LIBRARY

Mlchlgan State
Unlverslty

 

 

PLACE IN RETURN BOX
to remove this checkout from your record.
TO AVOID FINES return on or before date due.

 

DATE DUE

DATE DUE

DATE DUE

 

NOV 1 5 1999

APR 2 9 2005

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1“ mm.“

ANTHROPOMETRIC AND BIOMECHANICAL ASSESSMENT OF
SKELETAL STRUCTURAL ADAPTATIONS lN BlOARCHAEOLOGlCAL
POPULATIONS FROM MICHIGAN AND WESTERN NEw YORK
By

David Arthur BarondeSS

A DISSERTATION
Submitted to
Michigan State University

in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Anthropology
1 998

ABSTRACT
ANTHROPOMETRIC AND BIOMECHANICAL ASSESSMENT OF
SKELETAL STRUCTURAL ADAPTATIONS IN BIOARCHAEOLOGICAL
POPULATIONS FROM MICHIGAN AND WESTERN NEW YORK
By

David Arthur Barondess

This research investigates postcranial skeletal structural adaptations in
archaeological populations derived from Michigan and western New York. It
accomplishes this by examining differences in long bone structure that may
have coincided with changes in physical activity between the prehistoric and
historic periods in Michigan. Second, it explores long bone structural
differences between groups of hunter»gatherers and agriculturalists in western
New York state.

Two separate yet complimentary data sets, both Of which measure
dimensional and architectural characteristics Of the femur and humerus, but
which are predicated on different methodological approaches, are used. One
data set is derived from a whole bone anthropometric analysis of external bone
dimensions, the other is derived from a biomechanical analysis Of
computerized tomographic (CT) scan-generated diaphyseal cross-sectional

size, shape and strength properties.

The results Of this research demonstrate that there are significant
differences in the femoral and humeral dimensions and biomechanical
properties between the prehistoric and historic period groups in Michigan. For
both males and females, most measures Of diaphyseal size are smaller in the
historic period. Alternatively, the diaphyses are significantly stronger, arguing
for increased biomechanical demand, for both sexes, in the Michigan historic
period. The magnitude Of the femoral cross-sectional size and strength
differences between the prehistoric and historic periods is generally greater for
females than for males, suggesting that the level Of physical activity changed
more dramatically for females between the two periods. For the New York
hunter-gatherer and agricultural comparison, the results Of the biomechanical
analyses do not argue for any significant increase (or decrease) in workload in
one group compared to the other.

WIth the aid Of the archaeological record and historical documentation,
Patterns Of physical activity and their potential behavioral correlates within the
context Of culture contact are examined. Specifically, the influence Of the fur
trade on Native American subsistence pursuits during the historic period is
addressed. Results are evaluated against a broad backdrop Of comparable
research from other regions of North America, where similar analyses have
brought to light the general patterns of skeletal structural adaptation that can
be expected from modificatiOns in physical activity in a wide range Of

biocultural contexts. Finally, suggestions for future research are proposed.

This work is dedicated to my family, to whom my deepest gratitude can
never be adequately expressed. TO my parents, whose ceaseless love,
support, and encouragement sustained me throughout my many years
Of schooling, to my sister Lisa, whose courage continues to be my
greatest source Of inspiration, to my wife and best friend Margaret,
without whose unlimited patience and balanced perspective on matters
large and small this work would never have been completed, and to my
son Jacob and daughter Abigail, whose lives strengthen mine in more

ways than they will ever know.

iv

ACKNOWLEDGMENTS

Throughout the course of my research, many individuals and institutions
have supported my efforts, and I owe them all a great deal of thanks. This
research was generously supported by a National Science Foundation Doctoral
Dissertation Grant (BNS 9011522). I also want to thank the anthropology
department at Michigan State University (MSU) for supporting me with
numerous teaching assistantships and research and teaching awards.

I wish to thank my dissertation committee members, Drs. Norman
Sauer, William Lovis, Charles Cleland and Larry Robbins from MSU, Dr. Clark
Larsen from The University Of North Carolina at Chapel Hill, and Dr. Lorraine
Saunders Of the Rochester Museum and Science Center.

TO be singled out from all Of those who have shaped my development
as a physical anthropologist is my Chairman, Dr. Sauer. I am extremely
fortunate that my emergence as a teacher and scientist has been cultivated in

an atmosphere where his Office door knew no lock. Norm's patient, gentle,

determined guidance never failed me. AS a mentor, he unselfishly shared with
me his energy and wisdom about matters anthropological and otherwise.
Thank you Norm.

TO Dr. Lovis, I am especially thankful for the numerous archaeological
fieldwork experiences he has shared with me, as well as for our discussions
regarding anthropological issues too numerous to recount here. I have learned
much by Observing Bill’s approach to research, which is based on a critical eye
for detail, subtlety, and nuance.

On many occasions over the years, Dr. Cleland has shared with me his
thoughts on issues related to Michigan archaeology and ethnohistory. One of
Chuck’s seminars captured my interest in the contact period Of Michigan, and l
have gained much from his knowledge Of Great Lakes Indians, both past and
present.

I thank Dr. Robbins for encouraging me to broaden my geographical
horizons, and for his willingness to always share with me his thoughts on a
wide range Of anthropological topics. The broad perspective from which Larry
approaches anthropology has greatly helped "round out" this study.

I wish to thank Dr. Larsen, who encouraged me to consider a research
project on contact period bone biology. Without Clark’s tireless commitment to
the field of contact period bioarchaeology, there would be little intellectual and
theoretical backdrop for this study. I also appreciate Clark’s invitation to

participate in the St. Catherine’s Island bioarchaeology project.

vi

One Of the most exciting aspects Of this project was the collaboration
that developed with the Rochester Museum & Science Center (RMSC),
Rochester, New York. Foremost at this institution, I am deeply indebted to Dr.
Lorraine Saunders. Without Lori’s deep interest in Seneca bioarchaeology, the
New York portion of the research would not have been possible.

In a climate where gaining access to Native American skeletal samples
can be problematic, I am greatful for having had the opportunity to tap the
scientific research potential of the collections examined in this study.

From the RMSC, I am indebted to Charles Hayes, lll, former Museum
Director, for helping to secure permission to transport material to my home
institution for study. I also wish to thank the RMSC for naming me an Arthur
C. Parker Fellow, which provided financial support to help defray expenses
associated with my trips to Rochester.

From the Henry Ford Hospital (HFH) in Detroit, I wish to thank Dr.
Dianna D. Cody, Senior Research Scientist, who directly supervised all aspects
of the CT study, and provided much needed assistance throughout this highly
technical, often laborious and frustrating, phase Of data collection. I would also
like to express my gratitude to Dr. Michael Flynn, Director of Radiologic
Physics and Engineering at Henry Ford, for his attention to the details that
helped move the CT scanning work along. I am forever grateful to James
Ciarrelli, who patiently sat by my side for every CT scan and readied all Of the

scan raw data for analysis.

vii

Several individuals at The University of Michigan were also helpful to
me with regard to data collection. I thank Dr. John O’Shea for his helpful
insights during the very formative stages Of the research, and Dr. Henry
Wright, who helped me gain access to the Juntunen Site skeletal collection.

Thanks tO Dr. Richard Wilkinson at the State University Of New York,
Albany, for facilitating access to skeletal material at his institution. Dick was a
gracious host on my trips to Albany, and our discussions on a wide range of
bioarchaeological topics were an added bonus to my visits.

I am also grateful for the extensive assistance Offered me by Dr.
Christopher Ruff of The Johns Hopkins University School Of Medicine. I
gained much knowledge from Chris’ willingness and enthusiasm to discuss any
aspect of scanning archaeological bone with me. He has been a major figure
in laying the groundwork for those with an interest in CT and archaeologically-
derived skeletal remains, and his work provided a tremendously firm platform
from which this study could be launched.

I wish to thank Dr. Patricia Bridges of Queens College for her help with
the anthropometric component Of the study; Dr. Susan Pfeiffer, SchOOI Of
Human Biology, University of Guelf, Ontario, who graciously shared with me
her age and sex data from the Frontenac Island site; and Dr. Dean Anderson,
Office Of the Michigan State Archaeologist, whose enthusiasm to share with
me his knowledge Of the archaeological and ethnohistorical records as pertain

to the contact period in the Great Lakes region was immensely appreciated.

viii

I also want to express my deepest appreciation to Dr. Dorothy Nelson of
the Wayne State University School of Medicine. I have benefitted immensely
from my experience in Dorothy’s laboratory, having been afforded the Chance
to work on the "extant side" Of skeletal biology. The knowledge I have gained
in this regard has helped expand the perspective from which I pursue my

interests in bone biology, and for this opportunity I am continually greatful.

ix

TABLE OF CONTENTS

LIST OF TABLES .......................................... xiii

LIST OF FIGURES ......................................... xiv
CHAPTER 1

INTRODUCTION ............................................ 1
CHAPTER 2

BIOMECHANICS AND THE HUMAN SKELETON .................... 7

Wolff’s-Law..... ...................................... 7

. Animal. Models and Skeletal Adaptation ...................... 10

Computed Tomography (CT) Scanning ...................... 12

Theoretical and Practical Principles Of CT Scanning ............. 14

The Hounsfield Scale ................................... 17

CT Scanning vs. Radiography (X-Rays) ..................... 18

Biomechanical Principles ................................ 19

Skeletal Structure and Engineering Beam Theory .............. 22

Application Of CT to Bioanthropological Issues ................. 25
CHAPTER 3

BIOARCHAEOLOGY ........................................ 27

Previous Research .................................... 27

Current Research ..................................... 32

Michigan Prehistoric Period ................... . ........... 32

The Riviera aux Vase Site .......................... 32

The Juntunen Site ................................ 35

Michigan Historic Period ................................ 37

The Lasanen Site ................................ 37

, The Fletcher Site ................................. 39

New York Prehistoric Period .............................. 40

The Frontenac Island Site .......................... 40

The Harscher Site ................................ 41

New York Historic Period ................................ 42

The Tram Site ................................... 42

Hypotheses .......................................... 43

CHAPTER 4

MATERIALS AND METHODS ................................. 47
Skeletal Sample Selection ............................... 47
Femoral and Humeral Whole Bone Anthropometry .............. 53
Computed Tomography (CT) Scanning ...................... 55

Technicare 1440 control display parameters ............. 55
The role of the phantom in CT scanning ................ 58
The orientation box and CT scanning .................. 59
Femoral Orientation . . .' ................................. 64
The anteroposterior (A-P) axis ....................... 65
The mediolateral (M-L) axis ......................... 66
The rotational axis ................................ 68
Final positioning ................................. 68
Humeral Orientation .................................... 69
The anteroposterior (A-P) axis ....................... 69
The mediolateral (M-L) axis ......................... 69
The rotational axis ................................ 70
Processing the CT-Generated Cross-Sectional Image Data ....... 70
Biomechanical Properties ................................ 75
Normalization of Data .................................. 80
Statistical Approach .................................... 81

CHAPTER 5

RESULTS ................................................ 84
Anth ropometrics ...................................... 84

Michigan prehistoric vs. historic period: male femora ....... 84
Michigan prehistoric vs. historic period: female femora ...... 85

New York hunter—gatherer vs. agricultural groups:
male femora ............................... 86

New York hunter-gatherer vs. agricultural groups:
female femora .............................. 86
Michigan prehistoric vs. historic period: male humeri ....... 87
Michigan prehistoric vs. historic period: female humeri ...... 87

New York hunter-gatherer vs. agricultural groups:
male humeri ............................... 88

New York hunter-gatherer vs. agricultural groups:
female humeri .............................. 88
Biomechanics ........................................ 89
Michigan prehistoric vs. historic period: male femora ....... 89
Michigan prehistoric vs. historic period: female femora ...... 90

New York hunter-gatherer vs. agricultural groups:
male femora ............................... 91
Michigan prehistoric vs. historic period: male humeri ....... 92
Michigan prehistoric vs. historic period: female humeri ...... 92

xi

New York hunter-gatherer vs. agricultural groups:

female humeri .............................. 93
Testing Of Hypotheses ............................. 93
CHAPTER 6
DISCUSSION ............................................ 100
Summary and Conclusions .............................. 110
Suggestions for Future Research ......................... 112
APPENDICES
Appendix A: Femoral and Humeral Anthropometrics-
Measurements, lnstrumentation and Explanations . . . 117
Appendix B: Results for Femoral and Humeral Anthropometric Data-
Michigan and New York Samples ............... 125
Appendix C: Results for Femoral and Humeral Biomechanical Data-
Michigan and New York Samples ............... 138
Appendix D: Figures to Anthropometric and Biomechanical Results-
‘ Michigan'and New York Samples ............... 157
Appendix E: Figures to CT Scanning ...................... 189
LITERATURE CITED ............. ' .......................... 194

xii

Table 1
Table 2
Table 3
Table 4

Table 5
Table 6

Table 7

Table 8

LIST OF TABLES

Femoral Sample for Anthropometric Analysis:

Breakdown by Sex and Side ......................... 51
Humeral Sample for Anthropometric Analysis:

Breakdown by Sex and Side ......................... 51
Femoral Sample for CT Analysis:

Breakdown by Sex and Side ......................... 52
Humeral Sample for CT Analysis:

Breakdown by Sex' and Side ......................... 52
Femoral and Humeral Whole Bone Anthropometry ......... 54
Example Of CT Scan Number Calculation ............... 57

Summary of Changes in Femoral Diaphyseal Dimensions and
Biomechanical Cross-Sectional Measurements ............ 98

Summary Of Changes in Humeral Diaphyseal Dimensions and
Biomechanical Cross-Sectional Measurements ............ 99

xiii

LIST OF FIGURES

APPENDIX D
Figure D1 - Maximum length of the femur (mm): Michigan ............ 157
Figure DZ - Maximum trochanteric length of the femur (mm): Michigan . . . 158
Figure 03 - Subtrochanteric A-P diameter of the femur (mm): Michigan . 159
Figure D4 - Subtrochanteric M-L diameter Of the femur (mm): Michigan . . 160
Figure 05 -Femoral neck girth (mm): Michigan ................... 161
Figure D6 - Proximal epiphyseal breadth Of the femur (mm): Michigan . . . 162
Figure D7 - Midshaft A-P diameter Of the femur (mm): New York ...... 163
Figure D8 - Subtrochanteric A-P diameter of the femur (mm): New York . . 164
Figure 09 - Subtrochanteric M-L diameter of the femur (mm): New York . . 165
Figure D10 - Maximum length Of the humerus (mm): Michigan ......... 166
Figure D11 - Lesser tubercle-trochlear length of the humerus (mm):

Michigan ...................................... 167
Figure D12 - Head-medial epicondylar length of the humerus (mm):

Michigan ...................................... 168
Figure D13 - Maximum midshaft diameter Of the humerus (mm):

Michigan ...................................... 169
Figure D14 - Humeral head girth (mm): Michigan .................. 170
Figure D15 - Cross-sectional medullary area Of the femur (mmz):

Michigan ...................................... 171

xiv

Figure D16 - Cross-sectional total area Of the femur (mmz): Michigan . . . .

Figure D17 - lw (maximum bending strength Of the femur-mm‘):

Michigan .....................................
Figure D18 - J (torsional strength Of the femur-mm“): Michigan .......
Figure 019 - PCA (% cortical area Of the femur-mmz): Michigan .......
Figure 020 - Cross-sectional medullary area Of the femur (mmz): Michigan
Figure D21 - Cross-sectional cortical area of the femur (mmz): Michigan . .

Figure D22 - Cross-sectional total area Of the femur (mmz): Michigan . .

Figure 023 - lw (maximum bending strength Of the femur-mm‘):

Michigan .....................................

