is how these materials relate to one another and
why there is such a dramatic shift in Saginaw Drainage subsistence and settlement

towards this region at c. AD 1000 (Brashler et al. 2000).

140

One potential explanation is environmental change occurring between the
Middle and Late Woodland periods. While the early Middle Woodland demonstrates an
increase in flooding of the Saginaw Drainage, the Late Woodland period appears to
coincide with a ‘drying out’ of the area. The Medieval Warm Period, which begins c.
AD 1000, is manifested across the northern hemisphere with a trend of warmer and drier
temperatures. Causes for the onset of the Medieval Warm Period are uncertain but it
appears to be related to increased solar activity and/or a decrease in volcanism which may
have amplified other climatic factors elsewhere (Booth et al. 2006; Crowley 2000).
Other factors which appear to be affected by or have an effect on the drying trend
occurring during the Medieval Warm Period include sea surface temperatures (in the
Tropical Pacific, North Pacific, and North Atlantic) which had a combined effect on
circulation patterns in the atmosphere. These modified circulation patterns appear to
have played a role in the drought and drying trend seen across the western Great Lakes
and elsewhere (Booth et al. 2006:423-425).

Evidence for drought during this time period is seen at both Hole-in-the-Bog in
Minnesota and Minden Bog in southeastern Michigan. Paleohydrological reconstruction
fi'om these two bog sites, located approximately 1000 km apart, correlate to drought
during the Medieval Warm Period at AD 1000 and AD 1200 (Booth et al. 2006). This
correlates with other evidence for drought and a general drying trend post-AD 1000 as
demonstrated in the growth of sand dunes on the shores of Lake Michigan and Lake
Huron and in other areas of the Midwest (Hanson et al. 2009; Lovis et a1. 2009).

Finally, there is also evidence for a drop in water levels in Lake Huron and the

Saginaw Bay. Fluvial changes manifested in the Shiawassee River indicate that at this

141

point in time, the river shifts its channel and there is also evidence for both vertical and
lateral erosion (Lovis et al. 2001; Monaghan and Lovis 2005; Speth 1972). This pattern
of erosion is consistent with a faster stream flow and is indicative of a drop in water
levels for Saginaw Bay. This change in stream flow is indirectly corroborated with the
drop off in identification of wild rice phytoliths during the Late Woodland. Wild rice has
very specific ecological niche requirements including clear slow moving water and
muddy river bottoms (Vennum 1988). An increase in stream flow would more than
likely shrink any existing stands of wild rice, possibly to the point where this
carbohydrate-rich grain could no longer be relied upon as a staple.

Cumulatively, these environmental changes especially the evidence for drought
in conjunction with faster stream flow, indicates that the wetlands so characteristic of the
Saginaw Drainage, were likely decreasing in size and/or in productivity. All subsistence
evidence prior to the Late Woodland period indicates heavy reliance on aquatic and
wetland species (Lovis et al. 2001). Drought would have forced a shift in subsistence.
This shift is corroborated by the fine—grained recovery and analysis of sediments, flora
and fauna at Schultz in a small-scale re—excavation by Lovis in 1991; reliance on aquatic
species drops significantly (Lovis et al. 2001). Incorporating the subsistence data derived
from the residue analyses conducted in this study, residues from Middle Woodland
pottery demonstrate the presence of wild rice. Wild rice is a nutritious grain, high in
carbohydrates which is and was extensively utilized by Native American populations in
the Great Lakes region (Vennum 1988).

While the current range of wild rice is limited to the Upper Peninsula of

Michigan, northern Wisconsin, Minnesota and Ontario, there is good prehistoric evidence

142

that it occurred farther south within the Great Lakes. Moffat and Arzigian (2000), Egan
(2001), Ford and Brose (1975), and Parker (1996) have all recovered wild rice
macrobotanicals from sites in southern Michigan and Wisconsin. An important
characteristic of wild rice is its ecological sensitivity to habitat requirements. It prefers
clear, slow moving, sandy bottomed streams, lakes, and rivers (Vennum 1988). These
environmental conditions were present during the late Middle Woodland, once lake levels
had stabilized and intensive flooding in the Saginaw Drainage was no longer an issue
(Lovis et al. 2001; Monaghan and Lovis 2005).

However, once water levels in Lake Huron and Saginaw Bay dropped and the
drying trend brought on by the Medieval Warm Period intensified, these conditions for
wild rice would no longer have existed or would have been significantly diminished.
These conditions would have made reliance on wild rice as a carbohydrate source
untenable. The likely result is that maize, which had been incorporated in low levels in
the previous subsistence regime, intensification occurred. The ubiquity of maize
cultivation at around AD 1000 is most likely a response due to this onset of drier
conditions.

The realignment of settlement in the Saginaw Drainage with areas to the
southeast is also apparent at this point in time. Brashler’s (1978 and 1981) study of
ceramic types in southern lower Michigan demonstrates that while three distinct cultural
traditions (Allegan, Spring Creek, and Wayne Traditions) exist for the early Late
Woodland period, there is also substantial fluidity in social boundaries, especially for the
late Late Woodland (Brashler 1978:347). Brashler (1978:347) cites the later integration

of Younge and Western Basin Traditions into southeastern Michigan and the unclear

143

boundary relationship between southeastern Michigan and Saginaw Drainage groups as
evidence for this social fluidity. Younge Tradition ceramics appear to have stylistic
affinities to Glen Meyer ceramic types in southern Ontario and sharing of decorative
motifs and techniques is evident on Younge and Wayne Tradition ceramics (Brashler
1981:334). Finally Brashler (1981) also provides evidence that Western Basin Tradition
groups may have cultivated crops. Maize found at the Sissung site, which has clear
Western Basin affinities, has an associated l4C date of AD 700 (Brashler 1981; Prahl
1969; Prahl et al. 1976). Further inquiry into southern Michigan social boundaries and
subsistence is continued by Krakker (1983).

Krakker (1983) examines ceramic and settlement system change during the Late
Prehistoric period for southeastern Michigan. Krakker’s specific focus in this study is on
post-Wayne Tradition wares, specifically Riviere au Vase, Younge, Springwells, and
Wolf wares. These wares represent a temporal sequence starting c. AD 900 and
continuing until c. AD 1400 and, importantly, also demonstrate continuity from one
stylistic tradition to the next; transitional forms are present and abrupt changes in ceramic
styles do not appear to exist (Krakker 1983:184).

In Krakker’s settlement analysis, he finds there is evidence for an increase in
community size over time and that this occurs at a rate which suggests population
aggregation (Krakker 1983:521). Krakker (1983:522-525) attributes this trend towards
an increase in settlement size due to maize cultivation and the selection of sites which
have optimal characteristics such as well-drained arable soil, access to water, and are
(potentially) easily defensible locations. However the onset of the Medieval Warm

Period at O. AD 1000, and the climatic changes caused by it, could have also spurred on

144

community aggregation as a means to conserve resources and buffer risk. This is further
supported by the realignment seen in the Late Woodland period with areas in southern
Ontario and northeastern Ohio. These regions may not have been affected in the same
way as the Saginaw Drainage, and hence had resources which may have become scarce
or were not usually available in abundance. Southern Ontario groups may have also had
more experience with maize cultivation due to its early incidence in southwestern Ontario
and the nearby Finger Lakes region of New York and its likely spread east towards
southeastern Ontario.

To sum up the changes in subsistence discussed above, subsistence during the
Middle Woodland is based on intensification of wetland and riverine resources, the use of
Eastern Agricultural Complex (EAC) crops, and some use of maize likely present
through down-the-line exchange. Late Woodland subsistence demonstrates a decline in
the reliance on wetland and riverine resources, the ubiquity of maize cultivation, and a
shift in settlement patterns towards southeastern Michigan. With the data generated from
this study incorporated into the above summary, how does this new data answer the

research questions initially posed?

Research Results

Q1: How much earlier is maize being 'plugged into' traditional subsistence
practices and does this have any eflect on the seasonal settlement-subsistence round?
Earlier use of maize was found to be present as early as 500 years prior to the current
maize macrobotanical evidence for the Saginaw Drainage. However, it’s probable
presence through down-the-line exchange rather than low-level cultivation is not

consistent with the idea of maize being ‘plugged into’ the existing subsistence system.

145

Due to my current hypothesis that maize was not undergoing cultivation in the Saginaw
Drainage its presence via down-the-line exchange most likely did not have a drastic
impact on the settlement-subsistence round. If anything, the incorporation of maize in a
dried form may have enabled Saginaw Drainage inhabitants to remain more effective
foragers by allowing for more prolonged foraging and/or hunting trips (Raymond and
DeBoer 2006; Wills 1988).

Raymond and Deer (2006) discuss the use of maize in Amazonia among
various hunter-gatherer groups as a supplement to a collected diet. Examples of Gé
linguistic populations, especially the Shavante, provide evidence of highly mobile hunter-
gatherers who grow maize in fields along travel routes (Raymond and Deer 2006).
Maize harvest for the Shavante coincides with community rituals and as a result, fits into
their prescribed settlement-subsistence pattern. While some maize is used ceremonially,
it is also eaten daily after harvest, stored on the stalk, hung from but rafters, stored in
baskets or ground into meal (Raymond and DeBoer 2006:340). This mixed use of maize
in regards to context, daily versus ritual, is also demonstrated in eastern North America
(Callendar 1978; Trigger 1969; Trigger and Day 1994).

The use of maize in dried form as a travel and trade food is also seen throughout
the ethnographic and ethnohistoric literature of eastern North America. Accounts of the
Huron, Iroquois, Cherokee, Choctaw, Hidatsa, Sauk, and Ottawa (among many others) all
mention the growing of maize and its use in trade (green, mature, or dried kernels) or as
ground meal taken on long-distance trips (Bushnell 1909; Callendar 1978; F eest and
Feest 1978; Perdue 1998; Quimby 1962; Rogers 1994; Trigger 1969; Trigger and Day

1994; Wilson 1987). While many of these accounts identify EurOpean trading posts as

146

the centers for where exchange of maize for other goods occurs, these posts sometimes
acted as areas for social aggregation in ways comparable to what is hypothesized for
prehistoric earthwork sites (Garland and Beld 1999; Howey and O’Shea 2006; Milner
and O’Shea 1998; Nassaney et al. 2007; Thwaites 1896-1901).

Q2: Is this earlier use of maize continuing the trend of greater reliance on
collective resources? This is probably true since there seems to be little impact on the
subsistence system with the earlier presence of maize. As a result, the continuing trend
of reliance on collective resources remains unchanged. If exchange of early maize is
occurring at social integration sites like Arthursburg Hill, then this could potentially be
viewed as a collective resource. Sharing of maize during social integration rituals or as a
means to buffer risk between regions could have been a means to solidify group identity
and/or alliance (Howey and O’Shea 2006; O’Shea 1989; Raymond and DeBoer 2006).

Q3: T 0 what extent does the use of maize overlap with the utilization of Eastern
Agricultural Complex domesticates in the western Great Lakes? The overlap between
maize and EAC plants is not seen in the manner I originally expected. The earlier
presence of maize does not appear to represent an overlap in the cultivation of maize with
EAC cultigens at this point in time (AD 0-200). However, this earlier presence of maize
does indicate potential overlap of use in terms of cooking and its incorporation into the
diet. Based on the location within the site context of sherds identified with early maize,
these appear to be daily and not special use contexts. However, further residue testing
may prove otherwise. The site with greatest potential to do so is Schultz; the stockade
present at the site is interpreted as having a ritual function and could demonstrate

differential use of maize (Fitting 1972).

147

Q4: Is the low incidence of maize macrobotanicals anomalous? Most likely the
low incidence of maize macrobotanicals in the Saginaw Drainage is not anomalous.
Based on the interpretation of the archaeological data provided by the experimental
results, it appears that the earliest regional use of maize is in a dried form. If this dried
form is being both transported and used off the cob, and possibly ground into flour,
experimental results indicate that it is unlikely that any maize macrobotanicals would be

found or incorporated into the archaeological record.

Future Research

Clearly much work remains to be done in order clarify the origins of early
maize in the Saginaw Drainage. Down-the-line exchange is the most feasible explanation
for the presence of early maize, and while Illinois as an origin is the best explanation
given the current subsistence and cultural affinity data, firrther testing of ceramic residues
and from areas to the southeast and southwest is required. Additionally, I want to
incorporate lithic material for testing as well. Ideally this would be ground stone, such as
mortars and pestles, but formal and/or informal tools used for cutting could also have
potential for microbotanical recovery. Testing of lithics, in conjunction with ceramic
residues, would allow for the incorporation of more sites within the Saginaw Drainage
that were previously discounted either due to lack of ceramics or lack of ceramics with
residue present.

Southwest Michigan sites which could best elucidate connections between the
Saginaw Drainage and Illinois are those with evidence for Illinois Havana wares, such as
Moccasin Bluff, Jancarich, Sumnerville and Toft Lake. Testing of residues on these

wares for maize could potentially provide direct evidence for maize exchange into

148

Michigan. Other southwest Michigan sites which could be tested, due to evidence for
interaction between Illinois and Michigan, include Eidson, Prison Farm, Elam,
Armintrout-Blackman, and Simpson. However, this isjust a handful of potential sites.

Testing of ceramic residues for evidence to the southeast would include sites not
only from southeastern Michigan but also southern Ontario and northeastern Ohio. The
realignment which occurs during the Late Woodland does have precursors, primarily in
the exchange of lithic raw material, in the Middle Woodland which must be considered.
The Middle Woodland period in southeastern Michigan is not well known but one site
which could be targeted for residue analysis is Fort Wayne (Monaghan and Lovis 2005).
Examination of the significant collections from southeastern Michigan of Late Woodland
material to assess for the presence of material from earlier time periods would be
required.

Finally, testing of materials from sites ‘in-between’ southwest Michigan and the
Saginaw Drainage is warranted. These sites could represent possible exchange points
between the two regions. Potential sites for testing include Ayen and Arthursburg Hill,
which both have Middle Woodland components. Given the probable function of
Arthursburg Hill as a social integration site, testing of ceramic residues here would
especially prove beneficial. In obtaining the '4C dates from Arthursburg Hill, Beld
utilized residues from two ceramic rims (Garland and Beld 1999); whether any residue
remains is unknown.