Figure 024 - I (minimum bending strength of the femur-mm‘):

min

Michigan .....................................
Figure 025 - J (torsional strength Of the femur-mm‘): Michigan .......
Figure 026 - Cross-sectional total area Of the humerus (mmz): Michigan . .
Figure 027 - lx (M-L bending strength of the humerus-mm‘): Michigan . . .
Figure 028 - J (torsional strength of the humerus-mm‘): Michigan .....
Figure 029 - Cross-sectional total area Of the humerus (mmz): Michigan . .

Figure 030 - lx (M-L bending strength of the humerus-mm‘): Michigan . . .

Figure 031 - Cross-sectional medullary area of the humerus (mmz):

New York ....................................

172

173

174

175

176

177

. 178

179

180
181
182
183
184
185

186

. 187

LIST OF FIGURES (continued)

APPENDIX E
Figure 1 - Technicare 1440 CT scanner ......................... 189
Figure 2 - Cross-sectional CT images

(top to bottom 4 different left humeri) .................... 190
Figure 3 - CT long bone orientation box ......................... 191
Figure 4 - CT image Of long bone scanning apparatus ............... 192

Chapter 1

INTRODUCTION

Anthropology has a deep history in attempting to understand human
biological and cultural variability by examining data derived from the fossil record
as well as from extant populations. Indeed, an appreciation Of the evolutionary
forces that are responsible for mOlding the biocultural milieu within which humans
are constantly Changing is 'an essential precondition for Characterizing the
tremendOus biological diversity that typifies our species.

As a logical extension to this fOCus on adaptation, the past two decades
have witnessed a burgeoning interest, from a variety Of perspectives within
biological anthropology, in the contributions that archaeologically-derived human
skeletal remains can make with regards to fostering a more holistic understanding
of the processes that help shape biocultural evolution. For example,
bioarchaeological studies have addressed the impact Of dietary change on health '
and disease, as well as the association between nutritional adequacy and skeletal
growth and development. Paleodemographic studies are also now numerous,

especially those focusing on age- and sex-specific patterns Of population growth

2

and decline. Other important contributions have been made in the arena Of
infectious disease research, where the sequelae Of morbidity and mortality
experience can be examined. The evaluation Of skeletal pathological conditions
with respect to trauma and osteoarthritis is also an area of research that has
advanced our understanding Of the interaction between biology and culture.

This dissertation makes an additional contribution to this growing body Of
literature; that is, it examines postcranial skeletal structural adaptations in relation
to physical activity. It accomplishes this by utilizing two separate case studies.
First, it examines differences in long bone structure that may have resulted from
concomitant Changes in physical activity between the pre- and post-contact
periods in Michigan. Second, it explores long bone structural differences between
groups Of hunter-gatherers and agriculturalists in western New York State.

In order to test the hypotheses as outlined in Chapter 3, two separate yet
complementary data sets, both Of which measure dimensions of the femur and
humerus, but which are predicated on quite different methodological approaches,
are used. One data set is derived from a whole bone anthropometric analysis Of
external bone dimensions, the other is derived from a biomechanical analysis Of
computerized tomographic (CT) scan-generated diaphyseal cross-sectional size,
shape, and strength characteristics. As a means Of depicting skeletal anatomy,
CT is unique in its ability to (1) precisely measure size and shape-related changes
resulting from the application of biomechanical forces, (2) measure properties Of

bone shape and strength that metric-based (e.g., anthropometric) studies cannot,

3

and (3) "visualize" the internal architecture Of bone non-invasively.

A number of studies, including those of Bridges (1985, 19893, 1989b,
1991), Brock (1985), Brock and Ruff (1988), Larsen (1982, 1984, 1987, 1990,
1995, 1997), Larsen and Ruff (1991), Ruff (1981), Ruff and Hayes (1982, 1983a,
1983b), Ruff and Larsen (1990), Ruff and Leo (1986), Ruff et al. (1984), Sumner
(1984), and Sumner et al. (1985), have utilized similar methods to evaluate
human skeletal remains in terms of long bone adaptations to physical activity from
a variety Of geographic contexts vis-a-vis the pre- and post-contact period
transition, or in terms Of sUbsistence change between hunter-gatherer and
agriculturally—based populations. These studies provide the methodological and
theoretical foundation for the present research.

This study, being the first of its~kind to examine femoral and humeral
structural adaptations in skeletal samples derived from Michigan and western New
York State, can provide answers to questions related to the biocultural
adaptations Of the prehistoric and historic period Native American inhabitants of
these two regions. It is also the first to explore the possibility that males and
females from these two areas may have adapted differentially to the biocultural
changes taking place in their respective environments.

In order appreciate how it is that the mechanical demands to which humans
are subjected throughout the course Of their lives leave an indelible mark on bone,
it is important to clearly understand Wolff’s Law, the single most important

organizing principle within the field of skeletal biology.

4

Chapter 2 begins with a comprehensive overview Of Wolff’s Law, including
a discussion on the response Of skeletal tissue to mechanical stimuli. The nature
Of cortical and trabecular bone orientation with respect to long bone diaphyseal
strength is discussed. The role of in vivo animal models in laying the foundation
for skeletal structural analyses in humans is briefly described, as is previous
research on skeletal function and mechanical adaptation in a host Of non-human
primate and early hominid groups.

Next is a short historical overview Of CT development. The theoretical and
practical principles governing CT technology are discussed, including the
Hounsfield scale in relation to skeletal tissue. The advantages of CT to
conventional radiography in terms Of the three-dimensional imaging Of human long
bones are laid out, and the influence of biomechanical forces, including a
discussion that distinguishes loads from stresses, is presented. The distinction
is made between those aspects of bone biology that are genetically, as Opposed
to environmentally, determined. Engineering beam theory is introduced, and its
application in studies involving skeletal adaptations and the influence Of
mechanical loading is examined.

Chapter 3 outlines the bioarchaeological background to the current study.
It begins by summarizing the major conclusions from research into human long
bone diaphyseal structural adaptations and levels Of physical activity in
archaeological remains from North America. The focus here is on the prehistoric

American Southwest, northwestern Alabama, and the Georgia coast. The

5

Michigan and New York archaeological sites from which the osteological data is
derived for the current research are outlined. Each site is discussed in terms Of
its geographical setting, temporal context, and major subsistence adaptation.
Relevant previous physical anthropological work is summarized. Hypotheses,
accompanied by alternative expectations based on the general patterns Of skeletal
structural adaptation that have been elucidated in comparable studies, are
presented independently for the Michigan and New York samples. For these
regions, the anthropometric and biomechanical data sets are also evaluated
independently.

Chapter 4 discusses the specific criteria, and their rationale, for selecting
the Skeletal samples. Sample sizes are provided for both the anthropometric and
biomechanical analyses. A description Of all femoral and humeral anthropometric
measurements is given, and issues related to long bone orientation in CT studies
is laid out. The formulae used to derive the cross-sectional area and moments
of inertia properties are derived, normalization Of the biomechanical data is
discussed, and the statistical approach is given.

Chapter 5 presents the results Of all statistical tests for both the
anthropometric and biomechanical data sets. Major sections are broken down by
state, skeletal element (i.e., femur or humerus), and sex (i.e., male or female).
General patterns Of skeletal size, structure, and strength are presented, and all
statistically significant results are discussed.

Chapter 6 examines the anthropometric and biomechanical results

6
independently with respect to the Michigan and New York skeletal samples.

Differences in long bone structure that may have resulted from changes in
physical activity are addressed, and the sex-specific behavioral correlates
associated with such changes are suggested. The hypotheses are evaluated, and
the results are then discussed within the broader context Of comparable studies
from other areas Of North America. Finally, suggestions for future research are

considered.

Chapter 2

BIOMECHANICS AND THE HUMAN SKELETON

Wolff’s Law

It has been 'said with regard to the human skeleton that nothing is as
constant as change (Stein et al., 1955). Indeed, it is axiomatic in skeletal biology
that the internal and external architecture Of human long bones are remodeled in
response to the direction and magnitude Of mechanical stress (Burr, 1980; Enlow,
1968; Frost, 1964, 1973; Hayes and Carter, 1979; Hoyte and Enlow, 1966;
Lanyon, 1984; Lanyon and Baggott, 1976; Lanyon and Rubin, 1984; Martin and
Burr, 1989; Tate and Cam, 1982; Weinmann and Sicher, 1955). Skeletal
remodeling, then, can be seen as the way in which bone adapts, by changing its
size, shape, and structure, in response to externally applied loads, or forces
(Carter and Spengler, 1978; Currey, 1968; Larsen, 1997). This remarkable and
unique aspect of bone as a structural material is embodied in Wolff’s Law of
Transformation (Wolff, 1892), or the "Law Of BOne Remodelling," which holds that

bone tissue adapts to its mechanical environment in response to the stresses

8
placed upon it (Lovejoy et al., 1976). As such, Wolff’s Law provides the

theoretical underpinning for understanding how it is that a skeleton’s final form
provides a window through which the external stimuli (i.e., mechanical forces) to
which an individual’s skeleton is subjected throughout life can be examined and
interpreted.

It follows that Changes (i.e., adaptation) in the strength and/or direction of
force will be accompanied by concomitant changes in bone strength and/or
structure. While the exact mechanism explaining the relationship between
skeletal form and mechanical loading is not fully understood (Frost, 1995; Hert et
al., 1969; Kannus et al., 1996; Liskova and Hert, 1971), it is known, for example,
that in response to changes in mechanical demand, at the cellular level, a bone’s
collagen fibers will orient themselves parallel to the direction of the greatest
tensile and/or compressive forces to which they are subjected. Bone mass (a
material property) is also influenced by mechanical stimuli and is an important
factor in determining bone strength (Beck et al., 1993); however, the response of
skeletal tissue to changes in the mechanical environment is predominately an
architectural one, having to do with size‘and distribution (Ferretti et al., 1993;
Larsen, 1997).

Manifestations Of Wolff's Law extend throughout the entire human skeleton.
As an example, one needs only point to the longitudinal cross-section Of a
proximal femur to visualize the relative distribution of cortical bone and the

intricate alignment of trabeculae with respect to the direction Of the lines Of force

9
acting along the metaphysis and epiphysis. In order to appreciate the fact that

the physical forces generated by human activity leave an indelible mark on bone,
it is worthwhile remembering that bone functions in the dual role Of serving both
as the primary structural element Of the human body as well as an organ, and
hence possesses both biomechanical and physiological properties important in the
remodeling process. For example, bones help protect internal organs and soft
tissues. Bones that serve a protective function are primarily found within the axial
skeleton, and are composed Of two layers of cortex, in between which is found a
layer Of cancellous or trabecular bone. This particular orientation enables bone
to accommodate high levels Of mechanical force without incurring structural
damage. Indeed, the skeleton Of all terrestrial vertebrates represents several
million years of evolution having worked to maximize physical strength, structural
support, efficiency of locomotion and mineral availability.

Whether conducting empirical studies under carefully-controlled laboratory
conditions designed to test bone strength and rigidity in relation to various
extemalIy-applied loads (Enlow, 1968; Frost, 1964; Jones et al., 1977; Lanyon,
1986; Lanyon and Baggott, 1976; Lanyon and Boum, 1979; Lanyon and Rubin,
1984) or, as in the present study, generating research hypotheses designed to
compare a broad spectrum Of morphometric change in bioarchaeological remains,
skeletal structure and function research vis-a-vis morphological adaptation is
predicated on the knowledge that bone is a dynamic, living tissue, with the ability

to alter its structure in response tO changing mechanical requirements (Hayes,

10
1986; Hayes and Snyder, 1981; Martin and Atkinson, 1977; Martin and Burr,

1989; Townsley, 1948; WOO et al., 1981).

Animal Mmgls ang Skeletal Adaptation

There has been a plethora Of non-human—based research focused on the
systematic manipulation of a long bone’s loading environment in order to measure
the element’s response to mechanical demand (Lanyon and Rubin, 1983). The
results of such studies not only support Wolff’s Law, but they provide a conceptual
framework for interpreting the relationship between Skeletal remodeling and
alterations in mechanical loading. Many of these investigations incorporate one
or more Of the major principles addressed in the current research, including
engineering beam theory, mechanical forces, and skeletal cross-sectional area
and geometric properties.

Two classic studies involving surgically-induced alterations to the loading
environment illustrate the approach taken to examine skeletal modifications in
response to changes in the mechanical environment. One of these studies
examined laboratory dogs (Chamay and Tschantz, 1972), where removal Of the
radius resulted in an approximately one-third compensatory increase in ulnar
diaphyseal size within 16 days. After nine weeks, the ulnar diaphysis increased
in size by a minimum Of 60 percent. Bone apposition was manifest on both the
subperiosteal and endosteal surfaces, and it was apparent that an intermittent

applimtion Of loads, as Opposed to loads applied statically, had the greatest

11

impact on skeletal remodeling. In another study involving pigs (Goodship et al.,
1979), ulnar removal resulted in a rapid increased in bone apposition in the
radius.

Non-surgically induced approaches, including exercise regimes, have also
been used to examine the response of skeletal tissue to alterations in mechanical
forces (e.g., axial compression and mediolateral bending). A widely cited example
is a study Of sub-adult pigs (WOO et al., 1981) where significant endosteal
apposition and remodeling of bone, with a corresponding increase in bending
strength, was documented. Other animal studies have involved horses
(Piotrowski et al., 1983; Rybicki et al., 1977), rabbits (Hert et al., 1969) and rats
(Ferretti et al., 1993).

Numerous studies aimed at evaluating hypotheses regarding skeletal
function and mechanical adaptation have also been accomplished in several
groups of prosimians, monkeys, apes, and early hominids. Some specific
examples include research on the extinct giant prosimian Megaladapis edwardsi
and the largest extant prosimian Indri indri (Jungers and Minns, 1979), Malaysian
leaf-monkeys (F leagle, 1977), macaques (Burr et al., 1981 ; Runestad et al., 1 993)
and lesser bushbabys (Galago senegalensis), where femoral mechanics have
been examined in order to elucidate the structural adaptations Of the hindlimb tO
leaping (Burr and Piotrowski, 1982). Mechanical properties and their associated
characteristics related to bending strength have also been examined in the

mandibular corpus Of Otavipithecus namibiensis, a middle Miocene hominoid from

1-2

southern Africa (Schwartz and Conroy, 1996). Ruff (1989) has presented an
overview of how long bone structural analyses in many Of these groups can be
used to address issues Of body mass estimation, limb dominance, and cross-
sectional shape differences. In terms Of early hominid evolution, cross-sectional
material properties with respect to activity-related changes in femoral and humeral
morphology have been studied in fossil specimens representing the remains Of
Homo erectus, Neandertals, and archaic Homo sapiens (Ruff et al., 1993; Ruff
et al. 1994; Trinkaus et al., 1994).

While there is no shortage Of biomechanically-based studies on extant
humans, as alluded to above, the level Of analysis has traditionally been that Of
a single bone. This is appropriate, in that the primary goal Of such research has
been to further the understanding of human locomotor biomechanics. Only
relatively recently have such studies been extended to incorporate the anatomical
and behavioral characteristics within and between archaeologically-derived
populations, and an overview Of several Of these key works is presented in the

next chapter.

Computed Tomggraphy (CI) Scanning

Today, the modality Of choice for examining the intricate three-dimensional
structure Of skeletal elements is computerized tomography (CT) scanning. The
theoretical basis for the first CT scanner was developed in the late 1960’s by

Godfrey N. Hounsfield at the Central Research Laboratories Of EMI Limited,

13

London (Morgan, 1983). After initial experimentation with inanimate Objects, and
subsequently with diseased brain tissue, Hounsfield demonstrated that a machine
built to incorporate CT technology could produce cross-sectional images that
previously had been difficult to visualize with conventional x-ray techniques
(Ambrose, 1973; Ambrose, 1975; Ledley et al. 1974). These early machines,
known as "brain scanners,” were only capable of producing images Of the head,
and their primary use was in the production Of images Of abnormalities within the
cranium, such as brain tumors. For his successful incorporation of the theoretical
. and technological aspects of CT technology into a machine capable of providing
a range Of cross-sectional images in living people, Hounsfield is widely credited
with making CT scanning clinically feasible (Seeram, 1982). In 1972, he shared
the Nobel Prize in Physiology and Medicine with A. M. Cormack, a Tufts
University physicist who independently made significant strides while working with
CT and cross-sectional imaging technologies.

Shortly after Hounsfield’s early work, CT techniques were adapted to image
other anatomical regions, and a generation Of "whole body scanners" emerged
that were capable of producing transverse-oriented sectional images anywhere
within the body. The first two Of these CT scanning units in the United States
were installed at the Mayo Clinic and the Massachusetts General Hospital in 1973
(Gordon et al., 1975; Morgan, 1983), and within four years, some 760 CT
scanners had been ordered by hospitals, clinics, and physicians (Computed

Tomographic Scanning, 1977; Policy Implications Of the Computed Tomography

14

(CT) Scanner, 1978). CT was soon hailed as the most significant advance in
radiologic science since the discovery of x-rays by Roentgen in 1895. Several
more advanced generations of scanners have dramatically decreased both
scanning and image reconstruction time through numerous modifications related
to the x-ray source and radiation detectors. Today, CT technology is recognized
as an integral component of diagnostic radiology, relying on the expertise Of a

wide-range of professionals within the clinical and research communities.

Theoretical and Practical Principles'of CT Scanning

The word tomography is derived from the Greek "tomos," which means "a
slice or section." It follows that the primary goal Of tomography is to depict
several overlapping layers Of tissue selectively and distinctly (Gambarelli et al.,
1977), leading to the reconstruction Of an image that accurately replicates the
morphology Of a transverse cross-section, or slice, Of anatomy, be that structure
of soft (e.g., fat or muscle) or hard (e.g., bone) tissue origin (Boyd and Parker,
1983; Morgan, 1983; Ter-Pogossian, 1983).

The CT scanner is composed Of six basic elements: (1) a subject handling
table; (2) a scanning gantry, a movable frame on the CT machine containing an
x-ray source, collimators, which function to reduce the emergent radiation to the
size of a fine beam, and detectors; (3) the data acquisition electronics; (4) the x-
ray generator; (5) a computer, which collects, stores, and processes information

from the detector; and (6) a viewing console, with a television-like screen on

1 5
which the reconstructed image produced by the computer is displayed for viewing

and/or photographic recording.

The primary theoretical achievement Of CT imaging is predicated on its
ability tO direct radiation at an object from multiple angles. With the aid Of a
computer capable Of storing and processing x-ray transmission data, a transverse
section, or cross-sectional image, of that object, devoid of any superimposed
structures, can then be reconstructed (Perry, 1980; Seeram, 1982). During CT
scanning, the energy source emits a beam Of x-rays that passes through an
Object, and is collected by the detector. The energy readings at each position of
the scanning beam’s geometrical coordinates are stored by the computer, each
reading reflecting the amount of energy lost by the beam of x-rays. This
reduction in photon energy as the scanning beam passes through a given
structure is known as attenuation. The computer then reconstructs an image and
displays it on the screen as a cross-section, or "slice."

The practical basis Of CT technology lies in the domain Of digital image
processing, which involves the conversion Of raw data (e.g., from a human long
bone cross-section) by a computer-applied algorithm, or reconstruction technique,
into numerical data (T er-Pogossian, 1983). Specifically, structures are
differentiated by their ability to absorb different energy levels. The more dense
a structure is, the. more energy it will absorb. Since exposure to an energy source
blackens x-ray film (i.e., the receptor), when relatively little energy reaches the x-

ray film, the image of that structure will appear light. A structure such as bone,

16

which is dense relative to other body tissues, appears white on x-ray film; air,
much less dense than bone, appears black. Structures with intermediate
densities, such as fat and soft tissue, appear as shades of gray. CT images are
capable Of showing extremely minute variations (i.e., spatial and contrast
resolutions Of 51 line/mm and s 0.5%, respectively) in x-ray attenuation, and the
computer and monitor have the technical ability to display CT images across a
broad spectrum Of contrast and brightness values. In fact, it is customary tO view
CT images at high contrast levels compared with conventional x-rays, and by
varying the window level (brightness) and width (contrast), the viewer can
visualize the entire density range Of the Object being scanned.

The scanned images themselves are composed Of picture elements
(pixels), small squares arranged in two dimensions (rows and columns) to form
a matrix, which is typically 256x256 or 512x512 pixels in size. Each pixel
corresponds to a small volume Of tissue (i.e., bone). For each pixel of tissue, the
CT scanner measures the x-ray attenuation within that pixel, independently Of the
remainder Of the picture. The reconstructed image represents not the image at
the surface Of the slice, but rather an average Of the full thickness Of that slice
(Winter and King, 1983). CT images are displayed as cross-sectional views
oriented as if the scanned Object is "cut" horizontally. Since scans are always
taken tO approximate as precisely as possible a horizontal plane through the
Object Of interest, variation (i.e., an inaccurate image and, ultimately, the inability

to confidently compare scans from one Object to the next) is potentially introduced

17
by tilting the gantry and/or the Object. Chapter 4, Materials and Methods,

discusses how femora and humeri can be oriented relative to the CT table and

scanning beam in order to ensure consistent, reproducible, horizontal slices.

The Hounsfield Scale

For the purposes Of CT scanning, a measuring scale, calibrated in
Hounsfleld units, or HU, is used to represent a CT number. The HU, or CT
number, is defined by the expression pm-pmm/pwam x K, where n represents the
attenuation coefficient Of x-rays Of the imaged tissue, and um, represents the
attenuation coefficient of water. The magnifying constant, K, is assigned a value
Of +1 ,000, and is used tO'extend the HU scale so that a wide range of attenuation
values can be incorporated onto the scale (T er—Pogossian, 1983). Each CT
number designates the x-ray attenuation in each picture element Of the CT image,
and the HU scale records Objects with increasing calcification at progressively
higher densities; the units, or values, themselves represent a relative expression
referenced to the attenuation Of radiation in water. X-ray attenuation values are
thus used on an arbitrary numerical scale, where air is represented as -1000 HU,
soft tissue ranges from -100 to +100 HU, water is 0 HU, bone ranges from +300
to +1000 HU (Hounsfield, 1973; Morgan, 1983), and particularly dense bone may
measure more than +1000 HU. Using the Hounsfield scale, a change of 1 HU
corresponds to a change Of 0.1% in the linear attenuation coefficient relative to

water (Boyd and Parker, 1983).

18

CT Scannin vs. Radi ra h X-ra s

For the purposes Of imaging the three-dimensional anatomy of skeletal
tissues, CT scanning provides many advantages over conventional x-rays; only
those that are directly applicable to this research are outlined here. First, on x-ray
film, internal structures overlap, thereby obscuring each other’s image (Gordon
et al., 1975). By rotating its source beam to produce cross-sectional images, the
CT scanner eliminates this problem Of structural overlap. Additionally,
conventional x~rays Often do not clearly differentiate between adjacent structures
Of even moderately similar densities (e.g., the interface between cortical and
cancellous bone surfaces). Hence, on x-‘ray, adjacent structures that appear as
shades Of gray are not easily distinguishable from one another. By accomplishing
numerous x-ray exposures from different angles to produce a well-defined, single
image, extremely small density differences that are intrinsically present within a
single structure can be accurately discriminated with CT imaging technology
(Genant et al., 1981; Ter-Pogossian, 1983).

The research in the present study is predicated on the ability to obtain
precise measures Of femoral and humeral architecture. Radiographic techniques
are not capable Of accurately producing the cross-sectional images necessary for
evaluating the biomechanical properties examined in this study.

The noninvasive nature Of CT scanning makes it an especially timely and
appropriate modality for studying a broad range Of issues related to human

biocultural adaptations at a time when the disposition and scientific treatment Of

19
archaeologically—derived skeletal remains are being actively debated by both

Native and non-Native peoples. Given the sensitivity being accorded skeletal
collections such as those examined here, it is unlikely that the present study could
have been accomplished by utilizing a protocol that involved physically sectioning

bone in order to examine cross-sectional diaphyseal anatomy.

Biomechanical Principles

It is clear that any skeletal structural analyses of human bone have at their
core an appreciation Of the interaction between biology and engineering-based
mechanical concepts. The amalgamation of these two fields is known as
biomechanics, which has been described as the application Of engineering
principles to biology and physiology (Frost, 1967) and, specifically, to the
locomotor system of living animals who possess dynamic tissues that are modified
in response to mechanical loading (Dickie et al. 1984; Frankel and Burstein, 1970;
Larsen, 1997). As background for any discussion on how biomechanical forces
influence human long bone structure, a brief review Of how forces, in general, act
upon any Object in space, including bone, is worthwhile.

A force represents a motion causing the acceleration Of matter, as well as
the resistance of that matter to the acceleration (Frost, 1967). In biomechanical
studies forces are generally divided into two major categories, loads and
stresses. Loads represent a particular type of force that originate outside (i.e.,

external to) the structure in question, and either deform (Hayes, 1986) or are

20
sustained by that structure. The following example, albeit highly simplified, will

help differentiate between what is meant by loads and stresses.

In humans, synchronized movement between the legs and arms is
necessary for coordinated bipedal locomotion. In order to accomplish the
seemingly simple task of taking a forward step, major muscles Of the upper leg
contract about the femur, and flexion of the hip is accomplished. In essence, a
force, or a load, has been placed on the femur by soft tissues, including muscles,
tendons, and ligaments. Likewise, contracting tissues Of the upper axial skeleton
cause the humerus to flex about the shoulder, propelling the arm through space.

Altemately, the intensities Of the forces that are generated within, or
internal to, the structure, in response to loads are called stresses (Garn, 1970;
Hayes, 1986; Hayes and Carter, 1979). Stresses represent a given load per unit
area which develops on a plane surface within a structure in response to
externally applied loads. Hence, at the microstructural level, stresses (the three
principle types being tension, compression, and shear) can be thought Of as
representing the resistance Of the intermolecular bonds within bone to deformation
by the external (i.e., load) forces (Frost, 1967).

In order to interpret the potential that differences in physical activity have
in influencing skeletal structure, it is helpful to dichotomize those aspects of bone
morphology that are genetically as Opposed to environmentally (i.e., functionally)
determined. General bone type, that is, for example, whether a developing bone

becomes a femur or a calcaneus, and the location of ligamentous attachment

21

sites on external bone surfaces, are genetically determined. As such, they will
occur in the absence Of functional (i.e., load-bearing) influences (Lanyon, 1986).
However, a bone’s unique characteristics, for example, its cross-sectional shape
or its cortical or trabecular thickness at a given diaphyseal location, are largely
dictated by the given load-bearing or activity-related conditions confronted by
each individual during his/her lifetime. These non-genetically determined
characteristics Of bone morphology will continue to be manifest if functional load-
bearing is maintained with sufficient magnitude and/or duration. Indeed, a
biomechanically competent bone can be said to be one in which skeletal structural
and functional requirements are coordinated (Lanyon, 1986). In order to maintain
this competence, the skeleton’s adaptive responses must actively match its
structure to its load-bearing requirements (and hence the intricate relationship
between form and function) by ensuring that enough skeletal tissue Of proper
quantity and quality is maintained.

At one extreme, a lack Of function (i.e., insufficient loading) may result in
disuse atrophy, which may in turn lead to the loss of bone tissue, or osteoporosis
(Kerr et al., 1996). At the other extreme, abnormally increased functional
demands placed upon bone (i.e., overloading) may result in the formation Of more
than a normal amount Of bone, which can lead to osteosclerosis (Ortner and
Putschar, 1981). It is evident, then, that any functionally appropriate level of bone
tissue is only maintained under the influence of adequate load-bearing stimuli.

Finally, at the physiological level, a biomechanically competent bone also

22

represents an equilibrium having been attained between skeletal apposition and
resorption, by osteoblasts and osteoclasts, respectively. Although influenced by
factors other than weight bearing (e.g., Circulating hormonal levels associated with
mineral metabolism), there is no physiological need, per se, for the skeleton to
ultimately achieve any particular shape. Thus, it is logical that the primary stimuli
capable Of organizing bone architecture are those provided by the stresses

associated with load bearing itself (Lanyon, 1986, 1996).

Skeletal Structure and Engingring Beam Theorv

 

One approach to.the analysis Of the stresses, strengths, and rigidity
associated with human long bone diaphyses is the use Of conventional civil and
mechanical engineering beam theory. Here, the assumption is made that long
bones are structurally analogous tO hollow beams (Ruff and Hayes, 1983a). Such
beams are slender structures (hence the analogy tO long bones) that are designed
to resist bending loads. From a structural standpoint, and most germane to the
present study, is the fact that beams, like human long bones, are long in
comparison to their cross-sectional dimensions (Lanyon, 1986).

In the present study, two bones Of the appendicular skeleton, the femur and
humerus, are examined. Like all Of the body’s major long bones, they are
relatively long, tubular-like elements; as such, they tend to be dominated by
bending and torsionally-oriented forces directed toward the joint surfaces. Long

bone diaphyses, of course, are not replicas Of a perfect cylinder, which can be

23

thought of as representing an unloaded bone. Indeed, it is an interest in the
degree and direction Of departure from this idealized shape that propels the
analytical approach taken in studies Of human long bone biomechanical
properties. As they are irregularly shaped, albeit to varying degrees, long bones,
in vivo, are subject to a combination Of compressive and bending loads during
normal activity (Hayes, 1986; Larsen, 1997). Under these conditions, the most
efficient cross-sectional shape is the generalized cylindrical tube. A tubular
structure can better distribute the stresses imposed by bending and torsional
loading than can a solid structure Of the same cross-sectional area, since the
material is distributed at a greater distance from the central axis. The functional
significance Of these different forces from a biomechanical standpoint is discussed
below within the context Of beam analysis.

The above discussion Of mechanical loading suggests that there are key
components Of bone geometry that are important for understanding the structural
behavior of whole bones (Parfitt, 1988). These components can be measured
through beam analysis, whereby cross-sectional geometric properties, which
represent a measure of the amount and distribution of bone within a given
section, are useful measures Of skeletal adaptation. The specific cross-sectional
geometric properties included in this study represent measures Of cross-sectional
area and moments of inertia (also known as second moments Of area).

Long bone cross-sectional areas represent a measure of a section of bone

"cut" perpendicular to its longitudinal axis. These areas are generally divided into

24

three types: cortical, medullary, and total subperiosteal (or simply total), each
representing the amount Of space taken up by skeletal tissue (or empty space in
the case Of medullary area) in the given cross-section. As a relatively simple
example Of the importance Of cross-sectional area in maintaining skeletal integrity
is the recognition that, when subjected to pure axial loading, a bone with a larger
cortical area will less easily fracture (i.e., is stronger) compared with a bone Of
smaller area because the former is able to distribute internal forces over a larger
area. Also, under in vitro, carefully controlled laboratory conditions, the
mechanical properties of cortical bone have been tested by artificially loading
small, uniform specimens (Hayes and Carter, 1979; Hayes, 1986). Such testing
produces known stresses throughout the specimen, and bone deformation can be
measured and the stresses calculated. The mechanical properties Of the bone
can thus be determined for the specific loading conditions imposed. This general
engineering approach to biological materials, and to skeletal tissues in particular,
has elucidated the nature Of skeletal tissues under conditions Of tension,
compression, bending, and torsion (Burstein ef al., 1972; Burstein et al., 1975;
Hayes and Carter, 1979; Reilly and Burstein, 1974). From this work we have
Ieamed, for example, that cortical bone is stronger when oriented longitudinally
than it is when oriented transversely.