While the working hypothesis for this future work is that the early presence of
maize is due to down-the-line exchange from Illinois, it is also entirely possible that low-

level maize cultivation was occurring in these regions. The presence of maize phytoliths

149

in residues from these areas would be indicative of low-level cultivation, rather than
down-the—line exchange of maize in a dried form. Testing of lithic material for
microbotanical remains from Saginaw Drainage sites could also prove low-level
cultivation if maize phytoliths were identified to the same time period as the early dated
ceramic residues. There could also be implications for seasonality depending on where in
the Saginaw Drainage these sites occurred and possibly indicate the presence of ‘trail

fields’ as discussed in Raymond and DeBoer (2006).

150

Appendix A: Laboratory Protocol for Phytolith Extraction

Modified fi‘om the Archaeometry Laboratory, University of Minnesota—Duluth
Digestion procedure was used on samples taken from ceramic residues and for processing
plant material and experimental material. Methods for removing residues adhering to

ceramics discussed in Chapter 5.

Schulze Digestion Procedure

Part A: Water bath digestion

1) Prepare Schulze solution

-Prepare KClO3 solution by adding powdered potassium chlorate to distilled water
until saturated. Note: 10ml of water to 0.7g KClO3 is a saturated solution.

-Add 3 parts of concentrated nitric acid (HNO3) to 1 part saturated potassium
chlorate (KClO3) solution. Be sure KCIO3 dissolves completely; stirring by hand or
using a stirring plate aids in faster dissolution.

-Allow 15 to 20ml Schulze solution per sample for the initial digestion. If a
second digestion (after centrifuging) is required, allow 10 to 12ml per sample.

-ALWAYS ADD ACID TO OTHER CHEMICAL. Wear gloves and apron for
protection while preparing solution; a mask and breathing apparatus are also available.
Measure chemicals in graduated cylinder; pour from small containers only. Transfer
from large bottles to beaker if necessary. Glass rods control flow and prevent drips.

SCHULZE SOLUTION MUST BE MADE FRESH FOR EACH DIGESTION SET

2) Add 10-12 ml of Schulze solution to each tube.

151

-Set varipet to desired volume. To avoid contamination, tip must not come in to
contact with the sample.

- If there is a lot of plant material or sample, place a slender glass rod in each test
tube and stir to remove any air pockets that may develop.

-Place tubes in preheated water bath. The bath should be approximately 90-95°C.
Water may need to be added to the bath as digestion occurs since low water levels may
cause bumping and potential sample contamination.

-Submerge sample in acid with glass rod. The samples must be monitored very
closely for the first hour in the bath, since convection currents often cause undigested
material to rise to the top of the tube and occasionally overflow (especially during the
first half hour).

-As digestion continues, less constant attention is necessary. The acid level in the
tubes should be maintained at the 13ml level ( 1-2ml of the lip).

-Check water level and samples every half hour. Submerge samples and agitate
with rod to enhance digestion.

-Continue digestion, stirring frequently, until the sample has been completely
digested, or until additions of acid fail to produce a reaction (approximately 10 hours).
"Note: for small samples (often those characteristic of charred material taken from
ceramic residues) the water bath can be turned off after an hour or two and the reaction

will continue overnight. Small samples usually do not need a change of acid. NEVER

CENTRIFUGE HOT SAMPLES

3) Change acid

-Examine labels; if smudged, check against records and relabel.

152

-Remove completed samples from water bath and cover with parafilm.

-Prepare fi'esh Schulze solution (step 1).

-Submerge all plant material below liquid level; rinse rods and equalize tube
levels with excess Schulze. Remove rods. Centrifuge at 3000rpm for 15 minutes. If
material remains floating, return the tube to the water bath. If all plant material has
settled, decant, refill with fiesh Schulze solution, agitate with glass rods, and return tube
to water bath.

-Fresh acid increases the speed of digestion greatly. Additional changes of acid
should be done as needed to complete digestion efficiently.

-Do NOT digest all organics completely (i.e. until a clear liquid remains). A

small amount of plant material is desirable to aid centrifugation of phytoliths.

4) Rinse samples.

-Wear gloves for the first wash in this step.

-Centrifuge samples at 3000rpm for 15 minutes. Decant supernatant.

-Agitate remaining liquid and sample material with a glass rod. Rinse glass rod
with distilled water while removing it from the centrifirge tube.

-Equalize liquid levels and centrifuge samples again (3000rpm, 15 minutes).

-Repeat water wash until liquid in the centrifuge tubes is clear

-Rinse with 95% ethanol, centri firge and decant. DO NOT MIX ETHANOL

WASTE WITH SCHULZE SOLUTION AS A VIOLTENT REACTION WILL OCCUR.

5) Sample storage
-Rinse 1 dram storage vials with 95% ethanol. Label vials with sample number

and date and cover with scotch tape. Ethanol smudges labels; keep outside of vials dry.

153

-Using a disposable pipette, transfer sample material from the centrifuge sample
tube to the vial. Spray inside of centrifuge tube with ethanol. With the same disposable
pipette, suck up ethanol and squirt back into centrifuge tube. Suck ethanol back into the
disposable pipette and transfer to storage vial.

-If vials are too full, place stoppers at the bottom of each centrifuge well and place
vials in the centrifuge. Centrifuge at 3000rpm for 15 minutes. Tweezers may be required
to remove vials fiom the centrifuge. Use a new disposable pipette to remove the extra
liquid.

-Cover vial with parafilm to prevent evaporation and cap. Twist parafilm down

over vial top to reduce tearing.

Part B: Slide Preparation

1) Set slide warming tray at low to medium heat. Ethanol should evaporate

quickly but organic residue should not burn.

2) Label slide with information on storage vial and cover label with scotch tape.

Place on slide warming tray.

3) Using a disposable pipette, transfer 3-4 drops of material from bottom of

storage vial to the slide. Let residue dry.

4) Add 3-4 drops of either Permount or Canadian Balsam (either mounting
medium is acceptable) to dried sample material. Remove slide from heating unit and

carefully place cover slip over mounting medium. Gently press bubbles to one edge

using toothpick. Place flat in storage to dry.

154

List of ui ment

Scalpel
Kimwipes

Balance

Weighing paper
Centrifuge tubes

Test Tube Rack

Wax Pencil

Beakers (50ml and 100ml)
Graduated cylinder (25ml and 100ml)
Funnel

Chemical spoon

Watch glass

Gloves (non-powdered)
Apron

Lab Coat

Face Shield

Water Bath

Fume Hood

Nitric Acid

Saturated Kcro3
Distilled Water

Ethanol

155

Permount
Canadian Balsam
Glass rods

Varipet

Disposable Pipettes
Centrifuge

Labels

Vials

Slides (3x1)

Cover slips

Toothpicks

Appendix B, Figures Bl-B72: Artifact Photographs

Note: All scales in photographs are equal to 10 centimeters

Liber_ty Bridge (ZOBY77): Figures Bl-B3
Figure B1:N10E8.ISE-4

 

Figure BZ: 2.NOE0.28E-3

“‘1 8 3

 

156

Figure B3: 2.NOE0.2NE-2

 

Kantzler (20BY30): Figures B4-B7
Figure B4: 77127

 

157

Figure B5: 77112

 

 

Figure B6: 77102

 

 

158

Figure B7: 77087

 

 

20SA1276: Figures B8-Bl3

Figure B8: FS 394

   

 

159

Figure B9: FS 390

 

FigUre BIO: FS 356

160

Figure BlleS 305

 

 

Figure 812: FS 192

 

 

 

161

Figure 313: FS 153

 

ZOSA1251: Figure 1314

 

Figure 814: Shiawassee #2

 

162

Schultz (ZOSAZ): Figures BIS-B53

Figure BIS: 65888

 

 

 

Figure B16: 65826
llllllll' l‘ll'lfll l'll'llfl l'llill l lllllll'lp'l' l‘il El'l ll"f’li " E' E’ ftlf '
“I l 2 i a e . r ' e '

   
   
 
 

      

I63

Figure 317: 65557

 

Figure B18: 65409

 

164

Figure B19: 64718

 

Figure B20: 64676

 

 

165

Figure B21: 64622

 

Figure 322: 66361

 

166

Figure B23: 66160

 

Figure B24: 66020

 

 

167

Figure 825: 66004

 

 

Figure 326: 66003

 

168

Figure BZ7: 65961

 

lll‘:ll|l”‘;ll""l'lv' w

“Iv 2

      

Figure 828: 65920

 

169

Figure B29: 65908

 

Figure B30: 67878

HHIHHHHHHH
e ‘ ~

 

170

Figure B31: 67845

 

F igUre B32: 67674

 

l l

171

Figure B33: 67469

 

FigUre B34: 67275

 

172

Figure BBS: 67207

llllllll:ll'lllllll‘llllllll'll

 

Figure B36: 66839

 

173

Figure B37: 66656

 

FigUre B38: 66656

  
        

O
‘

111111111. llllllll: 111111111 1111,1111 1111:1111 111

1' -l'ltl“!”1r
7' .

   
 
 

I'mmm .. ’mw-n 'r'v
I . '

 

 

174

Figure B39: 69552

 

F igUre B40: 69477

 

175

Figure B41: 69276

 

F igUre B42: 69080

   

176

Figure B43: 68567

 

lllll'l" 111”""1"l"l'1"" ... ...
urn-co- E

   
 
   

J'x

Figure B44: 68559

' .1»
“‘“v ....‘1’? ’- * .

...o-r- "<§'_ "'

ll: lur' DI [IQI’Itt'rtfyrgtI‘i

ll‘l'lll‘ llll um H' " " "" "“"1"”l."”l‘
l

      
 
 

177

Figure B45: 67916

 

178

Figure B46: 67896

 

179

Figure B47: 67232

 

ll‘ 'lll"ll"1.‘* 'l I 'l H" ' ..- 1""!
We“

 

Figure B48: F29

 

180

Figure B49: 69771

 

’11 '1

 

Figure BSO: 69743

 

181

 

 

Figure BSI: 69760

 

FigUre B52: 69666

 

 

182

Figure B53: 69606

 

Tobico Marsh Sites (20BY192 and 20BY2 @: Figure BS4

 

Figure B54: 20BY210

1111""! 1' 1
‘W' 2 J 4

 

183

Simons 249 (20SA665 1: Figure BSS

Figure B55: Vessel 92

 

Solms (20SA57): Figures BS6-B62

Figure BS6: SAC 70 #1

 

184

Figure B57: SAC 70 #2

 

FigUre BS8: SAC 35

 

  
  
 

1111]“111111'11111111'1111”11,1111111113111111111'111111”ll"ll"ll.’"l”llllllllllll'f'l'"
I ‘ I ‘ ' ._..- ..r .

i ’ - .- '

  

     
 
  

   

185

Figure BS9: Vessel 435

 

186

Figure B60: SAC 31

 

urnmtmmimll"!

      
  

11’11111111111111111111”|
i I

‘111'1l"11711"17’11‘11léllt'tlf
. ‘1 l . I .

.‘lttllunrn'y
, ,. 7 '

      

Figure B61: SAC 15 #1

   

I87

Figure B62: SAC 15 #2

 

Stadelmeyer (208A 195 ): Figures B63-B68

Figure B63: 76986

 

 

 

 

 

 

188

Figure B64: 73995

111111111111 llll'lllllllllzlll'll'll
' i O i r

 

Figure B65: 73970

AI- n
u.‘ -!
4‘ .r
.A'.‘ '3!
..5‘ a ‘
)1.
.‘. . t
r
. - ‘ o
‘w‘i ; T 1'.
W. ‘ _
‘ ' 7.1' tr 1,.- ‘
O

 

ii’lllll-ls-|'-.__ ."l. “.1 '1 ll ' .‘ l, ' , .’ I l: ,: 4'

 

189

Figure B66: 73949

 

1.1.11'

Figure B67: 73942

 

190

Figure B68: 73931

 

  

 
  

  

’ """""’1'lT.’"l'11"1"I”1' 'll""','"l'.’1'l’ 1‘ ‘1 1‘1 1'1"'7""1"1"1:'1"l'7.'1i
‘ .

. 17 ., - : t i

      
  

Foster (20SA74 ): Figures B69-B7l

Figure B69: 76639

 

191

Figure B70: 76643

 

Figwe B71: 2861

 

 

 

192

 

r I. .0. .1! ...}! . t .‘r.’ 0'

 

 

r . .J‘u’OI

B72

igure

F

Clunie (208A722)

FS 433

Figure B72

 

 

193

 

 

I‘lhfrll‘I-I'; ... I, e1. lb

 

..”.t .....l Ii 11

 

 

 

 

 

 

 

 

 

 

88.2.2
.58 :85 @566
m e o em 38 82 3. as?
a 0 38¢
6% 3. .8 a _ 3. >2th
82-2w

o a e em 38 $2 an 236 8. ma SEEN

.8: 3. so
o o a use aeaem .see :85 BE 8932 2258
o e c seem e620 3: can we 82¢.ch

.§E>
e o e 5&8 >5 3:8 an me £258
6.985
8:: l

o c o .82 em 38 6333882: 32 .236 8». we ohm—$8

$2: 9.. .5

88788

_ e o em 38 8 9.. geese: «a me £258

39% em .8
o c 3 Ease... 8. we meadow

2.56:...— eflam 5:63....— Aaeuaé .
2: o o e e
8.: 2.3 use: out: 3.5 95. s 352 o 8... .— ..E. a E ..m 35

 

 

 

 

 

 

 

 

 

338m ember. seem an 5:65: ”To use.

0 Hafiz—Ade

194

 

326$

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

o c o o o o c n we _om3> bgmcn
o o N o o o c mm”. mm _ «median
c _ c c o c c got—tmotm "mg/Ewen
N d o c o c c wnm mm obmudsmcn
o o o c c o o non mm era—«swam
o o o _ o o o can mm chm—Aimed
_ o o o o o o 62 mm obu~<mou
c — o o o c 6 Re mm nh<mon
Emu”... g “EM flu.” 5am... tum...» its 2...... ....

 

 

3588 To 6.3.

195

 

€8.68

 

 

 

 

 

 

 

 

 

 

 

 

 

 

036

z o 2... > $58
2 z o 8. we . 3....ch
> a . 83-8... .2338
2 z c e2 m... 2.2.88
2 z a 8m 9. ee~.<mo~
z z e an m". $2....ch
z z o a. m... £2...ch
z z o 8. mm 2.38

5:95...
8 .— eufii n 0
SE .3. m m 9...: .58.: 2.15535 .2... ..m 35

 

 

eases". .-o 6:3

196

 

5....- Bmcfiam

 

 

 

 

 

 

 

 

 

o e. 8 e5 2. a 3 3.. .28 8%.. .2ch
o e o 3s. 3.8 93. «<ch
85-88 3. .28
o ... o .... .6. .... a< .56.. e86 988 Gem...
.3: a...
6-8 em .6
8E. 3. 3.36
c o o -83. .... .6. i.e.. :86 58 ~68
-32 ea.
8..-... 9. .6
88.8.3. A...
o e o .6. 8m. 9.. 3. . 88% ~<mo~
83-on
.6-3 um .6
88.8... .5
m o ... use... .6 .6. ..m 9.. .6 Beaaeeeefi 3.2 ....a 88. 2ch
-..: em .6
88.88.
. e o .5 .6. ..e 9.. 5565...: Ne. m”. 2268
6.8 em .6
c .. ... use... .6 3. m”. Ea...
2:33.... .26 3:33.... Assam-a.
3... .... 3 use... one: 3.5 m2... 2.? 3...an 8.8.. 2.5 ... 225 38

 

 

 

 

 

 

 

 

 

82.88 .-o 636

197

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

8 8. 8 8 8 8 88888 «8.888
8 . R 8 8 N 8 $888 «<88.
8 8 8 8 8 8 8 88888 «<88.
8 8 8 8 8 8 8 $888 2888
8 8 8 8 8 8 8 88888 «<88.
8 m 8 8 8 8 8 88888 «<88
. 8 8 8 8 8 8 88. 8”. 883.888
8 . 8 8 8 8 8 88.. 88 82888
588.3... 588...... .858 .258 2.88.. 8....
82:0 3.2.09 022.90 3.5an 58¢. 58:058. euca— Bo=o> 33...— xosn ‘8 anm 85

 

 

888888 .-o 8.88..

198

 

 

 

 

 

 

 

 

 

 

 

 

 

z z 8 88888 8<888
z z 8 82.88 8.8.88
2 z 8 88888 8<888
z z 8 88888 8<888
z z 8 88888 8<888
> z 8 88888 8<888
z z 8 88_ 8.8 8882888
2 z 8 888 88 82888

5:88....
2:: 8.8288 88.588 28¢ 2.85: 8.8888525 8 8.2.8 8:8

 

 

 

888888 .-u 8.8.;

199

F.
.8.

s

‘8'

l.“

V
llllfl‘.
‘ I

AVU

P!

h

n

.8.

d

h

2-

Fl.

 

EJJ

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

8 8 8 5.: 388.888 88888 8.8888
>5 8.5
8 8 8 .588 820 -3: 28. 88888 8.8888
8888.
-8888 ..8. .88
Ga
8 8 8 88. o< .80 8 .883 88888 8<888
-88. um .6
888.828 88.
. 8 8 .68 8. o< .80 .58.. 88.6 3.). 88888 8<888
-88. um .8
.888.
-88. 8 88. .68
8.18.8 um .8 88883888....8 .
.8 8.3
8 8 8 85 88888888 8.8.8888 3:. _ 88888 88.888
88. .68 888
-888 um .80
8888388388 as.
8 .8 8 .8 .588 52c 8.2-8.888 88888 8.888
88888
unocow
8 8 8 3: . 88888 8.888
2.8.8.8.... 8258 2.8.8.8.... 8858.888
8:. 8...: 8.8:. 88.82 8.88. 824. 888,8 8.5280 88.33. 85.8 8 8.2.8 3.8

 

 

82.8888 ..0 8.888

200

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

8 8 8 8 8 8 8 88888 888888

8 8 8 8 8 8 8 88888 88888
8 8 8 8 8 8 8 88888 8.8888
8 8 8 8 8 8 8 888 88 88888
8 8 8 8 8 8 8 88888 8.8888
8 8 8 8 8 8 8 88888 8.8888
8 8 8 8 8 8 8 .....88mmw8 8.8888

£838.... £838.... 89.8.8 88.8.8 853 8.3 8.58.. ..2. 8 8.2.8 8.8
888:0 £82.00 3.2.00 8.3.36 ...—348 saute—.28 ES.— Be=o> a .

 

 

82.8888 ..0 8.888

201

 

 

 

 

 

 

 

 

 

 

 

 

 

z z 8 88888 88.888
2 z 8 8888 88.888
2 z 8 88888 88.888
88 z 8 88888 8.888
88 z 8 88888 88.888
2 z 8 88888 8.8888
88888
2 z 8 .888888 8.8888
5:888:
.286 8.3.88 88.588 .238 8388... 8.8888525 8 8.2.8 8.8

 

 

888888 .-o 0.888

202

 

 

 

 

 

 

 

 

 

 

 

 

8888388§8 B8,. 8:88 8.888 8.8888
38,.
8 8 8 8.2.8.3» 8888 8.8888
.888.-88.8 88.
o .882.
8 8 8 .68 88 8.8 .6 38.4%... 88888 8.8888
-88. um .80
8 8 8 88888 88888
8 8 8 3.2 88888 8.8888
.888 .-8. .8 88.
8 8 8 .808 88 8.8 .80 .58.. :85 3.2 88888 88888
-88. um .8
8 8 . 8888388828 38,. 8:88 88888 88888
8 8 8 88883883.; 38,. 8:8 88888 8.8888
8 8 8 88883883.: as. 8:8 88,. 8.8888
888.8888
8 8 8 ..8. .88 88 Q8 38,. 8.8238 8888 8.8888
.8088 um .6
8 z 8.8 9.3 8 z 8.8 «83
out. a...” ”8.3% 888.58.“... “a: 888%. 2.88 8.826 88.88.. 28:. 8 8.2.8 3.8

 

 

 

 

 

 

 

 

 

82.8888 .-o 8.88.8

203

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

8 8 8 8 8 8 8 8.888 88888
8 8 8 8 8 8 8 88888 88888
8 8 8 8 8 8 8 88888 8.8888
8 8 8 8 8 8 8 88888 88888
8 8 8 8 8 8 8 88888 88888
8 8 8 8 8 8 8 88888 88888
8 8 8. 8 8 8 8 88888 88888
8 8 8 8 8 8 8 88888 8.8888
8 8 8 8 8 8 8 888 88888
8 8 8 8 8 8 8 88888 88888
5:38.... 5:38.... 8258 88.58 853 8....
889.0 3.550 9.82.5 eta—.00 F538 588.3858 ES.— Bc=o> 83.5 8.2.: .... anm 85

 

 

825888 .-u 8.888

204

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Z Z c 8.888 N<mo~
Z Z c 2.288 $63
2 Z . 88888 «(.ch
Z Z c 3888 $.ch
Z Z c 8.8888 N<mo~
Z Z o 388 ~<mo~
Z Z 8 838 N<mom
Z Z c 8388 ~488ch
Z Z .8 am". ~<mo~
Z Z o 888 $.ch
5:88....

:9:— oufix— 8333 o a .

.2388. . 8 8 Etc .8. 8.888.858.88— 88. ..8 ..8

 

 

82.8888 .-u 8.888

205

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

206

 

 

 

 

 

o S o e o: 888% ~<mo~
Aowowov
o 8 _ cm. 00
n a 3 8. 8 ~<mo~
o o o 5.: >55QO 32 22 $38 ~<mom
o o e E8 520 32 N23 N<mo~
o o o 32 $5 ~<mo~
o o o 83% .832 22803 5% ~<mo~
o o o ES. :86 32 338 ~<mo~
o o o 32 $88 ~<mo~
o o o 32 ES ~<mom
o o o 0833838 32 3.3 93% ~58
€8.25
o c o .59. 820 32 WRS ~<mo~
o o c 22895 88% ~<mo~
o o o E8 520 32 $88 m<mom
25.28.: :23 2.538.; 325.3
82 2;» 3:2 as: 8.5 ms? 2:8 2.528 3.5m 2:: 8 22m 25

 

 

 

 

 

 

 

 

 

80:23:00 TO 038,—.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

o o o c 88808 ~<mom
o 8 8 _ o o o o Awwwwwv N<mo~
o o o o o o c 88888 N<mom
o o o c o o c 88888 N<mom
o o o c o o o 88 888 m<mo~
o o o o o o o _8808 N<mom
o o o o o o o 8N888 N<mom
o c c o o o c 88888 N<mom
o o o o o o c 88888 m<mo~
o o o o o c c 80888 m<mom
o o o o o o o 88m88 m<mom
o o o o o o c 88888 N<mom
o o c o o o o @8888 m<mom
5:88.7— 5=3b5 49:38 89.88 maca— b: 33¢ a: .8 8.
8320 3.8.30 3.350 022.00 =..8< 5822.58 855 32.3» m v. G u 8 am .m

 

 

80:58:00 TU 038.8

207

 

 

 

:

au~fi-az5

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

z z c 888% ~<mo~
8883
2 z m a: 8 ~<mo~
z > a $38 ~<me~
z z o -~$ ~<mc~
z z o 82% ~<mc~
z z m ES ~<mc~
z z o £88 9.88
z z o 888 ~<mc~
z z o ES ~<mo~
z z o 825 ~<mo~
z z o 88$ ~<mo~
z z o 88% ~<mo~
z z o 888 ~<mc~
5:88.:
95: 8.838 99.2.8 25 38...: 2.32825 8 22.8 3.8

 

 

.8228 .6 2.3.8

208

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

25.888
8 8 .
8 8 8% 88<88~
885.888
8 8 ~
~8 8. 38 . 88<8o~
$5.888
~ 8

8 .8 8. 888 88<88~

1 885.888

~ 8 8 ~8 E 38 88<88~
95.888

8 0:: 8:88 8 :88
I ~ 8 8 Is 8 3 >3 W .888 BE 88<88~
\I 8 88 8 392.5. 8:83 >3 8:8 2.88.8888 .88<8o~
28>.» 8:883

F 8 8 8 8.8.2.8.. >5 88~88 ~<88~
F 8 8 8 as. 88888 ~<88~
F 8 8~ 8 a. o: .8888 ~<88~
_ 8 8 8 o. 8: 8~888 ~<88~
8 .. . 8.. o: .8888 ~<88~

8 - ~ 9. o: 8888 ~<88~

. 88 o a. o: 8.888 ~<88~

2.8.8.8.: 8258 2.8.8.8.... 3588-3
82 2.3 3.8:. 3.8:. 2.5 828 2.88 8.5.280 88.3,. 8...: 8 8.2.8 2.8

 

 

80:58:00 TO 038,—.

209

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

m5 0
8 8. 8. 8 8 8 8 88 988 M8<.8ww
we 0
8 8 8 8 8 8 8 88 8. 0.88 M8<.8wm
885.888
8 . 8 8 8 8 8 .8 8. 0.88 88.8888
885.888
8 8 88 8 8 8 8 88 88 988 8 8<888
885.888
8 8 8. . . 8 8 .8 88 98 8 8<888
8 . 88 8 . 8 . .- . 88.8888 .88<888
8 8 8 . 8 8 8 88888 8.8888
8 8 8 8 8 81 8 88888 8.8888
8 8 8 8 8 8 8 .8888 8<888
8 8 8. . 8 8 8 88888 8<888
8 8 8. 8 8 8 8 .8888 8<888
8 8. 88 . 8 8 8 88888 8.8888
8 8 88 8 8 8 8 8.888 8<888
5:88.: 88.38.: 828.8 88.58 853 8.3 8.8.8 2. .8 2.
889.0 3.3.30 2.3.30 2.3.30 =..8< 53.3548 ...—cm 32.8» m 8. a 8.. .8 am .m

 

 

8888888 ..0 8.888

210

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

888888
> 8 8 88 0.88 8 8.8888
.2588.

> > 8 88 8. 988 8 8.8888

885.88.

2 z 8 .8 8. 0% 88<888

886.888

.8 > 8 88 88 988 8 8.8888

, 885.88.

2 .8 . .8 88 0.88 8 88.888
2 z 8 .-..88.8888 .88<888

> z 8 88888 8<888

z z . 88888 8<888

z z 8 .8888 88.888

> 8 8 88888 8<888

z > 8 .8888 8.8888

.8 .8 8 88888 8.8888

.8 > 8 8.888 8<888

5828....
28.: 8.3.88 88.588 28¢ 8538.: 8.885.538: 8 8.2.8 3.8

 

 

82.8.38 .-u 8.88.8

21]

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

8 8 8 8888838888: 38,. 8:88 88888 88.8.888.
8 8 8 .588 8885 38,. 88.8.2 88.88.88
8-8.2

o v o w 88.

8 >1. .8 8 888.828. 88.88.88

8
8 8 8 88 8.88.828. 88.88.88
8 8 8 85 8.8888... 88.82 88.88.88
8 8 8 88888 5885.88.88
8 8 8 88888 5885.88.38
8 8 8 . 8888 8.885.838
8 8 8 88888 5885.88.88
8 8 8 >1. .8 38,. 88888 5885.88.38
8 8 8 .888: 858888,. 8.1. 8.8. 88888 5885.88.88
1 888.8883
8 8 8 8.88888 88. 88 88828888
USN—@003
8 8 . 8.88888 888 88 888. .8888
88.8.88.
8 8 8 .8 B88 8 8.88 88
888.888.. 8238 85828.... 3.88.8-8. 888.8 8:88.80 88:8 8:: E 8.. 8 8.. 8
8.: 2.3 88.82 88.82 8.8: 82.8 . . 8 .8 8 .

 

 

 

 

 

 

 

 

 

3:58:00 70 038,—.

212

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

8 8 8 8 8 8 8 88888 5.8.83.
8 8 8 8 8 8 8 88.8.2 88>m88
8 . 8 8 8 8 8 8. 888wn8. 88.88.88
8
8 8 8. 8 8 8 8 8.8 8.88.828. 88.588
8 8 8 8 8 8 8 88.82 88.588
8 8 8 8 8 . 8 88888 5885.88.88
8 8 . 8 8 8 8 88888 88.885.888.88
8 . 8 8 8 8 8 . 8888 5885.88.98
8 8 8 . 8 8 . 88888 3885.838
8 8 8 8 8 8 8 88888 5885.88.88
8 8 8. 8 8 8 8 88888 3885.88.38
8 8 8 8 8 8 8 88. 8,. 88828888
8 . 8 8 8 8 8 888 88. 8882888
8888.88.
8 8 8. 8 8 8 8 .8 8.8 8 8.88 88
88.88.... 8.88.888. 88.8.8 88.8.8 88.3 8.... 8.8.8 8 ..88 8 8.2.8 2.8
889.0 £3.30 £3.30 £3.30 ...:3< 5.2.3:}.~ ...—cm 36:8; a .