Long bone moments of inertia reflect a measure Of torque or the quantity
Of force necessary to angularly accelerate the bone through space. Moments Of

inertia are always expressed in relation to a reference axis; for example, one

25

situated at the center, or centroid, Of the area in question. Moments Of inertia
calculated with respect to transverse and vertical axes are denoted IX and l,,
respectively. The moment Of inertia is highly sensitive to the distribution Of the
bone area with respect to the reference axis. For example, Skeletal tissue that
is situated a greater distance from the neutral axis is more efficient in resisting
bending with respect to that axis. Hence, an area moment Of inertia (designated
l,) refers to a quantity Of force, and takes into account the cross-sectional area
and distribution Of material around a given axis during loading. The polar moment
Of inertia (designated J) also represents a quantity which takes into account the
cross-sectional area and distribution Of material around a neutral axis, but
specifically in torsional loading. These and other biomechanical properties are

discussed in greater detail in Chapter 4, Materials and Methods.

Appligtion Of CT to Bioanthropolpgical Issues

In a clinical setting, CT scanning has primarily been used to provide
information that contributes to the formulation of a diagnosis. This goal is Often
achieved when CT is used in conjunction with other diagnostic applications, such
as ultrasound, conventional x-rays, and radioisotope scans. Today, many
scientists outside Of clinical medicine utilize CT’s unique imaging abilities, and
there now exists a flourishing literature on the relationship between CT technology
and mathematics, engineering, physics, and to a lesser extent, to biological

anthropology. Research-based CT provides an opportunity for biological

26

anthropologists tO examine, from an evolutionary perspective, the synergism
between the biological and cultural (i.e., behavioral) forces that give each human

(and non-human) primate skeleton its unique architectural characteristics.

Chapter 3

BIOARCHAEOLOGY

In order to examine the postcranial skeletal structural adaptations
associated with the femora and humeri from the archaeological samples utilized
in this study, a comparative bioanthropological framework is needed. As outlined
below, there now exists a fairly extensive body of literature that can be used to

help generate a series of hypotheses and expectations for the current research.

Previous Research

One of the earliest attempts to study human long bone diaphyseal
structural adaptations in archaeological remains with the use Of computed
tomography was the work on the Pecos Pueblo sample from the prehistoric
southwestern United States (Ruff, 1981; Ruff and Hayes, 1983a,b; Ruff and
Hayes, 1982). This research showed that with a large enough archaeological
sample it was possible to identify differences in femoral and tibial geometric
properties, and to infer certain patterns Of human behavior based upon those

differences. For example, this work documented that cortical bone area in the

27

28

Pecos sample is greatest between midshaft and the proximal end for both the
femur and tibia, and that total subperiosteal and medullary areas are greatest in
cross-sections closest to the knee joint. In terms Of strength, the maximum
principal second moments Of inertia are greatest between midshaft and the
proximal end of the tibia. In the femur, the maximum bending strength (IN) is
found at the proximal and distal extremities; it is least at midshaft. Locational
differences in the femoral and tibial cross-sectional geometry generally
corresponded to those predicted from in vivo loading experiments, whereby bone
area is distributed so as to minimize loading stresses (see Chapter 2).

In terms of sex-specific differences in cross-sectional geometric properties,
Ruff found that male femora and tibiae were stronger in the direction Of
anteroposterior (A—P) bending, whereas female femora and tibiae were stronger
in a mediolateral (M-L) direction (Ruff, 1981). With respect to cross-sectional
areas, males were shown to have relatively greater cortical area (CA) in the mid-
distal femora and in the mid-proximal tibia, diaphyseal sites Of probable greatest
A-P bending stress, while females had relatively greater CA in the proximal femur
and distal tibia, sites of high ML and torsional stresses, respectively. These
differences were interpreted as likely being due to a combination Of pelvic sexual
dimorphism (i.e., variability in bending moments based on anatomical differences),
and to sex-specific differences in habitual activity levels within the Pecos
population. Here Ruff (1981) sites ethnohistorical and historic evidence indicating

that many groups Of the American Southwest, including the inhabitants Of Jemez

29

Pueblo, who are most closely allied to the Pecos males, participated in activities
(e.g., long distance running in the context Of informal races, institutionalized
ceremonial events, and formal betting races between individuals) that would have
placed high A-P bending stresses on the lower limb.

Also working with skeletal remains from the prehistoric Southwest, Brock
(1985) and Brock and Ruff (1988) have investigated femoral and tibial adaptations
from a biomechanical perspective by quantifying bone robusticity in terms Of shifts
in activity levels with respect to time, space, and inter- and intra-sexual
differences. The skeletal samples were derived from a temporally diverse series
Of archaeological sites represented by numerous Mogollon, Anasazi, Anasazi-
Chaco, and Anasazi-Gallina cultures spanning from AD. 900 to AD. 1540. The
primary hypothesis tested was that, through time, the shift from a hunting and
gathering to an agricultural adaptive strategy should produce associated adaptive
skeletal remodeling that reflects different mechanical needs. Specifically,
measures Of skeletal robusticity (e.g., size and cross-sectional moments Of inertia)
should be lower in the more sedentary agricultural groups. Their results indicate
that there was considerable temporal fluctuation with respect to skeletal
robusticity, and that the more mobile late prehistoric farming groups demonstrated
greater than expected lower limb strength, arguing against a reduction in physical
workload in these groups.

Sumner (1984) and Sumner et al. (1985) has utilized anthropometric,

computerized tomographic, and photon absorptiometric techniques to examine

30
femoral growth and aging in the prehistoric Grasshopper Pueblo population (A.D.

1275-A.0. 1400) from Arizona in order to examine the relationship between
changes in whole bone cross-sectional geometric properties and bone mineral
content. His results‘show that, during growth, the structural properties Of the
femur increase more than its material properties, that proximal femoral diaphysis
morphology depends more upon the angle of antetorsion than on the
cervicodiaphyseal angle (two angles related to femoral biomechanics), and that
the femur may structurally compensate for the loss Of bone mass with aging.

In another study, Bridges (1985, 1989a, 1989b, 1991) used computerized
tomography tO examine all of the major long bones Of the upper and lower
appendicular skeleton in archaeological populations from the prehistoric
southeastem United States. The primary focus Of this work was to examine
biomechanically-relevant skeletal structural adaptations accompanying the
transition from hunting and gathering (Archaic period) to agriculture (Mississippian
period) in the Pickwick Basin area Of northwestern Alabama. Bridges’s research
demonstrates that some diaphyseal external dimensions, for both sexes, are
larger in the Mississippian group, as are a number Of the biomechanical variables
of the femur and humerus. Taken in combination, the results indicate larger,
stronger, long bone shafts in the Mississippian compared with the Archaic groups,
and argue for an increased work load in the former.

Over the past several years, copious studies have been undertaken by

Larsen and co-workers that have sought to document human long bone skeletal

31

structural adaptations from prehistoric preagricultural, prehistoric agricultural, and
historic period agricuturalists from the Georgia and Florida Atlantic coast region
(Fresia et al., 1990; Larsen, 1982, 1984, 1987, 1990, 1995, 1997; Larsen and
Ruff, 1991; Ruff et al., 1984). In summary, this research has shown that femoral
structural strength declines between the precontact preagricultural and precontact
agricultural periods for both males and females. This trend is then reversed for
the contact period agriculturalists, where femoral strength increases. Increased
long distance travel in the post-contact (mission) period, is one Of the specific
behaviors implicated to help explain high strength-related cross-sectional
properties in the femoral midshaft of some males. This contention is supported
by ethnohistorical accounts from the region indicating that the Spanish in the area
sometimes forced some males into the repartimiento labor system, and
subsequently ordered them to make long distance trips tO St. Augustine and
elsewhere (Bushnell, 1981; Ruff and Larsen, 1990). Other important factors that
undoubtedly impacted the activity patterns (and hence the skeletal structure) of
the contact period inhabitants in the area included epidemics, labor demands,
retaliation by the Spanish following revolts, attacks by the English-occupied
Carolina colony, increased population centralization, and an overuse Of already
marginal soils (Larsen, 1990).

The pattern Of decrease, then increase, in diaphyseal strength between the
three time periods also holds for male humeri; for females, the initial decrease in

the precontact agricultural period holds, but does not reverse in the postcontact

32

agricultural group. It is also worthy of note that, in general, the results Of
Bridges’s work discussed above are contradictory (Ruff and Larsen, 1990) tO
those from the Georgia coast. This is especially true for changes in measures Of
long bone length and for cross-sectional geometric properties relative to length.
These differences likely reflect regionally-based adaptations to local environmental

and biocultural circumstances (Ruff and Larsen, 1990).

mm

Seven separate Native American skeletal samples were used in the current
research. Four are from Michigan, two each from the prehistoric and historic
periods, and three are from New York. State, with two from the prehistoric period
and one from the historic period. Within each geographical area, the sites are
presented from earliest to latest. Only a brief review of each site is given below.
Numerous aspects Of the activity-related behaviors that may pertain to the types
Of long bone skeletal adaptations focused on in this research are more fully

examined in Chapter 6, Discussion.

Michigan Prehistoric Period
The Riviera aux Vase Site
The Riviera aux Vase Site (ca. A.0. 1000-1300) is an extensive Late
Woodland mortuary locale in Chesterfield Township, Macomb County, in

southeastern Michigan. The site is located north Of the city of Mount Clemens,

33

just inland from the western shore Of Lake St. Clair. Geologically, Riviera aux
Vase lies within the Port Huron morainic system Of the Huron-Erie ice lobe, which
dates to the last stage Of Pleistocene glaciation (Harper, 1945). The site was
excavated by the University Of Michigan during two separate field seasons in the
summers of 1936 and 1937. The remains Of some 370 individuals from 145
burials were excavated (Wilkinson and Van Wagenen, 1993), with many having
associated grave goods (Bender, 1979).

Riviera aux Vase represents a multicomponent occupation, consisting of
several burials from the early Late Woodland (ca. AD. 800) period. The principal
Late Woodland occupation from which the burials examined for this research were
derived have been dated by means Of ceramic typology and assigned a date of
AD. 1000-1300 (Fitting, 1965; Fitting and Zurel 1976; Krakker, 1983).
Subsistence was likely a mixture Of hunting, gathering, and agriculture. The site’s
temporal position places it solidly within the rise Of agricultural pursuits in the
Great Lakes area (Bender, 1979; Wilkinson and Van Wagenen, 1993), and
Krakker (1983) has argued that the settlement pattern data points to agriculture
as being the prominent subsistence activity at the site. Fitting (n.d.) believes that
the site was revisited from time to time for the purpose Of mortuary
ceremonialism. The settlement pattern associated with the individuals using the
site as a cemetery is not known; no archaeological remains Of structural features,
save for burials and refuse pits, were recovered (Greenman, 1957). However,

based on analogy with roughly contemporaneous sites located in Ontario, a

34

seasonal settlement pattern is suggested, with groups concentrated along rivers
and lakes in spring and summer.

Much Of what can be learned from the site’s archaeological context must
be gleaned from the field notes of the 1936 and 1937 excavation seasons.
Although the total number Of individuals excavated appears to be 377, a more
reasonable estimate Of the total number Of individuals in the skeletal collection
available for study is actually 300-325, based on an examination of burial
photographs, and on a reconstruction Of age and sex (Raemsch, 1993). Adults,
children, and infants are all represented within the skeletal collection.

The burials were in a mixed state of preservation, ranging from excellent
to too poorly preserved to excavate. The burial types were numerous, in various
positions and arrangements. Burials were primary (both extended and flexed),
single and multiple, the latter with their heads oriented in the same direction or
at opposite points Of the compass. Some burials were nearly complete, others
consisted of only a cranium. Numerous crania and long bones exhibited post-
mortem modification, including drill holes and disc removal; some had clay placed
within and/or over the facial cavities and bones. Numerous femora had shaved
heads placed within the Obturator foramina, perhaps modified for the purpose of
re-articulation. Also, the ends Of numerous humeri and femora were perforated
by drilled holes.

Bender (1979) has reported on the paleodemography of the site, and

Wilkinson and Van Wagenen (1993) have examined evidence for cranial trauma

35

and interpersonal violence. Raemsch (1993) has generated a classification Of
cutrnarks based on postmortem skeletal modifications, their anatomical location
and morphology, discussing how cutmarks, disarticulation, and defleshing can be
used to reconstruct certain mortuary behaviors related to the Riviera aux Vase
population. A more general consideration Of these and other potentially related
post-mortem skeletal modifications, including their specific morphological
characteristics and geographical distribution, can be found in Hinsdale and
Greenman (1936), Hinsdale and Cappannari (1941), Raemsch (1993), and in
Wilkinson and Van Wagenen (1993).

The age and sex Of the Riviera aux Vase remains were determined by
standard gross morphological Observations, including pelvic morphology and
associated auricular surfaces and pubic symphyses, general patterns Of sexual
dimorphism, dental wear, ectocranial suture closure, and Observations Of tooth
loss and vault thickness and density in older individuals. Where cranial and post-
cranial elements were not associated (i.e., were found in multiple intemments),
craniO-facial and dental Observations were used to determine age and sex. Five-
year age categories were used for individuals between 16 and 50; 10 year

intervals were used for older individuals (Wilkinson and Van Wagenen, 1993).

The Juntunen Site
The Juntunen site (20MK1) is a Late Woodland period occupation located

along the northwestern edge Of BOis Blanc Island in the Straits Of Mackinac

36
between Lakes Huron and Michigan, Mackinac County, Michigan. The site was

excavated by the University Of Michigan Museum of Anthropology in 1960—61.
The occupation Of the site spans some 600 years, with radiocarbon dates ranging,
by occupation zone, from AD. 830 to AD. 1330. The skeletal remains examined
in this research come from the later time period. McPherron (1967) has written
the seminal work on the Juntunen site, wherein he provides extensive discussions
Of the archaeological background, paleoecology, and subsistence economy.

The site has been termed a typical woodland adaptation, with the utilization
Of a diversity Of the local fauna but with emphasis on fishing (McPherron,'1967).
Yamell (1964) has addressed the paleoethnobotonical aspects Of the site, and
has emphasized the relationship between the site’s location, climatic factors,
seasonality Of occupation, and subsistence. He reconstructs a summer fishing
and winter hunting and trapping subsistence economy, supporting his model by
ethnographic analogy with the Round Lake Ojibwa and Mistassini. Also, Wright
(n.d.) has written on the paleO-ecology of the site, Hamilton at al. (1984) has
reported on the perishables, and Cleland (1965) has examined animal burials as
evidence for ceremonial activity and has assessed the site’s paleoecology from
plant and animal remains.

There are some 55 individuals available for research. This group Of burials
dates to approximately AD. 1331 1 120 years (manuscript on file, University Of
Michigan Museum Of Anthropology).

The initial Juntunen site physical anthropology was done by Eyman and

37

Betteral (1965). Their determination of sex was primarily based on Giles and
Elliot discriminant function formulae, and all results were compared with non-
metric indicators of sex. Adult age categories were assigned as young, middle
and Old (nine young, 23 middle, and three Old), or 35 individuals.

Of particular interest to the present study are three observations made by
the authors. One, Eyman and Betteral (1965) feel that in terms Of stature, body
build, and general morphological attributes, the Juntunen Skeletal remains are
quite similar to other Michigan Woodland skeletal samples (e.g., Norton Mounds).
Second, they note that the Juntunen humeri display a marked hypertrophy in the
region Of the deltoid muscle attachment. They leave open-ended the question as
to whether this might reflect a specific skeletal adaptation to a lacustrine
subsistence economy. Finally, they report that the Juntunen people are
moderately short, at least compared to an Early Historic period ossuary sample

from St. Ignace, with robust faces and upper arms.