 

 

 

 

8888888 ..0 8.888

213

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

z 8 88888 5.8.88.
z z 8 88.8.2 88.88.88
8-8.2
z z . 8. 88. 82.8. 88.88.88
8

> z 8 8.8 8.88.828. 88888.88

2 z 8 88.82 88.88.88
2 z 8 88888 3885.258
z z 8 88888 3885.838
z z 8 . 8888 5885.838
z z 8 88888 5885.88.88
2 z 8 88888 5885.88.88
> > 8 88888 5885.88.88
2 z 8 88. 88 8882888
> z 8 888 88 882888

885.88.

> > 8 .8 0.88 8 88.8 88

858.88..
28: 88.8388 888888 2:: 8.88.8.8. 8.885.528: 8 .2888 8.8

 

 

 

tozczcoo TU 038,—.

214

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

o mm A C 88.8583 3.. No .388o> moo/8mg
8 8 8 . .8888888 2888.28
. 8.8588 88.88.88
:0: 3 on v
o o o 6:88:83. A n 8 E
.08. m 28.5.88 m~>mo~
co: .3 o 0 V
o o A. 63.8.8838 A a 8 E
.088 v 3.8.5.88 wm>mom
.8888.
o:
o E c Q. m. m8. o_m>mom
onumufl
o o o 33.8808. mom m. 8<
o o o omoon 3.80.8
o w o com. .808. _ :3: .820: 3.80...
38.8 3:88
o o o 3388.08.83 .cwm 3.80.8
8 8 ... 88.8588 88.88 88.8.83.
_ v o .Eom 3380 >2). N. :81 88.—SEQ
max—3.3m 8.3.35 85:38.: AaEuEév on» 2:33 $23 3:: .3 3..
35 33> 35¢: via—2 8.8: m3? h . U . 8— L. .... .8 :m .m

 

 

 

 

 

 

 

 

 

 

833E230 TO 338,—.

215

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

8 8 88 8 8 8 8 88 .8885 888.888
..8888888 82.8.8.8
8 8 8 8 8 8 8 . 8.88.88 88.88.88
:3. A3 03 V
8 . 8 8 8 8 8 88.8882 88.. 8......“
.2. m 29.38 >
:05
. .5 o
8 8 8 8 8 8 8 8.58882 .88.“...“8
38. v 29.38
8. o
8 . 88 8 8 8 8 8. 8... “.8...”me
8 8 8 8 8 8 8 888 8. 8.8
8 8 8 8 8 8 8 88888 .8888
8 8 8 8 8 8 8 . a... 88.8.. .888...
8 8 . 8 8 8 8 .888 .3888
8 . 8 8 8 8 8 88.88 5.8.888.
8 8 8 8 8 8 8 8., .88 3.8.888.
8.888.... 8.838.... 89.8.8 89.8.8 88.8.. 8....
EEO 9.3.30 9.3.30 333$ :33. 5.3.3:: 6.3.— Bo=o> 85...— 435: ..8 ...—35m 33m

 

 

 

 

8888888 .-u 8.888

216

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Z 8 No 38885 338mg
..8888888 . A6838.
2 .8 8 . 8.88.88 88.88.88
so:
> Z c 6.588.808 53:80—85
.08. m 29.38 wm>mo~
no:
> Z c 6:888:82 Cocoon—.5
38. v 29:88 wm>mom
A8395
2 2 o 2 m8. 93>ch
Z Z c mom m. 8<
z z 8 88888 .388...
Z Z c _ EB. .33: 3.80..
2 Z c Sam 3.80...
Z Z _ 8.2K 3.88:5—
2 Z c N. 2.8 3.556.
5:88....
23. 3:325 ouaam A23. 8:33.: 33:88:32.5 .... 1.3.5 3mm

 

 

388888 .-o 8.888

217

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

888.8. .8
.8088. om .80 . 8 .0 8.88.
2885.88 29.38
8
2 5.88 8038.. 8.330.... 80388. 32 23.3w A883 5.23m 33.338
888.8888 888.88
8.88- 88. .808 88 98 88.88.8888. 88. 8888888 888.888.8888 88. 88 888.888 888.8.>8. 8.8..
.8088 0.. .80 I
888-888. 38.. . 88.88
<2 88. .80. 888. 88. 88888: <2 88.88 88.0 888 .8888, 8828.28. 8.88.
-888 88.8 .80 8.8.
388952.882.
03:38 28.8588
8.8888 88.888 88.888 88.888 . 8.8.888 88888 888888 888.838.
38.. 8.888
. 8888.828 38. 888.88
. 88- 88. .80. 88 98 88. 8888888 88. 8888888 . 88888 N88888 88.888. -
8.8.888 38.. 8.8..
.8088. 08. .80
A8328=80 8.50
o 3.88. 88888.80. 888 85 8.8.. 82.8 88:88 8.5.8 8.. 5888:. 3.8 2:82 88.8.8.3.
8. 833.8350 3.355.850 35.23: 9:58.30 3.83.30 £35m
8.88.8 8 58..

 

 

G nan-3.34..

3.8.3. 8.8... 888888.888. ...: 8.88...

218

 

 

 

 

 

 

 

 

 

0.958 295.8 32 88—
8.8588 8.8.88.8 58.888 58.828 ......8 388.88 88888 N......88 8.8.8.2:
888.8888
as
8.88- .... .6. 88. 8.< 8.. 88388. 88 883888 88:.HM83 88888 5.2.8 8.8.8.22 888.88
Isom
.688. 88 .6 . .
88288.8
8.88- .... .688. ..< 88. 8888888 88. 8888888 8583.2 88888 N......88 :2838 8.88.88
. 8 =85 sum
.688. 68 .6
egg 0.9588 32 .
8.8.58 8.8.88.8 58.828 58.828 88.8.. :86 2.8.8.2 88.88.88 8.8.8.3..
.8.8. 8 32 8:8 858 888.8
<2 -888. 88: u. 88. 888888. <2 8888838888: 88888 888.<888 888. . -588
88.-8. o< .6
.888.-88.8 8m
.6. 88.8.8 08 38,. 8:8 888.88
<2 .6 85 .8888 88. 88888.8 <2 888838828 88888 N......88 888.832 -88..
.8mm .5 .80. 83.882.
888-888 08 .6
888.8888 888.88
8.88- A... .6. 88 8.< 88. 8888888 88. 883888 38,. 882.88 88888 N......88 888.838 -58..
.688 um .6
.882886
8880
u 3.2. 8888.6. 88< 8.0 3:. 8:2 88:88 28.8 ... 583...). o. 2.32 ... ....
8. nozznaau ...-53:98:80 10.588: 9.89.00 9.58.30 ..m 9.955 . .2.
88.8.8 8 38..

 

 

 

 

 

 

 

 

 

 

82.8.8.8 .-.. 8.888

219

 

88.» .6ch

3:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

@3- mm .80. cm Q< .5 ©8688 mm ©3558 .....om 50.0 88888 5.2.08 8.0.0.2.. o-WAHMMN
.8068 um .80 .
.888.
<2 -288. mm .80. .5 9886.8. <2 32 0.8. 88888 5.8208 8 .2038 8.”..0NMN
...-8.2 Q< .80
388.888. 888888
<2 mm .80. cm . Q< mm c.8688. <2 32 88.-8.880 cacao 8......om 8.0.0.2... -50m
.80...»- 0m .80 -
888.88.8-
<2 mm .80. cm Q< mm cvfiomom <2 dowua-MNMF 8838. 8......0m 8 .203... m-WWMMN
.8068. um .80 .
8.8.53
<2 mm .80. 388. .5 338m <2 .8 3... mmvmn— ~m8<mc~ 8.0.0.28. N0388-
_.80 883.888.. .80m
-838. Q< .80
€895.10 0
u 5.8.. 5.5.6. 88< 8.0 2.... 8:...V 88.3.. 8...... ... 588...: o _ 8.....2 8mm
8.. 3303.50 3.33.5280 $0858.80: 0.8.8.00 0.5.8.00 ..m 0.355 0.8.8— 5..
88.88 8 58..

 

 

82.8888 .-.. 8.88..

220

REFERENCES

Albert, DA. and SR. Denton, and B.V. Barnes

1986 Regional Landscape Ecosystems of Michigan. School of Natural Resources,
University of Michigan, Ann Arbor.

Allison, AL.

1972 Plant Remains. In The Schultz Site at Green Point: A Stratified Occupation Area
in the Saginaw Valley of Michigan, edited by, J .E. Fitting, pp.75-85. Memoirs 4.
Museum of Anthropology, University of Michigan, Ann Arbor.

Anderson, E.
1952 Plants, Man, and Life, pp. 136-185. Boston, Little Brown.

Andresen, IA. and J.A. Winkler

2009 Weather and Climate. In Alichigan Geography and Geology, edited by, RJ.
Schaetzl, J.T. Darden, and D. Brandt, pp. 288-314. Pearson Custom Publishing,
Boston, MA.

Asch, D.L. and NB. Asch

I977 Chenopod as cultigen: A reevaluation of some prehistoric collections from eastern
North America. Midcontinental Journal of A rchaeology 2:3-45.

1985 Prehistoric plant cultivation in west-central Illinois. In Prehistoric Food
Production in North America, edited by R.I. Ford, pp.l49-203.
Anthropological Papers No. 75. Museum of Anthropology, University of
Michigan, Ann Arbor.

Asch, N.B. and D.L. Asch

I978 The economic potential of Iva annua and its prehistoric importance in the lower
Illinois Valley. In The Nature and Status of Ethnobotany, edited by R.I. Ford, pp.
301-34.] . Anthropological Papers Vol. 67. Museum of Anthropology, University of

Michigan, Ann Arbor.

Asch Sidell, N. .
1999 Prehistoric Plant Use in Maine: Paleoindian to Contact Penod. In Current

Northeast Paleoethnobotany, edited by J. Hart, pp. 191-223. New York State
Museum, Albany. '

Barbour, M.G., J.H. Burk, W.D. Pitts, F.S. Gilliam, M.W. Schvvartz (eds)
1999 Terrestrial Plant Ecology, 3’“ edition. Benjamin/Cummings, Menlo Park, CA.

Barnes, B.V. and W.H. Wagner .
I981 Mchigan Trees: A Guide to the Trees of Michigan and the Great Lakes Region.

University of Michigan Press, Ann Arbor.

221

Barton, H. and R. F ullagar

2006 Microscopy. In Ancient Starch Research, edited by R. Torrence and H. Barton, pp.
47-52. Left Coast Press, Inc., Walnut Creek, CA.

Barton, H. and PJ. Matthews
2006 Taphonomy. In Ancient Starch Research, edited by R. Torrence and H. Barton, pp.
75-94. Left Coast Press, Inc., Walnut Creek, CA.

Beld, S.G.

1993a Site ZOSA665: A Late Woodland Site in Taymouth Township, Saginaw County,
Michigan. The Michigan Archaeologist 39(2):47-103.

l993b Site 201A37 (Arthursburg Hill Earthworks), Lyons Township, Ionia County,
Michigan. In Lyons Township Archaeological Survey, S-92-313, edited by S. Beld,
pp. 3-82. Office of the State Archaeologist, Lansing, Michigan.

Bender, B.

1978 Gatherer-Hunter to Farmer: A Social Perspective. World Archaeology lO(2):204-
222.

1981 Gatherer-Hunter Intensification. In Economic Archaeology: Towards and
Integration of Ecological and Social Approaches, edited by A. Sheridan and G.
Bailey, pp. 149-57. BAR International Series 96. British Archaeological Reports,
Oxford.

Bemabo, J .C. .
1981 Quantitative Estimates of Temperature Changes over the Last 2700 Years m

Michigan Based on Pollen Data Quaternary Research 15:143-159.

Bigony, B.
1970 Late Woodland Occupations of the Saginaw Valley. The Michigan Archaeologist
16(3-4):115-214.

Binford, L.R. .
1965 Archaeological Systematics and the Study of Cultural Process. American

Antiquity 3 l(2):203-210.

Booth, R.K., M. Notaro, S.T. Jackson, & J.E. Kutzbach . .
2006 Widespread drought episodes in the western Great Lakes region dunng the past
2000 years: geographic extent and potential mechanisms. Earth and Planetary

Science Letters 242:415-427.

Boyd, M.
2008 Personal Communication.

Boyd, M. and C. Surette

2010 Northemmost Precontact Maize in North America. American Antiquity 75(1):]17—
133.

222

Boyd, M., T. Vamey, C. Surette and .I. Surette
2008 Reassessing the northern limit of maize consumption in North America: stable

isotope, plant microfossil, and trace element content of carbonized food residue.
Journal of Archaeological Science 35:2 545—2556.

Brashler, LG
1973 A Formal Analysis of Prehistoric Ceramics fiom the Fletcher Site. Unpublished

Master's Thesis, Department of Anthropology, Michigan State University, East

Lansing.
I978 Boundaries and Interaction in the Early Late Woodland of Southern Lower

Michigan. Unpublished PhD Thesis, Department of Anthropology, Michigan State

University, East Lansing.
Early Late Woodland Boundaries and Interaction: Indian Ceramics of Southern

Lower Michigan. Museum Anthropological Series Vol. 3, Michigan State
University, East Lansing.

I981

Brashler, J.G, E.B. Garland, M.B. Holman, W.A. Lovis, and SR. Martin

2000 Adaptive Strategies and Socioeconomic Systems in Northem Great Lakes
Riverine Environments: The Late Woodland of Michigan. In Late Woodland
Societies: Tradition and Transformation across the Midcontinent, edited by
T.E. Emerson, D.L. McElrath, and AC. F ortier, pp. 543-582. University of

Nebraska Press, Lincoln.
Brashler, J.G., J.R. Halsey, M.B. Holman, JJ. Krakker, S.R. Martin, D.M. Stothers, and

RI. Zurel
I999 The Late Woodland: Prehistory's Finale, History's Prelude. In Retrieving

Michigan's Buried Past, edited by JR. Halsey, pp.l93-252. Cranbrook Institute of
Science, Bloomfield Hills, MI.

Brashler, .I.G, MJ. Hambacher, T]. Martin, K.E. Parker, and J.A. Robertson

2006 Middle Woodland Occupation in the Grand River Basin of Michigan. In,
Recreating Hopewell, edited by D.K. Charles and J.E. Buikstra, pp.26I-284.

University Press of Florida, Gainesville.

Brashler, .I.G. and MB. Holman

1985 Late Woodland Continuity and Change in the Saginaw Valley of Michigan. Arctic
Anthropology 22(2): 141-152.

Brashler, J.G., M.R. Laidler, and T]. Martin
1998 The Prison Farm Site (201A5 8): A Woodland Occupation in the Grand River

Basin of Michigan. Midcontinental Journal of Archaeology 23(2):143-197.