Michigan Historic Period
The Lasanen Site
The Lasanen site (20MA21) is located in Mackinac County, Michigan, within
the city limits Of St. Ignace in Michigan’s upper peninsula. The site was
excavated by the Michigan State University Museum in 1966 under the direction
of Charles Cleland.

Much Of what has been published on the Lasanen site can be found in

38

Cleland (1971), where there is a particularly useful discussion on the burial
excavations, the rich array of grave artifacts, and the time period Of the site with
respect to northern Michigan’s early post-contact period. Also, O’Shea (1988) has
reanalyzed the mortuary practice from the site, examining burial behavior within
the wider context of other Michigan mortuary sites.

Although the subsistence regime Of the Lasanen site inhabitants is difficult
to establish, it can be inferred from subsistence practices in the St. Ignace region
that the 1670-1700 time period is most likely represented by the site. Subsistence
practices in that region likely were based on hunting and fishing in the late fall
and early winter, with agriculture playing an important role during the late summer
and early fall seasons (Cleland, 1971). There are numerous accounts in The
Jesuit Relations (T hwaites, 1896-1901), indicating that agriculture, and specifically
the production of corn, beans, peas, and squash, was practiced in the region
extensively. By the time Of the Lasanen site occupation, the agricultural pursuits
were likely at least partially dependent on European implements (Cleland, 1971).

Overall, an AD. 1670-1700 date has been established for the site
(Cleland, 1971). This determination is based on artifact evidence (both European
and Native), comparative site studies (especially to the AD. 1685 - 1696 Pan site
from Onondaga County, New York), and burial type (Cleland, 1971).

The Lasanen site physical anthropology has been reported by Clute (1971).

In general, the skeletal material from the site is in excellent condition.

39
The Fletcher Site
The Fletcher Site (203Y28) is a large, multicomponent mid- to late-

eighteenth century burial site located on the western bank of the Saginaw River
in Bay City, Michigan (Lovis, 1985). The site has been dated primarily on the
basis Of comparative artifact evidence and on ethnohistorical grounds with respect
to population dynamics (Mainfort, 1977, 1979). The human remains from the Site
represent a northern or central Algonquian people (Sauer, 1974; Mainfort, 1977;
Nelson, n.d.).

The site yielded relatively few faunal or floral remains, hence the diet of the
people represented by the Fletcher remains cannot be reconstructed with any
degree of certainty. Based on ethnohistorical analogy to the Ojibwa or Ottawa,
the groups likely represented in the cemetery, Mainfort (1977) has constructed a
generalized subsistence model involving the hunting and gathering of wild
foodstuffs during the summer and winter, with the spring planting and fall
harvesting of cultigens such as squash and corn. Given the inclusion Of cultigens
in the diet, it has been argued by Nelson (1985) that subsistence at the Fletcher
site can be thought of as primarily agricultural. Lovis (1985) and Brashler and
Holman (1985) have discussed the role Of the Fletcher site within the larger
Saginaw Valley geographic zone with respect to seasonality, settlement patterns,
and cultural adaptation during the Late Woodland period.

Some 93 middle Historic Period (A.0. 1750-1770) individuals were

excavated by archaeologists from Michigan State University from 1967 to 1970;

40

these individuals represent the potential sample for the current study. Sauer
(1974) has reported on the physical anthropology Of the site, and Mainfort (1977,
1985) and O’Shea (1988) have addressed Fletcher site mortuary behavior.
Sauer’s (1974) major findings are that the Fletcher inhabitants were
impacted by nutritional stress as a consequence Of European contact, exemplified

by a high incidence Of osteoporosis, short stature, and decreased life expectancy.

New York Prehistoric Period

The Frontenac leland Site
The Frontenac Island site is located in Cayuga County, New York. An
extremely large site given its early time-frame, the Island (the only one in New
York’s Finger Lakes region) is located in Lake Cayuga. It was excavated in 1939-
40 and, in conjunction with the Lamoka Lake and Brewerton sites, it provided the
background for Ritchie’s reconstruction Of central New York’s Archaic period

(Ritchie, 1944, 1945, 1965).
Radiocarbon dates for the Frontenac Island site range from 2980 BC. 1
260 through 2013 BC. 1 80 to 1723 BC. 1 250. Excavations produced 159
burials, all belonging to the Archaic period. Just under 100 Of these burials were
available for study at the Rochester Museum and Science Center (RMSC),
Rochester, New York. Bone preservation was generally good; however, for the
present study the potential sample was greatly reduced due to numerous

"restorations" to the long bones (also see Pfeiffer, 1977). Many Of the long bones

41
have been altered (not by current RMSC staff) by having fragments glued

together and/or missing bone filled in with plaster. Given the Obvious potential
problems these artificial alterations might have on accurately determining external
morphometry and internal cross—sectional geometry, any such bone was
eliminated from study.

The Frontenac Island demographic data were taken from Pfeiffer (1977),
who was kind enough to share with me her assessment of age and sex,
independent from those data on file at the RMSC. Adult ages were divided into
four broad stages, primarily based on cranial suture closure. The sex ratio for the

sample is 54% male and. 46% female (Pfeiffer, 1977).

The Harscher SE a
The Harscher site (Can 38-1) is located in the town Of East Bloomfield,

Ontario County, New York. Most Of what is know Of the archaeology of the site
has been reported by Wray and Cameron (1968). All age and sex data have
been determined by Lorraine Saunders at the Rochester Museum and Science
Center. The site consists Of a village and cemetery, and having been dated to
AD. 1500, it represents a Late Prehistoric lroquoian site. NO evidence Of
European trade goods were recovered from the site. Thirty-three burials, 32
single, and one multiple (two individuals), were excavated by Charles Wray and
others during the summer and fall Of 1968. Burials were found in sandy loam at

depths ranging from 12 to 46 inches, with an average depth of 29 inches. Most

42

Of the individuals in the cemetery were oriented with their heads facing either east
or south. All Of the burials except one (that of an infant, which was extended)
were in the flexed position. Approximately one-third Of the burials were oriented
on their stomachs with the legs partially-flexed, with most Of the remaining burials
flexed on their sides.

Several burials had evidence Of ornamentation (e.g., jaws probably
associated with a ceremonial bobcat headdress, small mammal jaws, teeth and
bones (weasel) as ornamentation for a pipe pouch, small mammal teeth from
clothing, heron bills near the head-probably headdress ornaments). Only two Of
the burials had associated grave offerings per se, consisting Of worked deer antler
and cutting tools Of flint. It is assumed that at Harscher there was no formal

custom to include burial Offerings with the dead (Wray and Cameron, 1968).

New York Historic Period
The Tram Site

The Tram site (Hne 6-4) is located near the village of Livonia in Livonia
Township, Livingston County, New York. The site has been known since at least
1848, when E.G. Squier mapped Indian fortifications of western New York State
for the Historical Society Of New York and the Smithsonian Institution (Wray and
Cameron, 1970). The village portion of the site encompasses about 20‘ acres.
The stockaded enclosure is about 16 acres in size, consisting Of a long oval

earthwork with four gates, and is positioned on the summit Of a prominent hill,

43

with steep embankments and deep entrenchment ditches constructed for defense.

The cemetery portion Of the site is sandy clay loam, with pit depths ranging
from 20 inches to 58 inches; the average depth is 38 inches. Portions Of the site
have been excavated over the years; the burials reported on here were excavated
by Wray and Cameron in 1970, revealing some 62 individuals.

Slightly over one-half Of the burials had grave goods; 11 (18%) had
European goods. These were likely brought to the site through trade; there is no
evidence Of European peoples having visited the site (Saunders, personal
communication). At least 51 Of the burials were single, five were double, and one
was triple. Most were tightly flexed; approximately 60% had heads facing west.

Establishing the temporal position Of the site has been the subject of much
discussion and research for many years (see for example, Houghton, 1927; Wray
at al., 1991). The most recent estimation dates the site to A0. 1580 - 1600
(Wray at al., 1991), with European contact in the region at about AD. 1550. The
site, then, represents a very early Seneca Historic, or post-contact site. The
physical anthropology on the Tram Site skeletal remains has been presented in

great detail by Saunders (Wray at al., 1991).

W

As indicated in the Introduction and earlier in this chapter, differences in
femoral and humeral size, shape, and strength have been documented for

numerous skeletal populations from a wide range of temporal and geographic

44

contexts within North America. lmportantly, the results Of these past studies have
brought to light the general patterns Of skeletal structural adaptation that can be
expected to result from modifications in physical activity in a variety Of biocultural
contexts. These studies can provide an interpretive backdrop against which the
results and expectations of the present research can be evaluated. In keeping
with the fact that the Michigan and New York skeletal samples are being treated
as two separate case studies, the anthropometric and biomechanical data sets
are also evaluated independently, as follows.

The first set Of hypotheses relate to the anthropometric and biomechanical

data sets, respectively, for the Michigan skeletal samples.

Hypothesis 1

There will be no difference between the Michigan prehistoric and historic
period samples with respect to anthropometric measures Of long bone size.

Expe_ct_a1ion

If this hypothesis is rejected, the results Of the previously discussed
research will not help explain the direction Of change in femoral and
humeral size in the historic period sample. The only other comparable
work to the current study in terms Of specifically examining pre- and
postcontact skeletal adaptations is that from the Georgia and Florida coast
(see the above discussion, and especially Larsen, 1990). Many Of the
behavioral factors that have been implicated to explain the skeletal
structural changes in these regions Of the country do not have direct
parallels in Michigan.

45
Hypothesis 2

- There will be no difference between the Michigan prehistoric and historic
period samples with respect to long bone cross-sectional biomechanical
properties.

Expectation

The expectation, as outlined for Hypothesis 1, is repeated here, except that
"change in femoral and humeral size” is replaced with "Change in femoral
and humeral cross-sectional dimensions and strength-related properties."

The second set Of major hypotheses relate tO the anthropometric and

biomechanical data sets, respectively, for the New York skeletal samples.

Hypothesis 3

There will be no difference between the New York hunter-gatherer and
agricultural samples with respect to anthropometric measures of long bone
size.

Exmctation

If this hypothesis is rejected, the results Of the previously discussed
research support the expectation that femoral and humeral size will be
greater in the agricultural sample (see for example, Bridges, 1985, 1989a,
1989b, 1991; Brock, 1985; Brock and Ruff, 1988).

Hypothesis 4

There will be nO difference between the New York hunter—gatherer and
agricultural samples with respect to long bone cross-sectional
biomechanical properties.

46
mm

The expectation, as outlined for Hypothesis 3, is repeated here, except that
"femoral an humeral size” is replaced with "long bone cross-sectional
biomechanical properties."

Chapter 4

MATERIALS AND METHODS

Skeletal Sample Selection

In selecting the data for the femoral and humeral anthropometric and
biomechanical analyses, several important decisions had to be made regarding
inclusion and exclusion criteria for these two separate but complimentary data
sets. Initially, only adult'individuals were included, "adult" meaning that all
proximal and distal epiphyses were fused. Also, at least one bone, right or left,
had to be complete enough for measurement. By "complete" is meant that the
bone had to be intact to the extent that there was a reasonable degree of
confidence in discerning the majority Of the appropriate external anthropometric
and biomechanically-related anatomical landmarks. As such, fragmentary
remains were included, but no measurements were taken from those portions
Of femora or humeri manifesting a pathologic condition on any external
surface, and hence possibly altering one or more external dimension(s).

Because the long bone size, shape, and strength characteristics

examined in this research are potentially influenced by sex and age (i.e.,

47

48

genetic) and/or lifestyle factors (i.e., activity-related), the skeletal sample was
further reduced by eliminating those individuals for whom sex could not
confidently be determined. Additionally, it must be kept in mind that the
number Of age categories, as well as the age range encompassed within each
Of those categories, are different for some of the sites because different
physical anthropologists have been responsible for each of the samples (the
exception being the Tram and Harscher sites, where the age and sex data was
provided by Lorraine Saunders). In order to control for age and yet achieve a
respectable sample size in light of the demands on the data as outlined above,
age ranges were collapsed into two broad categories. The younger group is
comprised Of individuals approximately 20-40 years Old, and the Older group is
composed Of individuals 40+ years Old. The decision to use the younger group
was made for both statistical and biological reasons. Specifically, (1) using
this group resulted in a larger sample size and (2) overall, the long bones
comprising the younger subsample are potentially less affected by the type of
age-related bone loss (i.e., due to osteoporosis) that might make the results Of
certain Of the CT measurements (e.g., the cross-sectional cortical and
medullary area dimensions) difficult to interpret.

Since human long bones may also manifest dimensional differences
based on side, the decision was made to include only one side, the left,
thereby eliminating the potential effects Of bilateral asymmetry within

individuals. The decision to use a single side for analytical purposes is

49

supported by both historical and pragmatic considerations. For example, in
most studies Of human skeletal morphometry, if one side Of a bilaterally-
occurring element is Chosen for study, traditionally it is the left. For the
present study, it was possible to examine a total Of 383 femora derived from
225 different individuals. This represents 85% (383l450-the total number Of
possible femora from 225 individuals) Of the total potential sample of
individuals. Also, based on the inclusion criteria outlined above, a nearly
identical number Of femora by side (192 right and 191 left) were available for
examination (Table I). This tOO constitutes approximately 85% (191 left
femora/225- different individuals) Of the total possible femora; hence, there is
no femoral sample selection bias based on choosing the left side for study.
It was possible to measure a total Of 347 humeri derived from 187
different individuals. This represents 93% (347/374-the total number of
possible humeri from 187 individuals) Of the total potential sample Of
individuals. Similar to the case for the femora, a nearly identical number of
humeri, by side (170 right and 177 left), could potentially be examined after the
selection criteria were applied (T able II). This represents approximately 95%
(177 left humeri/187 different individuals) of the total possible humeri; hence,
there is no humeral sample selection bias based on using the left side for
study. Both long bone subsamples are representative of the larger skeletal
samples from which they are derived, there being approximately 10% more

individuals from whom humeri compared with femora could be studied.

50
AS such, this sample constituted the data for the first phase of the study;

that is, a whole bone anthropometric comparison Of changes in the external
dimensions Of femora and humeri between the prehistoric and historic period
Michigan groups, and between the hunter-gatherer and agricultural New York
State groups.

Given the need to establish a consistent orientation with respect to CT
scanning. and cross-sectional geometric property determination, the long bone
sample available for the biomechanical phase Of research was comprised Of a
selective subset utilized for the anthropometric analyses. Tables Ill and IV
provide a breakdown, by sex and side, Of the potential bomechanical sample Of
femora and humeri, respectively. These Tables demonstrate that, for femoral
and humeral side comparisons, within sex, in all but one case (i.e., that for left
vs. right male femora), left femora and humeri provided a larger sample size
compared to the right side. Comparing Table 1 with Table 3, and Table 2 with
Table 4, it can also be seen that, based on bone, sex, and left side, 51%
(50/98 for male left femora), 56% (51l91 for female left humeri), 57% (53/93 for
female left femora), and 63% (54/86 for male left humeri) Of the anthropometric
sample could be used for biomechanical analysis. Further details concerning

the manipulation Of both data sets for analytical purposes is discussed below.

51

Table 1

Femoral Sample‘ for Anthropometric Analysis: Breakdown By Sex and Side

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Males (n=1 18) Females (n=107)
Left I Right Left I Right
Site2 (n=98) i (n=102) E (n=93) i (n=90)
Michigan
Historic 22 21 24 24
Prehistoric 4O 44 37 31
New York State
Agricultural 22 22 25 27
_Hunter-Gatherer ._ 14 15 7 8 _=
‘ After selection criteria applied — see text.

2 Michigan historic period: Fletcher and Lasanen sites; Michigan prehistoric period:
Juntunen and Riviera aux Vase sites; New York agricultural: Tram and Harscher sites;

New York hunting—gathering: Frontenac Island site.

Table 2

Humeral Sample‘ for Anthropometric Analysis: Breakdown By Sex and Side

 

 

 

 

Michigan

y
Males (n=99) Females (n=88)
Left 5 Right Left ' Right I

 

   

        

 

  

 

  

 

 

 

 

  

 

  

 

    

 

‘ Alter selection criteria applied -
2 See Table 1 for explanation.

 

see text.

 

 

Historic 23 26 21 29 I
Prehistoric 39 33 36 31 E
New York State fl
Agricultural , 18 20 25 25
Hunter-Gatherer . 10 9 6

 

52
Table

3

Femoral Sample‘ for CT Analysis: Breakdown By Sex and Side

 

   
 
 

 

 

 

 

 

 

 

 

 

  

 

 

 

 

  

 

— L Malcs (n=111) - :males (n=101)
Left I Right Left I Right
(n=50) i (n=61) (n=53) : (n=48)
Historic 1 1 I 7 6 5
Prehistoric 15 I 26 24 15
New York State
19 19 23

 

 

 

 

‘ Alter selection criteria applied - see text. .
2 See Table 1 for explanation.

Table 4

 

 

 

 

Humeral Sample‘ for CT Analysis: Breakdown By Sex and Side

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Males (n=91) Females (n=83)
Left I Right Left I Right
Site2 (n=54) i (n=53) (n=51) I (n=45)
Michigan f
Historic 9 l 10 7 6
Prehistoric 28 I 23 21 19
New York State
Agricultural 13 15 I 17 I 17
Hunter-Gatherer 4 5 i I 6 l 3

 

 

 

‘ After selection criteria applied - see text.
2 See Table 1 for explanation.

 

 

53
Femoral and Humeral Whole Bpne Anthropometgy

Each Of the specific long bone measurements included in the present
study were chosen based upon both methodological and pragmatic criteria.
First, all measurements were associated with well-established, easily
recognizable, anatomical landmarks. The ability to precisely define and
describe measurement endpoints is critical to maximizing the consistency (and
hence minimizing intra- and inter-Observer error rates) required for these
measurements to be replicated by other researchers, and represents a sine
qua non for studies that rely heavily upon anthropometric analyses. Second,
there exists a growing body Of similarly-derived whole bone anthropometric
data from studies (e.g., Bridges, 1985, 1989a,b, 1991; Brock and Ruff, 1988;
Fresia at al., 1990; Larsen, 1995, 1997; Larsen et al., 1996; Larsen and Ruff,
1991; Ruff, 1981, 1984; Ruff and Hayes, 1982, 1983a,b; Ruff and Larsen,
1990; Ruff at al., 1984) against which the results Of this research are to be
compared.

Several standard anthropometric instruments were utilized in the present
study, including a sliding caliper, flexible tape, osteometric board, and the first
segment of an anthropometer. All measurements were recorded tO the nearest
millimeter (mm), and most can be found in Bass (1987) or White (1991).

Nineteen femoral and 14 humeral measurements were taken (Table 5 below).

Femoral and Humeral Whole Bone Anthropometry

 

Femoral Metrics

Measurement

length
maximum
bicondylar
biomechanical
maximum trochanteric

midshaft
anterOposterior diameter
mediolateral diameter
circumference

subtrochanteric
anteroposterior diameter
mediolateral diameter

head
maximum diameter
horizontal diameter
girth

neck
minimum height
horizontal breadth
girth

condylar length
medial
lateral

breadth
proximal epiphyseal
biepicondylar

Humeral Metrics

Measurement

length
maximum
biomechanical
lesser tubercle-trochlear
greater tubercle-medial
epicondylar
head medial epicondylar

midshaft
maximum diameter
minimum diameter
circumference

head
maximum diameter
horizontal diameter
girth

breadth
anterior trochlear
posterior trochlear
biepicondylar

 

55
Com uted Tom ra h . C Scannin

For the present research, a fourth generation Technicare 1440 high
performance scanner (Figure 1), manufactured by the Johnson & Johnson
Company, was used to generate all of the femoral and humeral cross-sectional
images. The CT unit’s gantry circumference measures approximately 176 cm,
with a diameter of approximately 57 cm at the center. Although a considerable
amount Of time and effort was expanded in order to ensure that each long
bone was properly positioned prior to scanning (see below), the actual
scanning time necessary to Obtain each "slice," or cross-sectional image, was
only about four seconds. At theonset Of each scanning session, a series Of
iterative scanner readings were manually entered into the CT machine’s control
display. For each femur and humerus that was scanned, the following 18
control display parameters were used (CT machine-prompts are denoted in
capital letters and the operator’s responses are in quotes, followed by
explanations where appropriate).

Technicare 1440 control dieplay parameters

1. AGE: "none": not applicable; used for the purpose Of patient
identification.

2. SEX: "m": denotes an arbitrarily defined male; also for patient
identification.

3. CONTRAST: "0": relevant only for patient scanning.

4. DIRECTION: "feet/supine" (for femora) and "head/supine" (for humeri).

56

These designations pertain to the orientation Of the long bones relative
to the scanning gantry. For each femur and humerus, the protocol was
set to accommodate the bone lying on the scanning box anterior surface
up with its distal and facing the gantry.

NUMBER OF SLICES: "1-3; 1-2” for all femora and "1-5" for all humeri.
Because the phantom is not long enough to fit under the most proximal
and distal scan positions for most adult femora, the CT machine was
initially set to take three scans, beginning distally and moving toward
midshaft. After the phantom was repositioned, the machine was set to
complete the remaining‘two scans, moving from midshaft toward the
proximal end Of the bone. Because the maximum distance needed to
perform all five hUmeral scans fell within the phantom’s length, the CT
machine was preset to complete all five scans, and nO repositioning Of
the phantom was necessary. Figure 2 illustrates, from top tO bottom,
CT images Of the left humerus from four different individuals.

TABLE DIRECTION: "in": the cross-sectional images were taken with
the scanning table mOving through (and not retracted from) the gantry.
DYNAMIC MODE: "no"

ALTER OTHER SPECIFICATIONS: "yes"

Since all femora and humeri were scanned at five locations, each
separated by a factor Of 15% Of biomechanical length (i.e., 20%, 35%,

50%, 65%, 80%), the computer was set to automatically scan at

57

precisely the correct distance along each long bone shaft. Not having to
manually enter all Of the five scanning distances saved considerable

time. Table 6 illustrates this point:

Table 6

Example Of CT Scan Number Calculation

 

humeral length: 300 mm

scan it: 1 2 3 4 5

% of total bone length: 20% 35% 50% 65% 80%
scanning positions (mm): 60 105 150 195 240
4 scan increments (mm): 45 45 45 45
(#1 -#2/#2-#3I#3-#4/#4-#5)

 

In this example, the scanning increments (i.e., the distance from
one scan to the next) are all 45 mm; hence, the positions for the first
two scans (60 mm and 105 mm) were entered into the computer, the
increment set to 45 mm, and the CT machine automatically and
sequentially (from scan #1 through scan #5) repositioned the movable
table (and hence the scanning box and bone-see below) to the correct
positions for the remaining three scans (i.e., at 150, 195, and 240 mm).
SLC: "1.0 mm": refers to the thickness of each slice or cross-sectional

image being generated.

10.

11.

12.

13.

14.
15.

16.

17.
18.

58
TLT: "0.0”: i.e., vertical, with no tilt to the gantry. As described in

Chapter 2, the scan beam is directed as perpendicular as possible to
the long axis Of the bone (see orientation box section below).

DIA: "202"

TBL-POSITION: locates the "zero" point from which the scanning
locations are calculated. NO actual scan is taken at this position, rather,
it refers to the anatomically-defined point (see bone orientation section
below) on the distal femur and humerus from which the first scan
position is determined.

SCAN NUMBER: "1-5": the total number Of scans performed on each
bone (Figure 2). See discussion below under Statistical Approach for
choice Of scans for analytical purposes.

PROTOCOL: "#308": performs a 400 mill‘iamp-seconds (MAS) scan.
SITE #: i.e., "Lasanen": the provenience Of the skeletal remains.

ID #: i.e., "4123-L-femur”: refers to the bone number (catalog/museum
number), side (left or right) and bone type (femur or humerus).

DATE: the format "xx-yy-zz"; the day, month and year Of the scan.

TIME: the format "xx-yy-zz"; the time Of the scan.

The role of the phantom in CT ecanning

The primary purpose Of the phantom is to allow for the calculation of

bone densities derived from the CT-generated cross-sectional images.

Although the analysis Of bone density was not within the immediate aims Of

59

this research project, relatively little extra time was needed in order tO have the
density data recorded onto disk. This was consistent with an important goal Of
the study, which was to maximize, within time and budget constraints, the
research potential Of the skeletal samples, given the possibility Of repatriation
for one or more Of the collections. The CT phantom used throughout the entire
study was approximately 300 mm long x 155 mm wide x 35 mm thick (at
center). The phantom has permanently embedded within it four calibration
columns, each about 20 mm in diameter. Each column is composed Of a
different density, arranged in the following sequence (viewed left to right in
cross-section): 200, 100, 50, 0 (open space, i.e., the phantom density itself),
and -100. These numbers are. in HU (Houndsfield Units). Along the bottom Of
the phantom (between the 200 and 100 and 0 and -100 density columns) are
positioned two permanently fixed tin (Sn) wires. These wires, which can be
readily seen on the computer monitor as it displays the reconstructed cross-
sectional image, allow the scanned Object to be reoriented relative to the
position of the phantom and its calibration columns. The ability to re-establish
the position Of each image in terms Of its original orientation on the scanning
box is important for accurately calculating the cross-sectional area properties

as well as the strength-related moments Of inertia.

The orientation box and CT sgnning

One Of the most challenging aspects of generating accurate cross-

sectional images from archaeologically-derived human long bones is devising a

60

consistent, reproducible, method for ensuring that all bones are scanned with
the same three-dimensional orientation in space. There are several different
ways by which human long bones from archaeological contexts have been
scanned in the past (see, for example, Ruff, 1981, and Bridges, 1985), each
investigator designing the scanning protocol to suite his/her particular study.
Because of the large number of scans (i.e., five) performed on each bone in
this study, an apparatus had to be built that would accommodate not only the
physical characteristics of the skeletal remains being examined, but also take
into consideration the number Of scans to be taken. After considerable trial
and error with respect to the most efficacious scanning system, and taking into
account both time and personnel considerations, the following orientation
apparatus (Figure 3 and Figure 4) and scanning protocol was devised, and

. ultimately proved to be a highly efficient and accurate means by which to
accomplish the CT scanning goals Of this project.

The primary purpose Of the orientation box developed for this study was
to provide a highly dependable means by which all femora and humeri could
be consistently oriented and safely propelled through the CT scanner. The
physical make-up Of the box itself (manufactured by N. Sauer)had to take into
account not only the highly sensitive nature of the scanning procedure in terms
of cross-sectional image generation and quality, but also the need for proper
long bone orientation with respect to three pre-determined reference axes.

The orientation box is composed Of numerous "layers" (Figure 4), each

61

providing an essential function and without which the entire scanning protocol
would be compromised. It is perhaps easiest to understand the nature Of the
box’s construction relative tO the goals Of CT scanning human long bones by
"taking" the reader through the entire scanning process.

The scanning box was constructed entirely Of wood. Because all femora
and humeri to be scanned would ultimately be placed on top Of the box’s
plexiglass platform (see description below), the orientation box would itself be
directly in the path Of the CT scan. Thus, it was essential that no metal/steel
fasteners Of any kind be used in the box’s construction. These and other very
high density Objects cause-"artifacts," or "streaking,” as the X-ray beam passes
through the Object being scanned. In order to avoid these problems, all box
corners were fitted with wooden dowels. The physical construction Of the
orientation box proved to be effective beyond expectation.

The first step in scanning each long bone involved retracting the
movable (patient) table from within the gantry to its fullest extent. Because the
table was conwve or "cup-shaped" in the middle, the orientation box did not
easily rest in a level plane across the tabletop. Hence, a rectangular sheet of
plywood (hereafter referred to as the platform), cut with straight edges so that
in width it was wide enough to securely hold the orientation box on top Of it,
but narrow enough to fit through the gantry Opening, was used. Several
centimeters from one edge of the platform was affixed a thin line Of black

marking tape. This line was carefully oriented parallel to the edge Of the

62

platform. Its purpose was to ensure that the platform could be properly and
consistently oriented relative to the X-ray beam at the beginning Of each
scanning session. This orientation was accomplished by turning the overhead
lights down in the scanning room, turning on a light that illuminated the
scanning path from within the gantry, and manually superimposing this beam
Of light onto the line Of black tape. Once the platform was properly aligned
relative to the scanning beam, it was fixed into position with straps that were
permanently attached to the movable scanning table. Care was taken not to
move the platform once it had been fixed into proper position. If accidentally
moved, the entire sequence described above had tO be repeated.

Directly on top Of the plexiglass platform was placed a phantom Slide.
This slide comprised one of the more unique features Of the overall scanning
apparatus. The phantom slide consisted of a roughly square section Of thin
plywood that rested directly on the top Of the platform. Two holes were drilled
into the ends of the slide that faced toward and away from the gantry opening.
Strong chords, long enough so that their ends were accessible when the
orientation box was in place (see below), were attached tO these holes and
pulled taut through square openings at each end of the box. On tOp Of the
phantom slide was placed a piece Of black felt measuring 20 mm long x 16
mm wide, the purpose Of which was to provide a modest amount Of friction
between the phantom slide and the phantom to prevent the latter from moving

once scanning commenced. The phantom was then centered on the slide with

63

the calibration columns parallel to the platform. The purpose Of the phantom
slide was to allow the phantom to be moved underneath the section Of femur
being scanned without moving either the platform or the box itself (see above
explanation for #5, NUMBER OF SLICES, under Technicare 1440 control
display parameters). On tOp of the phantom was placed a bolus bag,
measuring approximately 385 mm long x 190 mm wide. The bag has a
density Of zero, and functions to reduce scanning artifact as the beam passes
through the density Of the phantom and through the density Of air.

Next, the orientation box was placed on top of the platform. Initially, the
box was positioned to only. roughly fit within the inside edges Of four wooden
Chocks that had been permanently affixed to the platform and situated to
correspond with the four sides Of the box. Approximately one centimeter below
the tOp edge Of the box a single sheet Of plexiglass (hereafter referred to as a
plexiglass platform), upon which an individual bone was placed, was fixed into
all four Sides. The phantom slide, black felt, phantom and bolus bag all fit
under this plexiglass platform on tOp Of the wooden box platform. The top Of
the bolus bag was situated just below the underside surface Of the plexiglass
platform. Several centimeters from the end Of the apparatus facing into the
scanning gantry, a small section Of millimeter-scale graph paper was taped to
the plexiglass surface. This scale allowed the operator to find the precise
location or the "zeroing point" (see above explanation for #12, TBL

POSITION), that corresponded to the proximal and distal ends Of the

64

biomechanical length landmarks for each femur and humerus.

Finally, a fixed black tape line on the surface of the plexiglass platform
was aligned relative to the scanning beam in a procedure identical to that
described above for orienting the box. .Once the scanning light was
superimposed onto this black tape line, the box was secured into place relative
to the platform with a wooden wedge placed between the edge Of the box and
the inside edge Of one Of the wooden Chocks. At this point, the plexiglass
platform and the orientation box essentially functioned as a single unit when
moving in and out Of the scanning gantry. The scanning apparatus was thus
ready to accept a long bone. The next step was to orient each long bone on
top of the plexiglass platform, and the rather complex measurements needed

tO do so are described in detail below.