Brashler, LG and B. Mead
1996 Woodland Settlement in the Grand River Basin. In Investigating the

Archaeological Record of the Great Lakes State, edited by M.B Holman, J.G.

223

Brashler, and KB Parker, pp.181-250. New Issues Press, Kalamazoo, MI.

Braun, D.
1987 Co-evolution of sedentism, pottery technology, and horticulture in the central

Midwest, 200 BC. - AD. 600. In Emergent Horticultural Economies of the
Eastern Woodlands, edited by WP. Keegan, pp. 153-181. Occasional Paper No. 7,
Center for Archaeological Investigations, Southern Illinois University,

Carbondale.

Brose, D.S.
1972 The Mollusc Fauna. In The Schultz Site at Green Point: A Stratified Occupation

Area in the Saginaw Valley of Michigan, edited by J. Fitting, pp] 17-130. Memoirs
4. Museum of Anthropology, University of Michigan, Ann Arbor.

Brose, D.S. and N. Greber (eds)
1979 Hopewell Archaeology: The Chillicothe Conference. Kent State University Press,

Kent, OH.

Brown, J.A. and R. Sasso
2001 Prelude to History on the Eastern Prairies. In Societies in Eclipse Archaeology of

the Eastern Woodlands Indians, AD. I 4 00-1 700, edited by David S. Brose, C.
Wesley Cowan, and Robert C. Mainfort, Jr., pp. 205-228. Smithsonian Institution

Press, Washington, DC.

Bryant, V.M., Jr., and S. Hall

1993 Archaeological Palynology in the Unites States: A Critique. American Antiquity
58(2):277-286.

Bryant, V.M., Jr., and R.G Holloway
I983 The role of palynology in archaeology. In Advances in Archaeological Method

and Theory, Vol. 6, edited by MS. Schiffer, pp. 191-223. Academic Press, New
York.

Buikstra, J.
1992 Diet and Disease in Late Prehistory. In Disease and Demography in the Americas,

edited by J.W. Verano and DH. Ubelarker, pp. 87-101. Smithsonian Institution
Press, Washington, DC.

Bushnell, In, DJ.
1909 The Choctaw of Bayou Lecomb, St. Tammany Parish, Louisiana. Bulletin 48,

Bureau of American Ethnology. Washington, DC.

Callendar, C.
1978 Sauk. In Northeast, edited by B.G. Trigger, pp. 648-655. Handbook of North

American Indians, Vol. 15. Smithsonian Institution Press, Washington, DC.

224

 

.b'j'.m.<'..ll-m' ll: v. in!

Cashdan, EA.
1985 Coping with Risk: Reciprocity Among the Baswara of Northern Botswana. Man

20(3):454-474.

Chapman, J. and GD. Crites
1987 Evidence for Early Maize (Zea mays) from the Icehouse Bottom site, Tennessee.

American Antiquity 52(2):352-3 54.

Chapman, J. and P.J. Watson
1993 The Archaic Period and the Flotation Revolution. In Foraging and Farming in the

Eastern Woodlands, edited by CM. Scarry, pp. 27-38. The University Press of
Florida, Gainesville.

Charles, D.K. and IE Buikstra
2006 Recreating Hopewell. University Press of Florida, Gainesville.

Cheruvelil, J.
2010 Personal Communication.

Chilton, E.
2002 “towns they have none”: Diverse Subsistence and Settlement Strategies in Native

New England. In Northeast Subsistence-Settlement Change AD 700-1300, edited
by J .P. Hart and CB. Reith, pp. 289-300. Bulletin 496. New York State Museum,

New York State Education Department, Albany.

Clark, R.
1989 Towards the integration of social and ecological approaches to the study of early

agriculture. In The Beginnings of Agriculture, edited by A. Milles, D. Williams,
and N. Gardner, pp. 3-22. Series 496. British Archaeological Reports, Oxford.

Cleland, CE.
1966 The Prehistoric Animal Ecology and Ethnozoologr of the Upper Great Lakes

Region. Anthropological Papers No. 29. Museum of Anthropology, Unrversrty of
Michigan, Ann Arbor.

Cleland, N.N.
1993 A Preliminary Analysis of the Botanical Remains from Sites 20BY77 and

ZOBY79, Bay City, Michigan. In The Archaic, Woodland and Historic Period

Occupations of the Liberty Bridge Locale, edited by W. A. Lovis, pp. 165- 172.
Michigan Cultural Resource Investigation Series Vol. 3. Michigan Departments of

State and Transportation, Lansing.

Clifion, J .A.
1978 Potawatomi. In Northeast, edited by B.G. Trigger, pp. 725-742. Handbook of

North American Indians, Vol. 15. Smithsonian Institution Press, Washington, DC.

225

Canard, N., D.L. Asch, N.B. Asch, D. Elmore, H. Gove, M. Rubin, J.A. Brown, MD.

Wiant, K.B. Famsworth and TO. Cook
1984 Accelerator radiocarbon dating of evidence for prehistoric horticulture in Illinois.

Nature 308:443—446.

Cortella, A. R. and M. L. Pochettino
I994 Starch Grain Analysis as a Microsc0pic Diagnostic Feature in the Identification of

Plant Material. Economic Botany 48(2): 171-181 .

Cowan, FL.
2006 A Mobile Hopewell? Questioning Assumptions of Ohio Hopewell Sedentism. In

Recreating Hopewell, edited by D.K. Charles and J .E. Buikstra, pp.26-49.
University Press of Florida, Gainesville.

Cowan, CW. and P.J. Watson (eds)
1992 The Origins of Agriculture: An International Perspective. Smithsonian Institution

Press, Washington, DC.

Cowan, C. W. and P.J. Watson
1992 Some Concluding Remarks. In The Origins of Agriculture: An International

Perspective, edited by CW. Cowan and P.J. Watson, pp.207-212. Smithsonian
Institution Press, Washington, DC.

Crawford, GW., D. Saunders, and DC. Smith
2006 Pre-contact Maize from Ontario, Canada. In Histories of Maize, edited by J.E.

Staller, R.H. Tykot, and BE Benz, pp.549-559. Academic Press, London.

Crawford, GW. and D.G. Smith
2003 Paleoethnobotany in the Northeast. In People and Plants in Ancient Eastern

North America, edited by RE. Minnis, pp.l72-257. Smithsonian Institution Press,
Washington, DC.

Crawford, G.W., D.G. Smith, V.E. Bowyer
1997 Dating the Entry of Com (Zea Mays) into the Lower Great Lakes Region,

American Antiquity 62(1):l 12-1 19.

Crites, (1D.
1991 Investigations into Early Plant Domesticates and Food Production in Middle

Tennessee: A Status Report. Tennessee Anthropologist 16:69-87.
Domesticated Sunflower in Fifth Millennium B.P. Temporal Context: New

1993
Evidence from Middle Tennessee. American Antiquity 58(1): 146-148.

Crowley, T.J.
2000 Causes of Climate Change Over the Past 1000 Years. Science 289:270-277.

226

Crumley, CL.
1973 The Kantzler Site (20BY30): A Multi-Component Woodland Site in Bay City,
Michigan. The Illichigan Archaeologist I9(3-4):183-291.

Cummings, LS.
2006 Poverty Point Objects (Box 9.3). In Ancient Starch Research, edited by R.
Torrence and H. Barton, p. 183-184. Left Coast Press, Inc., Walnut Creek, CA.

Cushman, K.A.

1982 F Ioral Remains from Meadowcrofi Rockshelter, Washington County,
Southwestern Pennsylvania. In Meadowcrofl: Collected Papers on the
Archaeology of Meadowcrofl Rockshelter and the Cross Creek Drainage, edited
by RC. Carlisle and J .M. Adovasio, pp. 207-220. Department of Anthropology,
University of Pittsburgh, Pennsylvania.

Dancey, W.S. and P.J. Pacheco (eds.)
1997 Ohio Hopewell: Community Organization. Kent State University Press, Kent, OH.

del Pilar Babot, M.

2006 Damage on Starch from Processing Andean Food Plants (Box 4.4). In Ancient
Starch Research, edited by R. Torrance and H. Barton, p.66-67. Lefi Coast Press,
Inc., Walnut Creek, CA.

Dellar, D.B., C. J. Ellis and I. Kenyon

1986 Archaeology of the Southeastern Huron Basin Area. In Studies in Southwestern
Ontario Archaeology, edited by W. Fox, pp. 2-12. Occasional Publications of the
London Chapter, Ontario Archaeological Society No. 1. London, Ontario, Canada.

Denham, T., J. Iriarte, and L. Vrydaghs (eds)
2007 Rethinking Agriculture: Archaeological and Ethnoarchaeological Perspectives.
Left Coast Press, Inc., Walnut Creek, CA.

DeNiro, M.J.
1987 Stable Isotopy and Archaeology. American Scientist 75:182-191.

DeNiro, M.J. and ML. Schoeninger

1983 Stable Carbon and Nitrogen Isotopes of Bone Collagen: Variation Within
Individuals, Between Sexes and Within P0pulations Raised on Monotonous Diets.
Journal of Archaeological Science 10:199-203.

Densmore, F. ‘
I929 Chippewa Customs. Bulletin No. 86. Bureau of American Ethnology, Smithsonian
Institution, Washington, DC.

Dice, LR.
1943 The Biotic Provinces of North America. University of Michigan Press, Ann Arbor.

227

Dimbleby, GW.
1985 The Palynology of Archaeological Sites. Academic Press, London.

Doolittle, W.E.
2000 Cultivated Landscapes of Native North America, pp. 21-216; 428—452. Oxford
University Press, Oxford.

Dorr, J.A. and DP. Eschman
1970 Geology of Michigan. University of Michigan Press, Ann Arbor.

Egan, KC.

1990 Archaeobotanical Remains from 208A620. In The Bridgeport Township Site:
Archaeological Investigation at 20SA 620, Saginaw County, Michigan, edited by
John M. O’Shea and Michael Schott, pp. 203-231. Anthropological Papers No. 81 .
Museum of Anthropology, University of Michigan, Ann Arbor.

1993 Hunter-Gatherer Subsistence Adaptation in the Saginaw Valley, Michigan. Vol.
I&II. PhD Dissertation, Michigan State University, East Lansing. University
Microfilms, Ann Arbor.

Feest, J.E. and CF. F eest
1978 Ottawa. In Northeast, edited by B.G Trigger, pp. 772-786. Handbook of North
American Indians, Vol. 15. Smithsonian Institute Press, Washington, DC.

Fie, SM.
2006 Visiting in the Interaction Sphere: Ceramic Exchange and Interaction in the Lower

Illinois Valley. In Recreating Hopewell, edited by D.K. Charles and J.E. Buikstra,
pp.427-445. University Press of Florida, Gainesville.

Fischer, F. W.
1972 Schultz Site Ceramics. In The Schultz Site at Green Point: A Stratified
Occupation Area in the Saginaw Valley of Michigan, edited by J .E. Fitting, pp.

137-190. Memoirs 4. Museum of Anthropology, University of Michigan, Ann
Arbor.

Fitting, J.E. (ed.)
1972 The Schultz Site at Green Point: A Stratified Occupation Area in the Saginaw

Valley of Michigan. Memoirs 4. Museum of Anthropology, University of
Michigan, Ann Arbor.

Fitting, J.E.

1969 Late Prehistoric Settlement Patterns in the Upper Great Lakes. Ethnohistory
16(4):289-302.

I972a Excavations at the Schultz Site. In The Schultz Site at Green Point: A Stratified
Occupation Area in the Saginaw Valley of Michigan, edited by J.E. Fitting, pp.5-
42. Memoirs 4. Museum of Anthropology, University of Michigan, Ann Arbor.

228

1972b Lithic Industries at the Schultz Site. In The Schultz Site at Green Point: A
Stratified Occupation Area in the Saginaw Valley of Alichigan, edited by J .E.
Fitting , pp. 191-224. Memoirs 4. Museum of Anthropology, University of
Michigan, Ann Arbor.

1972c Occupational History. In The Schultz Site at Green Point: A Stratified Occupation
Area in the Saginaw Valley of Michigan, edited by J .E. Fitting, pp.253-266.
Memoirs 4. Museum of Anthropology, University of Michigan, Ann Arbor.

1975 The Archaeology of Michigan: A Guide to the Prehistory of the Great Lakes
Region, 2"d edition. Cranbrook Institute of Science, Bloomfield Hills, MI.

Ford, R.
1985 The Processes of Plant Production in Prehistoric North America. In Prehistoric

Food Production in North America, edited by R.I. Ford, pp. 1-18. Anthropological
Paper No. 75. University of Michigan, Museum of Anthropology, Ann Arbor.

Fowler, ML.
1957 The Origin ofPlant Cultivation in the Central Mississippi Valley: A ijothesis.

Paper presented at the annual meeting of the American Anthropological

Association, Chicago.
1971 The Origin of Plant Cultivation in the Central Mississippi Valley: A Hypothesis.

In Prehistoric Agriculture, edited by Stuart Struever, pp. 122-128. Natural History
Press, Garden City, New York.

Fox, W.A.
I990 The Middle Woodland to Late Woodland Transition. In The Archaeology of

Southern Ontario to AD. 1650, pp.17 I -188. Occasional Publications of the
London Chapter, Ontario Archaeological Society no.5. London.

Franz, P.J.
I993 The Faunal Assemblage from Sites ZOBYTT and 20BY79, Bay City, Michigan. In

The Archaic, Woodland and Historic Period Occupations of the Liberty Bridge
Locale, edited by W.A. Lovis, pp. 165-172. Michigan Cultural Resource
Investigation Series Vol. 3. Michigan Departments of State and Transportation,

Lansing.

Fritz, GJ.
1990 Multiple Pathways to Farming in Precontact Eastern North America Journal of

World Prehistory 4:38 7—435.
1993 Early and Middle Woodland Period Paleoethnobotany. In Foraging and Farming

in the Eastern Woodlands, edited by CM. Scarry, pp. 39-56. Ripley P. Bullen
Monographs in Anthropology and History, University of Florida Press, ‘

Gainesville.
1999 Gender and the Early Cultivation of Gourds in Eastern North America. American

Antiquity 64(3):417-429.

229

Fullagar, R., J. Field, T. Denham and C. Lentfer

2006 Early and mid Holocene tool-use and processing of taro (Colocasia esculenta),
yam (Dioscorea sp.) and other plants at Kuk Swamp in the highlands of Papua
New Guinea. Journal of Archaeological Science 33(5):595-614.

Gallagher, J.P., R.F. Boszhardt, R.F. Sasso, and K. Stevenson
1985 Oneota Ridged Field Agriculture in Southwestern Wisconsin. American Antiquity

50(3):605—612.

Gallagher, J .P.
1989 Agricultural intensification and ridged-field cultivation in the prehistoric upper

Midwest of North America. In Foraging and Farming: The evolution of plant
exploitation, edited by D. Harris and G. Hillman, pp. 572-584. Unwin Hyman,

London.

Gardner, PS.
1987 New Evidence Concerning the Chronology and Paleoethnobotany of Salts Cave,

Kentucky. American Antiquity 52(2):358-367.

Garland, EB. and S.G Beld
1999 The Early Woodland: Ceramics, Domesticated Plants, and Burial Mounds F oretell

the Shape of the Future. In Retrieving Michigan ’3 Buried Past, edited by JR.
Halsey, pp.125-l46. Cranbrook Institute of Science, Bloomfield Hills, MI.

Garland, EB. and AL. DesJardins

I995 The Strobel Site (ZOSJ180), A Havana Middle Woodland Encampment on the
Prairie River in Southwestern Michigan. The Michigan Archaeologist 41(1):I-40.

2006 Between Goodall and Norton: Middle Woodland Settlement Patterns and
Interaction Networks in Southwestern Michigan. In Recreating Hopewell,
D.K. Charles and J .E. Buikstra, pp.227—260. University Press of Florida,

Gainesville.

Gamier, W.G.
1999 Late Woodland Landscapes of Wisconsin: Ridged Fields, Effigy Mounds and

Territoriality. Antiquity 73: 671-683.

Gilmore, MR.
1931 Vegetal Remains of the Ozark Bluff-Dweller Culture. Papers of the Michigan

Academy of Science, Arts, and Letters 14183-102.

Greenlee, D.M. -
2006 Dietary Variation and Prehistoric Maize Farming in the Middle Ohio Valley. In
Histories of Maize, edited by J.E. Staller, R.H. Tykot, and BF. Benz, pp.215-233.

Academic Press, London.

230

Gremillion, KJ.
2003 Eastern Woodlands Overview. In People and Plants in Ancient Eastern North

America, edited by RE. Minnis, pp. 1 7—49. Smithsonian Institution Press,
Washington, DC.

Gremillion, KI. and K.D. Sobolik

1996 Dietary variability among prehistoric forager-farmers of eastern North America.
Current Anthropology 37:529-539.

Gott, B.

2006 How Typha Starch Enters the Archaeological Record (Box 4.3). In Ancient Starch

Research, edited by R. Torrence and H. Barton, pp. 62-64. Left Coast Press, Inc.,
Walnut Creek, CA.

Gott, B., H. Barton, D. Samuel, and R. Torrence

2006 Biology of Starch. In Ancient Starch Research, edited by R. Torrence and H.
Barton, pp. 3546. Left Coast Press Inc., Walnut Creek, CA.

Hall, R.

1976 Ghosts, water barriers, corn, and sacred enclosures in the Eastern Woodlands.
American Antiquity 4 l (3):359—364

Hall, S.A.

I981 Deteriorated pollen grains and the interpretation of Quaternary pollen diagrams.
Review of Paleobotany and Palynology 32:193-206.

Halsey, J.R. (ed.)

1999 Retrieving Michigan's Buried Past. Cranbrook Institute of Science, Bloomfield
Hills, MI.

Hammett, J.E.

I997 Interregional Patterns of Land Use and Plant Management in Native North

America. In People, Plants, and Landscapes: Studies in Paleoethnobotany,

edited by KJ. Gremillion, pp. 195-216. The University of Alabama Press,
Tuscaloosa.

Hanson, P.R., A.E. Arbogast, W.C. Johnson, RM. Joeckel, A.R. Young
2009 Megadroughts and late Holocene dune activation at the eastern margin of the
Great Plains, north-central Kansas, USA. Aeolian Research 1(3-4):101-110.

Harlan, J. R., J.M.] de Wet, and E.G. Price
1973 Comparative Evolution of the Cereals. Evolution 27:311-325.

Harmon, JR.

2009 Plant Geography. In Michigan Geography and Geology, edited by RJ. Schaetzl,
LT. Darden, and D. Brandt, pp. 330-345. Pearson Custom Publishing, Boston.

231

Harris, DR.

1989 An Evolutionary Continuum of People-Plant Interaction. In Foraging and

Farming: The Evolution of Plant Exploitation, edited by DR. Harris and GC.
Hillman, pp. 11-26. Unwin Hyman Ltd., London.

,Domesticatory Relationships of People, Plants, and Animals. In Redefining

Nature: Ecology, Culture, and Domestication, edited by R. Ellen and K. F ukui,
pp. 437-463. Oxford University Press, Oxford.

Agriculture, Cultivation and Domestication: Exploring the Conceptual

Framework of Early Food Production. In Rethinking Agriculture: Archaeological
and Ethnoarchaeological Perspectives, edited by T. Denham, J. Iriarte, and L.
Vrydaghs (eds.), pp. 16-35. Left Coast Press, Inc., Walnut Creek, CA.

Hart, J .P.

1999 Maize Agriculture Evolution in the Eastern Woodlands of North America: A

Darwinian Perspective. Journal of Archaeological Method and Theory 6(2): 137-
180.

1996

2007

Hart, J .P. and N. Asch Sidell

I997 Additional Evidence for Early Cucurbit Use in the Northern Eastern Woodlands
East of the Allegheny Front. American Antiquity 62(3):523-53 7.

Hart, JP. and R.G Matson

2009 The use of multiple discriminant analysis in classifying prehistoric phytolith
assemblages recovered from cooking residues. Journal of Archaeological Science

36(1):?4-83.
Hart, .I.P., R.G. Thompson, and HI. Brumbach

2003 Phytolith Evidence for Early Maize (Zea Mays) in the Northern Finger Lakes
Region of New York. American Antiquity 68(4):619-640.

Hart, J.P., W.A. Lovis, J.K. Schulenberg, and GR. Urquhart

2007 Paleodietary Implications from Stable Carbon Isotope Analysis of Experimental
Cooking Residues. Journal of Archaeological Science 34(5):804-813.

Hart J.P., GR. Urquhart, R. Feranac, and W.A. Lovis

2009 Non-linear relationship between bulk 813C and percent maize in carbonized

cooking residues and the potential of false-negatives in detecting maize. Journal
of A rchaeolo gical Science 36(10):2206-2212.

Haslam, M.

2004 The Decomposition of Starch Grains in Soils: Implications for Archaeological
Residue Analyses. Journal of A rchaeological Science 31:1715-1734.

Hastorf, CA. and M. J. DeNiro

1985 Reconstruction of prehistoric plant production and cooking practices by a new
isotopic method. Nature 315:489-491.

232

Hayden, B.

1995 A New Overview of Domestication. In Last Hunters, First Farmers: New

Perspectives on the Prehistoric Transition to Agriculture, edited by T.D. Price and

AB. Gebauer, pp. 273-299. School of American Research, Santa Fe, New
Mexico.

Hegrnon, M.
1991 The Risks of Sharing and Sharing as Risk Reduction: Inter-Household Food

Sharing in Egalitarian Societies. In Between Bands and States, edited by S.A.

Gregg, pp. 309-329. Occasional Paper No. 9, Center for Archaeological
Investigations, Southern Illinois University, Carbondale.

Heron, C. and R. Evershed

1993 The Analysis of Organic Residues and the Study of Pottery Use. In

Archaeological Method and Theory, edited by M.B. Schiffer, pp. 247-284.
The University of Arizona Press, Tuscon.

Hinsdale, W.B.
I931 Archaeological Atlas of Michigan. University of Michigan Press, Ann Arbor.

Holman, M.B., J .G Brashler, and K.E. Parker (eds)

1996 Investigating the Archaeological Record of the Great Lakes State. New Issues
Press, Kalamazoo, MI.

Holman, MB. and R.G. Kingsley

I996 Territoriality and Societal Interaction During the Early Late Woodland Period in
Southern Michigan. In Investigating the Archaeological Record of the Great

Lakes State, edited by M.B. Holman, J.G Brashler, and K.E. Parker, pp. 341-382.
New Issues Press, Kalamazoo, MI.

Holst, 1., J.E. Moreno, and DR. Pipemo
2007 Identification of teosinte, maize, and Tripsacum in Mesoamerica by using pollen,

starch grains and phytoliths. Proceedings of the National Academy of Science
104(45):I7608-176l3.

Howey, M.

2006 Ritual, Resources, and Regional Organization in the Upper Great Lakes, AD
1200-I600. PhD Dissertation, University of Michigan, Ann Arbor.
2007

Using multi-criteria cost surface analysis to explore past regional landscapes: a

case study of ritual activity and social interaction in Michigan, AD 1200-1600.
Journal of Archaeological Science 34: 1 830-1846.

Howey, M. and J.M. O'Shea
2006 Bear's Journey and the Study of Ritual in Archaeology. American Antiquity

233

7 I (2)1261 -282.

Hupy, CM. and CH. Yansa
2009 Late Holocene Vegetation History of the Forest Tension Zone in Central Lower
Michigan, USA. Physical Geography 30(3):205-235.

Jefferies, R.W. and BM. Butler (eds)
1982 The Carrier Mills Archaeological Project: Human Adaptation in the Saline

Valley, Illinois. Research Paper No. 33, Center for Archaeological Investigations,
Southern Illinois University, Carbondale.

Jones, V.H.

1936 The vegetal remains of Newt Kash Hollow shelter. In Rock Shelters in Menifee
County, Kentucky, edited by W.S. Webb and W.D. F unkhouser. University of
Kentucky Reports in Archaeology and Anthropology 3(4):]47-167.

Justice, ND.

1987 Stone Age Spear and Arrow Points of the Midcontinental and Eastern United
States. Indiana University Press, Bloomington.

Kapp, RD.
1969 How to Know Pollen and Spores. Brown, Dubuque, IA.
1999 Michigan Late Pleistocene, Holocene, and Presettlement Vegetation and Climate.

In Retrieving Michigan '3 Buried Past, edited by J .R. Halsey, pp. 31-58. Cranbrook
Institute of Science, Bloomfield Hills, MI.

Katzenberg, MA.

2006 Prehistoric Maize in Southern Ontario. In Histories of Maize, edited by J.E.
Staller, R.H. Tykot, and BF. Benz, pp. 263-273. Academic Press, London.

Katzenberg, M.A., H.P. Schwarcz, M. Knyf, and J. Melbye
1995 Stable Isotope Evidence for Maize Horticulture and Paleodiet in Southern
Ontario, Canada. American Antiquity 60(2):335-350.

Keeley, L.H.

1995 Protoagricultural Practices among Hunter-Gatherers: A Cross-Cultural Survey. In
Last Hunters First Farmers, edited by T.D. Price and AB. Gebauer, pp. 243-272.
School of American Research Press, Santa Fe, NM.

Kelly, R.L.

I995 The Foraging Spectrum: Diversity in Hunter-Gatherer Lifeways. Smithsonian
Institution Press, Washington.

Ketchum, 8., MR. Schurr, and RC. Gamiewicz

2009 ATest for Maize Consumption by Fauna in Late Prehistoric Eastern North
America. North American Archaeologist 30(1): 87-101.

234

Kidder, T.
1992 Timing and Consequences of the Introduction of Maize Agriculture in the Lower

Mississippi Valley. North American Archaeologist I 3: 15-41 .

King, RB.
1985 Early cultivated cucurbits in eastern North America In Prehistoric Food

Production in North America, edited by R.I. Ford, pp. 73-97. Anthropological
Papers No. 75. Museum of Anthropology, University of Michigan, Ann Arbor.

Kingsley, R.G, D.S. Brose, and M.J. Hambacher
1999 The Middle Woodland: A Golden Age of Mound Builders and F isherman. In
Retrieving Michigan's Buried Past, edited by JR. Halsey, pp. 147-192. Cranbrook

Institute of Science, Bloomfield Hills, MI.

Kinietz, W.V.
1940 The Indians of the Western Great Lakes 1615-1 760. Occasional Contributions No.

10. Museum of Anthropology, University of Michigan, Ann Arbor.

Knauft, B.
2010 The Gebusi: Lives Transformed in a Rainforest World, 2'” edition. McGraw Hill,

New York.

Krakker, J.J.
1983 Changing Sociocultural Systems During the Late Prehistoric Period in Southeast

Michigan. Vol. I&II. PhD Dissertation, University of Michigan, Ann Arbor.
University M icrofilms, Ann Arbor.

Larsen, CE.
1985 Lake Level, Uplift and Outlet Incision: The Nipissing and Algoma Great Lakes. In

Quaternary Evolution of the Great Lakes, edited by P. F. Karrow and P. E. Calkin,
pp. 63-7 7. Special Paper 30. Geological Society of Canada, St. John‘s,

Newfoundland.

Lee, RB.
2003 The Dobe Ju/ ’hoansi. Wadsworth, Toronto.

Lovis, W.A. (ed.)
1993 The Archaic, Woodland and Historic Period Occupations of the Liberty Bridge

Locale. Michigan Cultural Resource Investigation Series Vol. 3. Michigan
Departments of State and Transportation, Lansing. -

Lovis, W.A.
1985 Seasonal Settlement Dynamics and the Role of the Fletcher Site in the Woodland

Adaptations of the Saginaw Drainage Basin. Arctic Anthropology 22(2):153-170.
1990 Curatorial Considerations for Systematic Research Collections: AMS Dating a

235

Curated Ceramic Assemblage. American Antiquity 55(2):382-387.

1993a Background of Investigations at Sites ZOBYTT, ZOBY78, and ZOBY79, Bay City,
Michigan. In The Archaic, Woodland and Historic Period Occupations of the
Liberty Bridge Locale, edited by W.A. Lovis, pp. 1-19. Michigan Cultural
Resource Investigation Series Vol. 3. Michigan Departments of State and
Transportation, Lansing.

I993b Features at Sites 208Y77 and ZOBY79, Bay City, Michigan. In The Archaic,
Woodland and Historic Period Occupations of the Liberty Bridge Locale, edited
by W.A. Lovis, pp. 185-224. Michigan Cultural Resource Investigation Series Vol.
3, Michigan Departments of State and Transportation, Lansing.

I993c Discussion and Conclusions. In The Archaic, Woodland and Historic Period

Occupations of the Liberty Bridge Locale, edited by W.A. Lovis, pp. 225-229.

Michigan Cultural Resource Investigation Series Vol. 3. Michigan Departments of

State and Transportation, Lansing.