Femorel Orientation

When one considers the highly precise association between the beam Of
X-rays generated by a CT scanner and the unique nature Of a human long
bone’s orientation vis-a-vis that beam that is necessary to achieve an accurate
cross-sectional image, it becomes patently clear that the complex three-
dimensional morphology Of the adult human femur presents several major
Obstacles. A femur’s proximal (head) and distal (condylar) articular surfaces
do not lie in the same horizontal plane, nor do its most posterior-projecting

contact points with respect to a flat horizontal surface. Positioned on a level

65

reference surface, posterior side down, the shaft of a femur can by no means
be considered to be level. Also, due to its-anterior curvature, the full length of
the diaphysis does not lie within the same coronal plane.

Since one Of the overriding considerations involved in deriving accurate
CT cross-sectional images from long bones is the assurance that the scanning
beam is as perpendicular as possible to the point on the bone being scanned
(in theory, the X-ray beam should intersect the bone surface at 90° to the
scanning location), a number Of external reference points and orientation axes
have to be defined that allow the femur to be consistently positioned as it is
being scanned. Toward this and, three separate but overlapping planes or
axes (sagittal or anteroposterior (A-P), mediolateral (M-L) and rotational) were
determined with the aid Of well-defined, easily reproducible (albeit after
considerable practice) reference points. How the precise orientations were
accomplished will be discussed within the context of describing how these
reference points were determined. A record Of an entire scanning cycle,
beginning with turning on the machine and including the orientation Of the
bones with respect to the three reference axes, was recorded on videotape.

The anteroposterior (A-P) axis

"Levelling" the femur involved manually raising the proximal end Of the
bone while allowing the distal end to lie directly on the plexiglass platform. In
order to determine how high to raise the proximal end, two reference points

(marked by small erasable pencil dots, and placed on the lateral edge Of the

66
bone while completing the whole bone anthropometrY). were established. The

proximally-positioned reference point was determined by measuring, with
sliding calipers, the A-P midpoint, just distal to the lesser trochanter. The
distally-positioned reference point was determined by taking the midpoint, also
in the A-P direction, just proximal to the distal condyles. Because the femoral
surface is concave just proximal to the most proximally projecting aspects Of
the condyles, spreading calipers were used. Once these two points were
established, the actual ”levelling" of the femur took place by placing a small
wedge of modelling clay under the proximal extremity, roughly under the lesser
trochanter. A hand-held, transparent quilting ruler, measuring 60 cm long x 13
cm high, with numerous parallel horizontal lines permanently affixed to its
surface, was placed against the lateral side of the bone, and the two reference
points were aligned by gently pressing the proximal end Of the bone into the
clay until the two orientation points were aligned with respect to one of the
fixed lines on the quilting ruler. Since the entire bone still had to be moved on
the plexiglass platform to its proper zeroing position, the modelling clay was
carefully separated from the platform, yet still left adhering to the underside Of
the bone. In this way, the bone, clay in tow, could be accurately repositioned.
The mediolateral (M-L) axis

The next step involved properly aligning the femur side-tO-side, or in an

M-L plane. As with the A—P orientation, two reference points were needed to

effect proper positioning. The proximal point corresponded to the

67

biomechanical length reference points as described above in the whole bone
anthropometry section. Briefly, the proximal fixed point is located along the
superior surface of the femoral neck, just medial to where the neck joins the
greater trochanter (near the point Of insertion Of the Obturator intemus muscle).
The distal reference point is located at the midpoint Of the intercondylar notch.
In order to understand how these two M-L reference points were
aligned, additional features Of the orientation box and plexiglass platform must
be described. Extending from both sides Of the center Of the box were two
vertical pieces Of wood. Near the tOp Of these extensions were drilled small
holes through which a length of monofilament fishing line was passed. At one
end Of the box (facing the gantry) the line was knotted and secured so that it
would not move. At the opposite and Of the box, a small wedge Of wood, also
with a small hole in it, was placed flush with the top Of the extension. The line
was sent through both Of these sections and knotted, snugly but not so tightly
that the wooden wedge could not be moved. Once connected, the line ran
down the center Of the orientation box, some 25 centimeters above the
plexiglass platform. Directly beneath this monofilament, a thin line Of black
tape was secured to the platform, and positioned along its center. By standing
above the box, one could superimpose the monofilament line onto the black
tape by gently pulling the movable wooden section side-tO-side. The femur
was then slid into place directly beneath the monofilament line, and the two

reference points were aligned with respect to the monofilament and black lines.

68
T r tational axis

This was by far the easiest alignment to accomplish. The posterior
edges Of the condyles were laid directly on the plexiglass platform (the need
for this specific placement constituted the primary reason that any femora with
evidence Of erosion on this surface were eliminated from the CT protocol—see
section on sample exclusion criteria above), and the proximal end was left to
rest on the base of modelling clay.

flrLaLDQSJILOMIQ

Having achieved the three axes as described above, the final step was
to move the bone to its zeroing point so that scanning could begin. The bone
was "zeroed" by sliding it into final position by aligning the midpoint Of the most
projecting edges Of the condyles with the small section Of graph paper affixed
to the plexiglass platform. Once this was done, the bone was ready for
scanning. The bone number and Side were relayed to the CT scanner
operator from a sheet previously prepared with the five cross-section locations.
The operator entered these positions into the console; after the first three
scans were completed (20%, 35% and 50%), the phantom was slid to the
proximal end of the bone so that the 65% and 80% positions could be
scanned. Again, this step was necessary because the distance between the
20% and 80% femoral scanning locations was greater than the overall length

Of the phantom.

69

W

Most Of the Obstacles outlined above regarding proper femoral
orientation do not hold for the humerus. Although it is also necessary to
establish three separate axes in order to properly orient the humerus on the
plexiglass platform, the sagittal (A-P) axis, in particular, is easier to orient, in
that the humeral diaphysis is not curved like that Of the femur.

W

NO true "levelling" of the humerus was necessary. The bone was simply
laid on the plexiglass platform, posterior side down. In this position, the
diaphysis presents itself as nearly parallel to the platform, and as such
approaches perpendicularity with respect to the CT scanning beam. Also,
because no modelling clay was needed, a final repositioning like that described
above for the femur was not necessary.

The mediplateral (M-L) axis

Two reference points were needed to properly orient the humerus along
the M-L axis. Both points were represented by M-L midpoints along the shaft,
one located proximally at the level of the surgical neck, the other located
distally at a point just proximal to the most superior aspect of the olecranon
fossa. Once established, these two reference points were aligned relative to
the monofilament line positioned above the orientation box. The line was then

superimposed against the black tape and secured to the plexiglass platform.

70

The rotational axis
The humerus, properly positioned, rested on the head and greater
tubercle proximally, and on the anterior surface of the capitulum and trochlea

distally. As such, the bone was not free to rotate out Of proper orientation.

Processing the CT-Genereted Cross-Segional Image Data
Approximately 2.5 megabytes Of disk memory were needed to store the

raw data that comprised each cross-sectional image. Given that the research
protocol designed for the present study called for all femora and humeri to be
scanned at five separate locations, a total Ofapproximately 12.5 megabytes
(2.5 megabytes/scan x 5 scans) Of disk memory were needed to store the raw
data for a single long bone. In order to prepare each image for analysis, the
raw data images needed to be reconstructed. The image reconstruction
process reduced the amount of disk space needed per image by about four-
fifths, in that a single reconstructed image occupied approximately 0.5
megabytes Of memory. Thus, each long bone’s reconstructed images
potentially totalled 2.5 megabytes of memory (0.5 megabytes Of memoryfimage
x 5 images).

Numerous steps are necessary to process the raw CT data so that the
areal and biomechanical properties can be calculated. All Of the raw CT data
was collected using a small field Of view and a high kilovoltage (KV) protocol.

Acquisition time for the CT data was approximately 40 seconds for each axial

71

slice. The total acquisition interval included time for the x-ray tube to cool prior
to further scanning (about 20 seconds per axial slice); the remaining time was
allocated for data transfer from the analog to the digital converters, then to a
virtual disk and, finally, to an actual raw data file. Each of these files contains
the information needed to accomplish the axial reconstructions with a host Of
specified parameters. The final version Of the axial images (about 0.5
megabytes each) were reconstructed from the raw data set by application Of a
"sinc500" filter (Cody and Flynn, 1989). The reconstruction process took
approximately 60 seconds per axial image, or typically five minutes per study,
given that each long bone was scanned at five separate locations. All
reconstructed cross-sectional images were archived onto 3M DC (data
cartridge) tapes, each with 60 megabytes of storage capacity.

The next step was to transfer the cross-sectional image data from the
PDP—11 based CT scanner to a Sun Microcomputer via an ethemet link. This
transfer took approximately one minute per axial slice. The large size Of each
raw data file (approximately 2.5 megabytes/slice) and the relatively small
amount Of disk space available on the CT scanner resulted in a relatively slow
processing rate for each specimen.

TO analyze the cross-sections, each image was initially converted into a
data format compatible with the software created for the analysis routine. The
original cross-sectional image format was inherent to the Technicare HPS 1440

scanner, and contained a descriptive header Of 2048 bytes and a 512 pixel by

72
512 pixel by 2 words per pixel raw image (512x512x2 bytes). The conversion

software transformed each image into a standard format used by the Physics
and Engineering Division at Henry Ford Hospital (HFH), known as HFH format.
A 128 byte descriptive image header was created for each image using the
actual information found in the HPS image. The data were then appended to
this header to create a HFH format image.

The next step was to convert the cross-sectional image’s pixel units (in
Houndsfield Units) to a measure that has more physical relevance, such as
physical density (mg/cm3) or concentration of bone mineral, calcium
hydroxyapatite. This conversion was performed using the known densities
within the calibration phantom (Image Analysis Inc., Columbia, Kentucky), with
respect to two fine tin guide wires. These wires were located relative to each
Of the cross-sectional images, the distance between them was calculated, and
the value was compared to the actual value in order to determine a scaling
constant for all phantom-related measures. Once each cross-sectional image’s
actual distance relative to the guide wires was known, the position Of each
region in space was found using a combination Of both known distances, the
scaling constant, and the guide wire positions.

Once the calibration regions were determined for each axial image for
an individual specimen, a linear regression was applied to each image. This
regression produced a direct relationship between the measured Houndsfield

Units and the known physical density of the calibration phantom’s regions.

73

The regression was then applied to each pixel value on each image. In this
way, the long bone diaphyses did not have to be sectioned, and the cross-
sectional outlines did not have to be traced and manually digitized in order to
calculate the biomechanical properties. By using a "thresholded” raw CT
number to determine the presence or absence Of cortical bone (see below),
potential errors due to image distortion in relation to discriminating between the
periosteal surfaces is eliminated. Finally, each image was reduced from the
512 pixel x 512 pixel image size to a 128 pixel x 128 pixel image size (or, for
several larger cross-sectional images, a 180 pixel x 180 pixel image size)
containing the region Of interest (i.e., bone tissue).

Analysis Of the cross-sectional images began by isolating each
individual region of interest, which included the cortical area (CA), the
medullary area (MA), and the total subperiosteal area (TA) within the bone. All
"salt and pepper noise," or drop-out pixel values, were screened by applying a
3x3 median filter. The purpose Of the filter was to eliminate any individual
spurious values that might alter the next, or segmentation step, which was
performed in order tO threshold the image at a value Of 60% Of the maximum
pixel value. This produced a binary image with a background value of zero,
and the regions of interest (i.e., bone tissue) with a value of one.

The binary image was then dissected into the two components of
interest, cortical and medullary regions. This separation utilized the unique

properties Of each component. For example, a line bisecting any portion Of the

74

medullary cavity of the cross-sectional image of a long bone will pass from the
periosteal surface, through the cortex, to the endosteal surface of the
medullary cavity, through the medullary cavity, to the endosteal surface of the
other side Of the medullary cavity, through the cortex, and to the periosteal
surface Of the other side. Hence, the cortical component is characterized by a
continuous "shell" with four transitions along a line bisecting the bone. It
follows then, that a measurement with four transitions would imply the
presence Of both medullary and cortical bone. The total area between the first
and second, and the third and fourth transitions, is the total cortical bone area
(CA) component; the'area between the second and third transitions comprises
the total medullary area (MA) component. The sum Of the two components
represents the total subperiosteal area (TA). Each Of the two components was
given a unique region defining number (i.e., 0=background; 1=cortical;
2=medullary). The assignment of these numbers completed the process Of
segmentation and the classification Of the region(s) Of interest (ROI).

Each moment Of inertia calculation was performed on both the binary
and binary-masked gray scale images. This delineated the cross-sectional
area components and the centroids necessary for determining the

biomechanical properties, as described below.

75
Biomechanical Properties
Cortical Area (CA) describes the total area occupied by cortical bone in

a given one millimeter cross-section. CA is calculated by summing the total
number of pixels whose CT numbers are above the given threshold value, then
multiplying this number by the size Of each pixel within the matrix
superimposed on the cross-section.

Mada/lazy Araa (MA) describes the total area occupied by the medullary
cavity in a given one millimeter cross-section. TA is calculated by summing
the total number Of pixels whose CT numbers are below the threshold value,
then multiplying this number by the size Of each pixel within the matrix
superimposed on the cross-section.

Total Area (IA) represents the total subperiosteal area in a given one
millimeter cross-section. It is determined by summing the values for CA and
MA, as described above.

The ability of a given long bone cross-section to resist axial or
compressive loads is directly proportional to its cross-sectional area. It is too
simplistic, however, to assume that the cross-section in question is primarily
subject to pure compressive loadings. Human long bones are irregular, in that
they depart from a perfectly smooth cylindrical shape, due to the muscular
forces that are placed upon them. Hence, bending and torsional forces must
be accounted for, and to do so for biomechanical analyses, it is necessary to

determine the centroid for each cross-sectional image.

76

The centroid, or "balancing point" for a cross-section, was calculated

using the relative density Of each pixel according to the following formulae.

C, = 2(CT)(d,,)/zcr Cy = z(CT)(d,)/zCT

In the above equations, the symbols are represented as follows:
- C), is the position Of the centroid relative to the X axis,
- C, is the position Of the centroid relative to the Y axis,
0 CT is the CT number of each pixel, a measure Of relative
density,
- d, is the distance Of the centroid from the X axis,
- d, is the distance Of the centroid from the Y axis.
Additionally, it is necessary to understand the stress generated by the
bending forces along a given cross-sectional axis. The relationship is

described by the following equation:

a = (M)(Y)/l.
In this equation, the symbols are represented as follows:

- 0 denotes stress,

- M represents the bending moment of inertia along the given

axis,

. Y represents the maximum fiber length

- I, represents the area moment Of inertia.

It follows that the degree Of stress is inversely proportional to l,, the

area moment Of inertia, which is in turn a good estimate Of the strength Of the

cross-section with respect to the forces applied relative to the axis in question.

We can thus derive a formula to calculate area moments Of inertia for any

77

particular axis as:

I. = 2mm?)
In this equation, the symbols are represented as follows:
- Ill represents the area moment Of inertia,
- A represents the area Of each pixel,
- d represents the perpendicular distance from each pixel’s center
to the axis Of the centroid.

The equations for determining the area moments about the X

(mediolateral) and Y (anteroposterior) axes can now be derived as follows:

I, = Z(A)(d,‘°')
In this equation:
- I, = the area moment Of inertia about the X (mediolateral) axis,
- A represents the area Of each pixel,
- d, represents the perpendicular distance from the center Of each
pixel to the X axis Of the centroid.
Similarly, for the Y axis:
I, = Z(A)(dy2)
In this equation:
- IV = the area moment of inertia about the Y (anteroposterior)
axis,
0 A represents the area Of each pixel,
- d, represents the perpendicular distance from the center Of each
pixel to the Y axis of the centroid.