Clay effigy representations of the bear and Mishipishu: Algonquian iconography

from the Late Woodland Johnson site, northern lower Michigan. Midcontinental

Journal of Archaeology 26(1): 105-1 19.

2001

Lovis, W.A., A. Arbogast, and GW. Monaghan
2009 Geoarchaeology and Taphonomy of Lake Michigan Coastal Dunes: Activation,

Stabilization, Cycling and Archaeological Site Formation Processes. Report
submitted to the Environmental Section, Michigan Department of Transportation

(Contract IDS 2006-0034-ZI). Michigan State University, East Lansing.

Lovis, W.A. and E. Bogdan-Lovis
2004 Ridged Fields, Catastrophic Wet Season Planting, and Germination Rates: Notes

from Libby's Garden. The Wisconsin Archeologist 85(1):78—83.

Lovis, W.A. and RM. Davis
1993 The Ceramic Assemblage from Sites 2OBY77 and 20BY79, Bay City, Michigan.

In The Archaic, Woodland and Historic Period Occupations of the Liberty Bridge
Locale, edited by W.A. Lovis, pp. 98-123. Michigan Cultural Resource
Investigation Series Vol. 3. Michigan Departments of State and Transportation,

Lansing.

Lovis, W.A., K.C. Egan, GW. Monaghan, B.A. Smith, EJ. Prahl
1996 Environment and Subsistence at the Marquette Viaduct Locale of the Fletcher

Site. In Investigating the Archaeological Record of the Great Lakes State, edited
by M.B. Holman, J .G Brashler, and K.E. Parker, pp. 251-306. New Issues Press,

Kalamazoo, MI.
Lovis, W.A. K.C. Egan-Bruhy, B.A. Smith, and GW. Monaghan

2001 Wetlands and Emergent Horticultural Economies in the Upper Great Lakes: A
New Perspective from the Schultz Site. American Antiquity 66(4):615—632.

236

Lovis, W.A. and GW. Monaghan
2008 Chronology and Evolution of the Green Point Floodplain and Associated

Cucurbita pepo. In Current Research in Northeast Paleoethnobotany I], edited by
J. P. Hart, pp. 141-150. Bulletin 512. New York State Museum, Albany.

Lu, T.
2006 The Survival of Starch Residues in a Subtropical Environment (Box 5.3). In
Ancient Starch Research, edited by Robin Torrence and Huw Barton, p. 80-81.

Left Coast Press, Inc., Walnut Creek, CA.

Mainfort, RC.
2001 The Late Prehistoric and Protohistoric Periods in the Central Mississippi Valley.
In Societies in Eclipse, edited by D.S. Brose, C.W. Cowan, and RC. Mainfort, Jr.,

pp. 173-190. Smithsonian Institution Press, Washington, DC.

Mangold, W.L. and MR. Schurr
2006 The Goodall Tradition: Recent Research and New Perspectives. In Recreating

Hopewell, edited by D.K. Charles and J .E. Buikstra, pp. 206-226. University Press
of Florida, Gainesville.

Maxwell, MS.
1952 The Archaeology of the Lower Ohio Valley. In Archaeology of Eastern United
States, edited by J.B. Griffin, pp. 176-189. University of Chicago Press, Chicago.

Messner, T.
2008 Woodland Period People and Plant Interactions: New Insights fiom Starch Grain
Analysis. PhD. Thesis, Department of Anthropology, Temple University,

Philadelphia, PA.

Messner, T. and R. Dickau
2005 New Directions, New Interpretations: Paleoethnobotany in the Upper Delaware
Valley and the Utility of Starch Grain Research in the Middle Atlantic. Journal of

.Middle Atlantic Archaeology 21 271-82.

Milner, CM.

1991 Localization in Small-Scale Societies: Late Prehistoric Social Organization in the
Western Great Lakes. In Between Bands and States, edited by S.A. Gregg, pp. 35-

57. Occasional Papers No. 9. Center for Archaeological Investigations, Southern

Illinois University, Carbondale.

Minnis, P.E. (ed.)
2003 People and Plants in Ancient Eastern North America. Smithsonian Institution
Press, Washington DC.

Mofl‘att, CR. and CM. Arzigian
2000 New Data on the Late Woodland use of Wild Rice in Northern Wisconsin.

237

Midcontinental Journal of A rehaeology 25 ( 1):49-8I .

Monaghan, GW. and WA Lovis
2005 Modeling Archaeological Site Burial tn Southem Michigan. Michigan State

University Press, East Lansing.

Monaghan, GW., W.A. Lovis, and KC. Egan-Bruhy
2006 Earliest Cucurbita from the Great Lakes, Northern USA. Quaternary Research

65 :2 16-222.

Morgan, R.G
1952 Outline of Cultures in the Ohio Region. In Archaeology of Eastern United States,
edited by J.B. Griffin, pp. 83-98. University of Chicago Press, Chicago.

Morton, JD. and HP. Schwarcz
2004 Palaeodietary implications from stable isotopic analysis of residues on prehistoric

Ontario ceramics. Journal of Archaeological Science 31 :503-517

Mulholland, SC.
1986 Classrf' cation of grass silica phytoliths. In Plant Opal Phytolith Analysis in

' r
Archaeology and Paleoecology, edited by I. Rovner, Vol. I, pp. 41-52. Occasronal
Papers of the Phytolitharien, North Carolina State University, Raleigh.
I989 Phytolith shape frequencies in North Dakota grasses: A comparison to general

patterns. Journal of Archaeological Science 16:489-511
2006 Personal Communication.

Mulholland, SC. and C. Prior
1993 AMS radiocarbon dating of phytoliths. In Current Research in Phytolith Analysis:
Applications in Archaeology and Paleoecology, edited by D.M. Pearsall and DR.

Pipemo, Vol. 10, pp. 21-23. MASCA Research Papers in Scrence and
Archaeology, University Museum of Archaeology and Anthropology, University

of Pennsylvania, Philadelphia.

Mulholland, SC. and G Rapp Jr.
1992 Phytolith systematics: An introduction. In Phytolith Systematics: Emerging Issues

edited by G Rapp Jr. and SC. Mulholland, pp. 1-13, Plenum, New York

Mulholland, S.C., G. Rapp Jr., and AL. Ollendorf
1988 Variation in Phytoliths from Corn Leaves. Canadian Journal of Botany 66.2001-

2008.
Murphy, C. and N. Ferris
1990 The Late Woodland Western Basin Tradition in Southwestern Ontario. In The

Archaeology of Southern Ontario to AD. 1650, pp. 189-278. Occasronal
Publications no.5 London Chapter, Ontario Archaeological Society, London, ON

238

Nassaney, M., J.A. Brandao, W.M. Cremin and B. Giordano
2007 Archaeological evidence of daily life at an 18th century outpost in the western

Great Lakes. Historical Archaeology 14(4): 1-1 7.

O’Leary, M.H.
1988 Carbon Isotopes in Photosynthesis. Bioscience 38(5):328-336.

O’Shea, J .M.

1981 Social Configurations and the Archaeological Study of Mortuary Practices: A
Case Study. In The Archaeology of Death, edited by R. Chapman, 1. Kinnes,
and K. Randsborg, pp. 39-52. Cambridge University Press, Cambridge.

1989 The role of wild resources in small-scale agricultural systems: tales from the
Lakes and the Plains. In Bad Year Economics: Cultural Responses to Risk and

Uncertainty, edited by P. Halstead and J. O’Shea, pp. 57-67. Cambridge

University Press, Cambridge.
2003 Inland F oragers and the adoption of maize agriculture in the upper Great Lakes of

North America. Before Farming: the archaeology and anthropology of hunter-
gatherers, 1(3): 1-21

Ozker, DB.

1982 An Early Woodland Community at the Schultz Site 20SA 2 in the Saginaw Valley

and the Nature of the Early Woodland Adaptation in the Great Lakes Region.
Anthropological Papers No. 70. Museum of AnthrOpoIogy, University of

Michigan, Ann Arbor.

Parker, K.E.
1996 Three Corn Kernels and a Hill of Beans: The Evidence for Prehistoric
Horticulture in Michigan. In Investigating the Archaeological Record of the Great

Lakes State, edited by ME. Holman, J.(1 Brashler, and K.E. Parker, pp. 307-340.

New Issues Press, Kalamazoo, MI.

Payne, J.
1994 A Preliminary Report of the Prehistory of the Tobico Marsh Area: The 1993 Field
Season. Manuscript on file, Office of the State Archaeologist, Lansing, Michigan.

Pearsall, D.M.

2000 Paleoethnobotany: A Handbook of Procedures, 2nd edition. Academic Press, San
Diego, CA.

Perkle, BE.

1998 Cucurbita pepo from King Coulee, southeastern Minnesota. American Antiquity

63:279—288.

Perdue, T.
1998 Cherokee Women: Gender and Culture Change, 1 700-1835. University of

239

Nebraska Press, Lincoln.

Perry, L.
2001 Prehispanic Subsistence in the Middle Orinoco Basin: Starch Analysis Yield New

Evidence. Ph.D. dissertation, Southern Illinois University.

2002a Starch Analyses Reveal Multiple Functions of Quartz "Manioc" Grater Flakes
Item the Orinoco Basin, Venezuela. Interciencia 27(11):635-640.

2002b Starch Granule Size and the Domestication of Manioc and Sweet Potato.

Economic Botany 56(4):335-349.

Perry, L., R. Dickau, S. Zarrillo, I. Holst, D.M. Pearsall, D.R. Pipemo, M.J. Berman, R.G
Cooke, K. Rademaker, A.J. Ranere, J .S. Raymond, D.H. Sandweiss, F. Scaramelli, K.

Tarble and J.A. Zeidler
2007 Starch Fossils and the Domestication and Dispersal of Chili Peppers (Capsicum

spp. L.) in the Americas. Science 315:986-989.

Pipemo, DR.
2006 Phytoliths: A Comprehensive Guide for Archaeologists and Paleoecologists.

Altamira Press, New York.

Pipemo, DR. 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 analyses. In
Encyclopedia of Plant and Crop Science, edited by R. M. Goodman, pp. 330-332.

Marcel Dekker, Monticello.

Pipemo, D.R., A.J. Ranere, I. Holst, and P. Hansell
2000 Starch grains reveal early root crop horticulture in the Panamanian tropical forest.

Nature 407:894-897.

Prahl, EJ.
1969 Preliminary Comparison of Three Prehistoric Sties in the Vicinity of the Western

Lake Erie Shore. Toledo Area Aboriginal Research Club Bulletin 1:32-62.

Prahl, E.J., D.S. Brose, and D.S. Stothers
1976 A Preliminary Synthesis of Late Prehistoric Phenomena in the Western Basin of

Lake Erie. In The Late Prehistory of the Lake Erie Drainage Basin, edited by
D.S. Brose, pp. 251-282. Cleveland Museum of Natural History, Cleveland.

Price, T.D. and AB. Gebauer (eds.)
1995 Last Hunters First Farmers. School of American Research Press, Santa Fe, NM.

Price, T.D. and AB. Gebauer
1995 New Perspectives on the Transition to Agriculture. In Last Hunters First Farmers,

240

edited by T.D. Price and AB. Gebauer pp. 3-20. School of American Research
Press, Santa Fe, NM.

Quimby, GI., Jr.
1941 Hopewellian Pottery Types in Michigan. Papers of the Michigan Academy of

Science, Arts, and Letters 26(4):489—94.
1943 The Ceramic Sequence within the Goodall Focus. Papers of the Michigan

Academy of Science, Arts, and Letters 28(4):543—48.
1944 Some New Data on the Goodall Focus. Papers of the Michigan Academy of

Science, Arts, and Letters 29(4):419—24.
1962 A Year with a Chippewa Family, 1763-1764. Ethnohistory 9(3):217-239.

Raviele, M.E. .
2007 Subsistence Transformation and Thermal Pretreatment of Lithic Raw Materlal.

Lithic Technology 32(1):99-l 14.

Raymond, J .S. and W.R. DeBoer
2006 Maize on the Move. In Histories of Maize: Multidisciplinary Approaches to the

Prehistory, Linguistics, Biogeography, Domestication, and Evolution of Maize,
edited by J. Staller, R. Tykot, and B. Benz, pp. 337-343. Academic Press, San

Diego, CA.

Reber, E. and R. Evershed .
2004 How did Mississippians prepare maize? The application of compound-specrfic

carbon isotope analysis to absorbed pottery residues from several Mississippi
Valley sites. Archaeometry 46: 19—33.

Redmond, B.G.
2007 Hopewell on the Sandusky: Analysis and Description of an Inundated Ohio

Hopewell Mortuary-Ceremonial Site in North-Central Ohio. North American
Archaeologist 28(3): 189-232.

Reimer, P.J., M.G.L. Baillie, E. Bard, A. Bayliss, J.W. Beck, C.J.H. Bertrand, P.G
Blackwell, C.E. Buck, GS. Burr, K.B. Culter, P.E. Damon, R.L. Edwards, R.G. Fairbanks,

M. Friedrich, T.P. Guilderson, A.G HOgg, K.A. Hughen, B. Kromer, G McCormac, S.
Manning, C.B. Ramsey, R.W. Reimer, S. Remmele, J .R. Southon, M. Stuiver, S. Talamo,

F.W. Taylor, J. van der Picht, C.E. Weyhenmeyer
2004 IntCalO4 Terrestrial Radiocarbon Age Calibration, 0-26 Cal Kyr BP. Radiocarbon

46: 1029-1058.
Riley, T.J., GR. Walz, C.J. Bareis, A.C. Fortier, K.E. Parker

1994 Accelerator Mass Spectrometry (AMS) Dates Confirm Early Zea Mays in the
Mississippi River Valley. American Antiquity 59(3):490-498.

Rindos, D.
1984 The Origins of A griculture: An Evolutionary Perspective. Academic Press,

241

London.

Robertson, J.A.
I987 Inter-assemblage variability and hunter-gatherer settlement systems: a

perspective fiom the Saginaw Valley of Michigan. PhD Dissertation, Michigan
State University, East Lansing. University Microfilms, Ann Arbor.

1993 The Lithic Assemblage from Sites 20BY77, ZOBY78, and ZOBY79, Bay City,
Michigan. In The Archaic, Woodland and Historic Period Occupations of the
Liberty Bridge Locale, edited by W.A. Lovis, pp.41-97. Michigan Cultural
Resource Investigation Series Vol. 3. Michigan Departments of State and

Transportation, Lansing.

Rogers, ES.
1994 The Algonquian Farmers of Southern Ontario, 1830-1945. In, Aboriginal Ontario:

historical perspectives on the first nations, edited by ES. Rogers and DB. Smith,
pp. 122-166. Dundum Press, Toronto, Ontario, Canada.

Rose, F.
2008 Intra-Community Variation in Diet during the Adoption of a New Staple Crop in

the Eastern Woodlands. American Antiquity 73(3):413-440.

Rosenberg, M.
1990 The Mother of Invention: Evolutionary Theory, Territoriality, and the Origins of

Agriculture, American Anthropologist 92(2):399-41 5.

Samuel, D.
2006 Modified Starch. In Ancient Starch Research, edited by R. Torrence and H.

Barton, pp. 205-216. Left Coast Press, Inc., Walnut Creek, CA.

Sasso, RF.
2003 Vestiges of Ancient Cultivation: The Antiquity of Garden Bends and Corn Hills in

Wisconsin. Midcontinental Journal of Archaeology 28(2):233-25 7.

Scarry, C.M. (ed.)
1993 Foraging and Farming in the Eastern Woodlands. University Press of Florida,

Gainesville.

Scarry, CM.
2003 Patterns of Wild Plant Utilization in the Prehistoric Eastern Woodlands. In People

and Plants in Ancient Eastern North America, edited by P. E. Minnis, pp. 50-1 04.
Smithsonian Institution Press, Washington DC.

Schoeninger, M., M.J. DeNiro, and H. Tauber
1983 Stable Nitrogen Isotope Rations of Bone Collagen Reflect Marine and Terrestrial

Components of Prehistoric Diet. Science 220(4604): 1381-1383.

242

Schoeninger, M. and K. Moore
1992 Bone Stable Isotope Studies in Archaeology. Journal of World Prehistory 6(2):
247-296.

Schurr, MR. and B.G Redmond
1991 Stable Isotope Analysis of Incipient Maize Horticulturalists from the Gard Island
2 Site. Midcontinental Journal of Archaeology 16(1):69-84.

Schwarcz, H.P., J. Melbye, M.A. Katzenberg and M. Knyf
1985 Stable Isotopes in Human Skeletons of Southern Ontario: Reconstructing
Paleodiet. Journal of A rchaeological Science 12:187-206.

Seeman, M.F.
1992 Woodland Traditions in the Midcontinent: A Comparison of Three Regional
Sequences. Research in Economic Anthropology Supplement 6:3-46.

Sievers, C. and L. Wadley
2008 Going underground: experimental carbonization of fruiting structures under
hearths. Journal of Archaeological Science 35(1 1):2909—2917.

Smith, B.D.

1992 Rivers of Change: Essays on Early Agriculture in Eastern North America.
Smithsonian Institution Press, Washington, DC.

1995 Seed Plant Domestication in Eastern North America. In Last Hunters First
Farmers, edited by T.D. Price and AB. Gebauer, pp. 193-214. School of
American Research Press, Santa Fe, NM.

2001 Low-Level Food Production. Journal of Archaeological Research 9(1):I-43.

Smith, B.D. and CW. Cowan .
1987 Domesticated Chenopodium in Prehistoric Eastern North America: New

Accelerator Dates from Eastern Kentucky. American Antiquity 52(2):355-357.
2003 Domesticated Crop Plants and the Evolution of Food Production Economies in

Eastern North America. In People and Plants in Ancient Eastern North America,

edited by RE. Minnis, pp. 105-125. Smithsonian Institution Press, Washington,

DC.

Sommer, J. . .
1994 Faunal Exploitation at Tobico Marsh. Honors TheSls submitted to the Department

of Anthropology, University of Michigan. Manuscript on file, Office of the

State Archaeologist, Lansing, Michigan. .
2002 Report on the Archaeology of the Shiawassee National Wildlife Refiige: The 2001

Field Season. Manuscript on file, Office of the State Archaeologist, Lansing,

Michigan. . . . .
2003a A Middle Woodland AMS Date from Saginaw County, Michigan. The Michigan

Archaeologist 49(3-4):87-95. .
2003b Report on the Archaeology of the Shiawassee National Wildlife Refiige: The 2002

243

Field Season. Manuscript on file, Office of the State Archaeologist, Lansing,
Michigan.

2005 Report on the Archaeology of the Shiawassee National Wildlife Refuge: The 2004
Field Season. Manuscript on file, Office of the State Archaeologist, Lansing,
Michigan.

2007 Report on the Archaeology of the Shiawassee National Wildlife Refuge: The 2006
Field Season. Manuscript on file, Office of the State Archaeologist, Lansing,
Michigan.

2008 Report on the Archaeology of the Shiawassee National Wildlifir Refiige: The 2007
Field Season. Manuscript on file, Office of the State Archaeologist, Lansing,

Michigan.

Spence, M.W. and W. Fox
1986 The Early Woodland Occupations of Southern Ontario. In Early Woodland

Archaeology, edited by K.B. F amsworth and TE. Emerson, pp. 4-46. Center for
American Archaeology Press, Kampsville, IL.

Spence, M.W., R.H. Pihl, and C. Murphy
1990 Cultural Complexes of the Early and Middle Woodland Periods. In The

Archaeology of Southern Ontario to AD. 1650, pp.125-l 70. Occasional
Publications no.5. London Chapter, Ontario Archaeological Society, London, ON.

Speth, J.
1972 Geology of the Schultz Site. In The Schultz Site at Green Point: A Stratified

Occupation Area in the Saginaw Valley of Michigan, edited by J .E. Fitting, pp. 53-
76. Memoirs No. 4. University of Michigan, Museum of Anthropology, Ann

Arbor.

Staller, J.E., R.H. Tykot, and BF. Benz (eds.)
2006 Histories of Maize: Multidisciplinary Approaches to the Prehistory, Linguistics,
Biogeography, Domestication, and Evolution of Maize. Academic Press, San

Diego, CA.

Stone, CD. and CE. Downum
I999 Non-Boserupian Ecology and Agricultural Risk: Ethnic Politics and Land Control

in the Arid Southwest. American Anthropologist 101:113-138.

Struever, S. (ed.)
1971 Prehistoric Agriculture. American Museum of Natural History Press, New York.

Struever, S.
1968 Flotation techniques for the recovery of small-scale archaeological remains.

American Antiquity 33 :353-362.

Struever, S. and K.D. Vickery
1973 The Beginnings of Cultivation in the Midwest-Riverine Area of the United States.

244

American Anthropologist 75(5): 1 l 97- 1220.

Tankersley, KB. and RA. Tench
2009 Riker-Todd: A Salvaged Ohio Hopewell Mound. North American Archaeologist

30(2):]95-217.

Theler, IL. and RF. Bozshardt
2006 Collapse of Crucial Resources and Culture Change: A Model for the Woodland to

Oneota Transformation in the Upper Midwest. American Antiquity 71(3):433-472.

Therin, M.
2006 Starch Movement in Sediment (Box 5.5). In Ancient Starch Research, edited by

R. Torrence and H. Barton, pp. 91-93. Left Coast Press, Inc., Walnut Creek, CA.

Thompson, R., J.P. Halt, H.J. Brumbach, and R. Lustek
2004 Phytolith Evidence for Twentieth-Century B.P. Maize in Northern Iroquoia.

Northeast Anthropology 68:25-40.

Thwaites, R.G. (editor)
1896-1901 The Jesuit Relations and A llied Documents. 73 Volumes. The Burrows

Brothers Company, Cleveland, OH.

Torrence, R.
2006 Starch and Archaeology. In Ancient Starch Research, edited by R. Torrence

and H. Barton, pp. 17-34. Left Coast Press, Inc., Walnut Creek, CA.

Torrence, R. and H. Barton (eds)
2006 Ancient Starch Research. Left Coast Press, Inc., Walnut Creek, CA.

Trigger, B.G
1969 The Huron Farmers of the North. Holt, Rinehart and Winston, New York.

Trigger, B.G and GM. Day
1994 Southern Algonquian Middlemen: Algonquin, Nipissing, and Ottawa, 1550-1780.

In, Aboriginal Ontario: historical perspectives on the first nations, edited by ES.
Rogers and DB. Smith, pp. 64-78. Dundum Press, Toronto, Ontario, Canada.

Van der Merwe, NJ.
1982 Carbon Isotopes, Photosynthesis, and Archaeology. American Scientist 70:596-

606.

van der Merwe, NJ. and J.C. Vogel

I978 13C Content of Human Collagen as a Measure of Prehistoric Diet in Woodland
North America. Nature 276:815-8 I 6.

245

Vennum, Jr., T.
1988 Wild Rice and the Q'ibway People. Minnesota Historical Society Press, St. Paul.

Vrydaghs, L. and TR Denham
2007 Rethinking Agriculture: Introductory Thoughts. In Rethinking Agriculture:

Archaeological and Ethnoarchaeological Perspectives, edited by T. P. Denham, J.
Iriarte, and L. Vrydaghs, pp. 1-15. Left Coast Press Inc., Walnut Creek, CA.

Wagner, GE.
2003 Eastern Woodland Anthropogenic Ecology. In People and Plants in Ancient

Eastern North America, edited by RE. Minnis, pp. 126-171. Smithsonian
Institution Press, Washington, DC.

Watson, P.J.
1976 In pursuit of prehistoric subsistence: A comparative account of some

contemporary flotation systems. Midcontinental Journal of Archaeology 1:77-100.
I995 Explaining the Transition to Agriculture. In Last Hunters First Farmers, edited by
T.D. Price and AB. Gebauer, pp. 21-38. School of American Research Press,
Santa Fe, NM.
1997 The Shaping of Modern Paleoethnobotany. In People, Plants, and Landscapes:
Studies in Paleoethnobotany, edited by K.J. Gremillion, pp. 13-22. The University

of Alabama Press, Tuscaloosa.

Webb 111, T., B. Shuman, and J.W. Williams
2004 Climatically Forced Vegetation Dynamics in Eastern North America during the
Late Quaternary Period. In The Quaternary Period in the United States, edited by

AR. Gillespie, S.C. Porter, and BF. Atwater, pp. 459-478. Elsevier, Amsterdam,
The Netherlands.

Williamson, B.
2006 Investigation of Potential Contamination on Stone Tools 030x 5.4). In Ancient

Starch Research, edited by R. Torrence and H. Barton, p. 89-90. Left Coast Press,
Walnut Creek, CA.

Wills, W.H.
1988 Early Prehistoric Agriculture in the American Southwest. School of American

Research Press, Santa Fe, NM.

Wilson, (1L.
I987 Buflalo Bird Woman Is Garden: Agriculture of the Hidatsa Indians. Reprinted.

Minnesota Historical Society Press, St. Paul. (Originally published 1917,
University of Minnesota).

Winterhalder, B. and C. Goland
1997 An Evolutionary Ecology Perspective on Diet Choice, Risk, and Plant

Domestication. In People, Plants, and Landscapes, edited by K.J. Gremillion,

246

 

pp. 123-160. The University of Alabama Press, Tuscaloosa.

Wright, GA.
1967 Some Aspects of Early and Mid-Seventeenth Century Exchange Networks in the

Western Great Lakes. The Michigan Archaeologist 13(4): 181-197.

Wright, HT.
1964 A Transitional Archaic Campsite at Green Point (ZOSAI). The Michigan

Archaeologist 10( 1 ): 1 7-22.

Wymer, DA.
1987 The Middle Woodland-Late Woodland Interface in Central Ohio: Subsistence

Continuity Amid Cultural Change. In Emergent Horticultural Economies of the
Eastern Woodlands, edited by W.F. Keegan, pp. 201-216. Occasional Paper No.
7, Center for Archaeological Investigations, Southern Illinois University,
Carbondale.

Cultural Change and Subsistence: The Middle Woodland and Late Woodland
Transition in the Mid-Ohio Valley. In Foraging and Farming in the Eastern
Woodland, edited by GM. Scarry, pp. 138-156. The University Press of Florida,

Gainesville.

1994 The Social Context of Early Maize in the Mid-Ohio Valley. In Corn and Culture
in the Prehistoric World, edited by S. Johannessen and CA. Hastorf, pp. 422-436.
Westerview Press, Boulder, CO.

1996 The Ohio Hopewell Econiche: Human Landscape Interaction in the Core Area. In
A View fiom the Core: A Synthesis of Ohio Hopewell Archaeology, edited by P.J.
Pacheco, pp. 36-52. Ohio Archaeological Council, Columbus.

1997 Paleoethnobotany in the Licking River Valley, Ohio: Implications for
Understanding Ohio Hopewell. In Ohio Hopewell Community Organization,
edited by W.S. Dancey and P.J. Pacheco, pp. 153-171. Kent State University

Press, Kent, OH.

I 993

Wymer, DA. and E. Abrams
2003 Early Woodland Plant Use and Gardening: Evidence from an Adena Hamlet in

Southeastern Ohio. Midcontinental Journal of Archaeology 28(2): 195-232.

Yamell, RA.
1964 Aboriginal Relationships between Culture and Plant Life in the Upper Great

Lakes Region. Anthropological Paper No. 23. Museum of Anthropology,

University of Michigan, Ann Arbor.
I972 Iva annua var macrocarpa: Extinct American Cultigen? American Anthropologist

74:335-341. .
I993 The Importance of Native Crops during the Late Archaic and Woodland Periods.

In Foraging and Farming in the Eastern Woodlands, edited by CM. Scarry, pp.
13-26. University Press of Florida, Gainesville.

247

 

Yerkes, R.W.
2005 Bone Chemistry, Body Parts, and Growth Marks: Evaluating Ohio Hopewell and

Cahokia Mississippian Seasonality, Subsistence, Ritual, and Feasting. American

Antiquity 70:241-266.
2006 Middle Woodland Settlements and Social Organization in the Central Ohio
Valley: Were the HOpewell Really Farmers?. In Recreating Hopewell, edited by

D.K. Charles and J .E. Buikstra, pp. 50-61. University Press of Florida,
Gainesville.

Zarillo, S. and B. Kooyman
2006 Evidence for Berry and Maize Processing on the Canadian Plains from Starch

Grain Analysis. American Antiquity 71(3):473-500.

Zvelebil, M. and M. Lillie
2000 Transition to Agriculture in Eastern EurOpe. In Europe ’s First Farmers, edited by

T.D. Price, pp. 57-92. Cambridge University Press, Cambridge.

248

 

63 8682

‘1'
‘Il
|
‘I‘I
‘
Ill-
\III
|||||
II‘II
I‘ll
l
|\
‘
I‘ll
.II‘II
|I||I
I“
ll‘
‘
‘hl
\
\II
.II‘
‘
|
‘l
‘I
\l
‘
|||

1 ll lllllllllll 111110

3

ll