Although it is theoretically and technically possible to determine the

moments Of inertia for an infinite number Of horizontal and longitudinal axes, it

78

is more useful, and conventional, to determine what are known as the

maximum and minimum moments Of inertia of a given cross-section with the

following formulae:

1,“, = (I, + i,)/2 + (l, - 1,)12 + i;

IIt'lin = (Ix + ly)l2 - (Ix - Iy)2 + lxyz

In these equations:

= the maximum area moment Of inertia,

= the minimum area moment of inertia,

the area moment of inertia about the X axis,

the area moment of inertia about the Y axis,

= a measure Of regularity Of distribution between quadrants.

max
mi

X

II II:

- I
- l
- l
s |y
s Ixy
lxy itself can be calculated from the following equation:
1., = i/(A)(d)(d,)]
In this equation:
- A = the area Of each pixel,
- d, = the perpendicular distance from pixel center tO the X
axis,
0 d, = the perpendicular distance from pixel center to the Y
axis.
In addition to the equation above that describes bending forces, it is

necessary to derive an equation that will provide a good estimate Of shear, or

torsional loading. The formula representing the maximum shear stress within a

given cross-section is:

S=£Dm
J

79

Here:
- 8 denotes maximum shear stress,
- T denotes the applied torque,
- Y denotes the maximum fiber length (i.e., the distance between
the centroid and the most distal point on the cross-section),
- J denotes the polar moment Of inertia (torsional strength).
Similar to the case for bending stress (a), shear stress is inversely

proportional to a moment Of inertia-4n this case to J. J is therefore a good

estimator Of torsional loading, and can be derived from the following equation:

J=L+g
Here:
. J denotes the polar moment Of inertia,
- I, denotes the area moment Of inertia along the X axis,
0 I, denotes the area moment Of inertia along the Y axis.

In addition to the calculation Of CA, MA, TA, lm, lmm, I,, l,, and J, three
"shape" ratios, PCA, l,,,,,/l,,,,,, and IJI, (Ruff and Larsen, 1990) were calculated.
One Of these expresses the amount of cortical bone in a cross-section relative
to the total subperiosteal area, and is defined as:

PCA = (CA/TA) x 100

Where:

- PCA = percent cortical area,

0 CA = cortical area,
- TA = total subperiosteal area.

FOr the two other ratios, we have:

80
[ma/Imin and All,
Where:

- IWII",in = the ratio of the maximum to the minimum bending
strength;

- IJI, = the ratio Of the area moments Of inertia about the x and y
axes.
Normalization Of Data

Prior to a statistical analysis of the biomechanical variables, it is
necessary to account for the effect Of overall body size on individual long bone
dimensions. This process Of compensating, or "standardizing," for size
differences is known as normalization. It is not a simple task to accommodate
for the potential influence Of body size on bone size in studies Of
bioarchaeological remains (Albrecht et al., 1993; Bridges, 1985; Ruff and
Larsen, 1990). Body weight has been used for this purpose, but
archaeological samples Obviously preclude this approach. Long bone length
has been used, and while incorporating this relatively simple measure into a
statistical protocol is not difficult, it is by no means Clear that length per se is a
good "adjuster" for overall bone size.

Several studies have demonstrated that in biomechanical analyses the
best way to accommodate for size differences is to divide the dimension in
question by the same power as the units Of the moments of inertia, or mm‘
(Bridges, 1985; Lovejoy at al., 1976; Ruff, 1984; Ruff et al., 1984).

Additionally, Bridges (1985) has tested various methods Of normalization,

81

including dividing the biomechanical properties by maximum fiber length, femur
length, and (femur length)“. The results indicate that while changing the
denominator does alter the differences between the Archaic and Mississippian
groups for some variables, the overall pattern Of change remains the same
(Bndges,1985)

A relatively new method Of normalization for the femur has been
developed (Ruff at al., 1993) whereby cross-sectional areas and second
moments of area are divided by (length)3 and (Iength)5'33, respectively.
Normalization in the present study, however, was accomplished by dividing
cross sectional areas by (biomechanical length)”, and also by dividing second
moments of area by (biomechanical length)‘- Cross-sectional area and
moments Of area variables are thus reported in min2 and mm‘ units,
respectively (see key to Table C1). This particular method Of normalization is
consistent with that Of Bridges (1985) and Ruff and Larsen (1990), two
important comparative Native American bioarchaeological samples against

which the results of the current research are evaluated.

tatistical A roach
As discussed previously, seven separate skeletal samples were
examined in this study, four from Michigan and three from New York. Of the
four Michigan samples, two were derived from the prehistoric period (Riviera

aux Vase and Juntunen) and two from the historic period (Lasanen and

82
Fletcher). Of the three New York samples, two were derived from the

prehistoric period (Frontenac Island and Harscher) and one (Tram) from the
historic period.

Analyses were carried out independently for the whole bone
anthropometric and biomechanical data sets. Nineteen femoral and 14
humeral measurements were tested. For the biomechanical data, 11
measurements at each Of three separate CT scanning locations (i.e., 20%,
50% and 80% Of biomechanical length) were tested. These three sites
correspond to scan #1, #3 and #5, respectively, as described above.

All statistical work was performed with SPSS for Windows (NorUSIS,
1990, 1993). Two-tailed t-tests (non-directional) for independent (pooled)
samples were utilized to test whether there are any long bone size, shape or
strength differences between the pooled temporal groups, controlling for sex
and age. Test variables consisted of the femoral and humeral anthropometric
and biomechanical measurements discussed above. For the two data sets
combined, 30 separate femoral (19 anthropometric and 11 biomechanical) and
25 separate humeral (14 anthropometric and 11 biomechanical) variables were
analyzed.

For the independent samples routine, grouping variables consisted Of
the recoded or combined archaeological sites. For all anthropometric
analyses, the data was filtered for bone side, sex, age and site. For all

biomechanical analyses, the data were filtered for bone side, sex, age, site and

83

scan position. For statistical purposes, the four Michigan samples were
collapsed into two groups based upon prehistoric (i.e., Riviera aux Vase and
Juntunen sites) and historic (i.e., Fletcher and Lasanen sites) contexts, and the
New York samples were grouped such that the Archaic period Frontenac
Island site was compared to the two protohistoric (i.e., Tram and Harscher)“
Seneca sites.

For each significance test, a variable mean, standard deviation,
standard error Of the mean, and a mean difference (between the two sample
means) was calculated. T values and their corresponding two-tailed
probabilities, for both equal (i.e., pooled) and unequal (i.e., separate) variance
estimates, were also calculated (Norusis, 1990, 1993). Results for all tests
were considered statistically significant for p<0.05, on the basis that the
findings Of this research are compared to studies that all report their results

significant (although not exclusively so) at a p<0.05 level.

Chapter 5

RESULTS

The following is a summary Of the results for the femoral and humeral
anthropometric (see Appendix B, Tables B1-B8) and biomechanical analyses
(see Appendix C, Tables C1-C12). Each Of these two data sets is addressed
separately for males and females for all femora and humeri. For Michigan,
results are discussed for prehistoric vs. historic period comparisons. For the
New York samples, results compare the hUnter-gatherer vs. agricultural
groups. Overall trends in the measurements (i.e., increases or decreases) are
examined (Table 7 and Table 8), and statistically significant differences are
indicated. Due to insufficient sample sizes, no biomechanical results are

available for female femora or for male humeri within the New York sample.

Anthropometrics

Michigan prehistoric vs. historic perigz male femora

There is a consistent pattern of decreased femoral dimensions when

84

85

historic and prehistoric period Michigan males are compared (T able B1;
Figures 01-06). This pattern is not localized, but is seen along the diaphysis
in nearly all measures Of length, circumference, and breadth. The only ‘
exceptions are in the midshaft A-P diameter, where there is no change, and in
the midshaft M-L and subtrochanteric A-P diameters (Figure 03), where the
Historic period males have larger diaphyseal dimensions. For all four length
measurements, the subtrochanteric M-L diameter, and the proximal epiphyseal

breadth, there is a statistically significant decrease in the historic period.

Michigan prehistoric vs. hi§t_oric period: female femora
Similar to the case for males, the Michigan female sample also shows a

statistically significant decrease for all measures of femoral length in the
historic period (T able B2; Figures 01 and 02). However, the remaining pattern
Of femoral change within the female sample is quite different compared with
that seen for the males. Specifically, the general pattern of decreased size
and robusticity experienced by historic period males is not repeated. In fact,
the general pattern along the femoral diaphysis, other than for midshaft A-P
diameter and circumference, subtrochanteric M-L diameter (Figure 04), lateral
condylar length, and proximal epiphyseal breadth (Figure 06), is that of
increased size (significantly so for neck girth-Figure D5) in the historic vs.

prehistoric period female sample.

86

New York hunter;gatherer vs. agricultural groups: male femora
The general pattern of Change seen when comparing the New York

hunter-gatherer and agricultural groups is one Of slightly increased size in the
latter (T able B3), although the changes are not as large or as consistent
across the femur compared with the Michigan male sample. For example,
while all of the femoral head measurements are smaller in the agricultural
group, all Of the neck measurements are larger. However, none Of these
differences reach statistical significance. The two measurements that do
significantly differ between the two groups also do so in opposite directions.
Specifically, the subtrochanteric A-P and M-L diameters show a significant
decrease (Figure 08) and increase (Figure 09), respectively, in the agricultural

compared with the hunting-gathering group.

New Yprk hunter;gatherer vs. agricultural groups: female fempra
The pattern borne out in the New York male femora sample is largely

repeated for females (T able B4; Figures 07-09). All femoral length, neck, and
- condylar length measurements are larger in the agricultural group, while the
femoral head measurements show either no change, as is the case for
maximum diameter, or a decrease in size. While there is a statistically
significant increase in the midshaft A-P diameter in the agricultural group
(Figure 07), there is a statistically significant decrease in the subtrochanteric
A-P diameter (Figure 08). This latter difference parallels that seen in the male

sample in terms of statistical significance and direction Of change.

87

Michigan prehistoric vs. historic period: male humeri
There is a generally consistent pattern of decreased humeral

dimensions in the historic vs. prehistoric Michigan male humeri sample (T able
B5; Figures 010-014). This holds for all of the diaphyseal length, head, and
breadth measurements, however, this pattern of change is reversed for all Of
the midshaft dimensions (Figure 013), although not significantly SO. All lengths
(maximum, biomechanical, lesser tubercle-trochlear, greater tubercle-medial
epicondylar, head-medial epicondylar) and head girth are significantly smaller
in the historic period sample.

There is also a striking parallel between the changes seen in the femora
and humeri within the Michigan male sample. Not only are both bones shorter
in the historic period, but the majority of the remaining measures Of size show

a pattern Of decreased robusticity in the historic period.

Michigan prehistorie vs. historip period: female hpmeri
The pattern Of change in the female humeri sample is similar to that for

the males, with all length (Figure 010-012) and head (Figure 014)
measurements decreasing in the historic period (T able B6). The major
difference is that, unlike for the males, the magnitude Of the differences
between prehistoric and historic period females is not as large, and none Of the

length differences are statistically significant.

88

New York hunter:gatherer vs. agricultural groups: inele mimeri
The pattern of change Observed in the New York male humeri sample is

 

mixed (T able B7). While all Of the length measurements are larger in the
agricultural group, all of the midshaft diaphyseal, head (save girth) and breadth
measurements are smaller in the agricultural group. None of the differences in
the humeri measurements between these two groups reach statistical
significance. It is interesting to note that compared to the Michigan male
prehistoric group, the New York male hunter-gatherers have much (on an
absolute basis) shorter humeri. Alternatively, when the same comparison is
made between the New York agriculturalists and individuals from the Michigan
historic period, the reverse is true, i.e., the post-contact Michigan males have

shorter arms.

New York huntergatherer vs. agricufiural groups: female humeri

In the New York female humeri sample, there is no clear pattern to the
direction Of the differences between the hunter-gatherer and agricultural groups
(T able B8). For example, two Of the five length measurements are smaller in
the agricultural sample, while three are larger. There is little change in the
midshaft measurements, and evidence for a slight increase in the anterior and
posterior trochlear breadth measurements. Humeral head measurements are
all smaller in the agricultural sample. However, like all Of the Changes between

these two samples, none are statistically significant.

89

   

BorrLctha

Michigan prehistoric ve. hietorigperioi male femora
As can be seen from the top Of Table C1, at the distal location (i.e., at

scan site #1; Figures D15-D19), nearly all of the geometric and biomechanical
properties in the Michigan male sample are larger in the historic group,
indicating a pattern Of increased robusticity and strength compared with the
prehistoric sample. Two measurements, MA and TA, are significantly larger in
the historic period, while PCA (Figure 019) is significantly smaller.

At the midshaft position (i.e., scan site #3-top Of Table C2), the cross-
sectional areas Show the same general pattern Of change as they do for the
distal region, that is, a decline in CA and increases in MA (Figure 020) and TA
dimensions in the historic period group. This argues for a larger cross-section
with relatively less cortical bone as a percentage Of total area. Additionally, as
' is the-case for scan #1, MA and PCA are significantly smaller in the historic
period. Unlike the distal location, at midshaft all but one (i.e., lmin) Of the
strength measures, and all but one (i.e., IJIY) of the strength indices, decrease
in the historic period.

Proximally (scan site #5-top Of Table C3), the general pattern Of change
(Figures 021-025) is more similar to that seen at midshaft than it is to that at
the distal location in that historic period males manifest relatively weaker

femoral diaphyses compared with the prehistoric sample. Only MA and the

90

two strength indices, l,,,,,/l,,,,,, and lle, are larger in the historic period Michigan
male group at this subtrochanteric femoral location. None Of the geometric or

biomechanical measures are significantly different between the two groups.

Michigan prehistoric vs. historic period: femele femora
Paralleling the general increase seen for the male sample, the female

group (bottom Of Table C1) also have larger and stronger femora at the distal
location in the historic period (Figures 015-019). The magnitude of the
differences, however, is generally much larger in the female sample, and,
unlike their male counterparts, many more Of the geometric and strength
measures reach statistical significance in the female sample. Interestingly, the
only three measurements for the male sample that decreased in the historic
period, namely CA, PCA (Figure 019) and l,/I, (a measure Of bending strength
about the x and y axes) are also the only three that decrease in the historic
period female sample.

At midshaft, the male and female (bottom Of Table C2) results depart
considerably. For the females, all but two of the femoral measurements
increase in the historic period, with only I, (bending strength about the y axis)
significantly larger, whereas the general pattern for the males is one of
decreased midshaft diaphyseal strength in the historic period.

As the bottom Of Table C3 indicates, the most striking differences for the

prehistoric and historic period femoral data are seen at the subtrochanteric

91

location. First, for females, two Of the three area measurements, CA (Figure
021) and TA (Figure 022), and all of the strength-related geometric properties
(Figures 023-025), are significantly larger/stronger in the historic period. This
general pattern Of increase is also Observed at the distal and midshaft
locations, but the male-female comparison, by scan site, is quite different. For
example, at scan #1, both males and females are larger/stronger in the historic
period, with females more so in terms Of more measures reaching statistical
significance. At midshaft, historic period female femora are larger and stronger
compared with prehistoric period females, but the males demonstrate a trend

in terms of weaker diaphyses during the historic period.

New York hunter;gatherer vs. agricultural groups: male femora
The New York hunting-gatherer and agricultural groups do not Show the

same level Of significant differences compared to the Michigan samples. For
example, at the distal location (scan #1 -Table C4), while there is a general
pattern Of decline in proximal femoral strength for the male agricultural group,
none Of the biomechanical measures are significantly different. The same
holds at midshaft (T able CS), with two fewer measures (i.e., In, and l,)
decreasing in the agricultural group. At the subtrochanteric location (T able
C6), only lmflmm, an index Of bending strength, is significantly different, in a

negative (i.e., weaker) direction, in the agricultural vs. hunting-gatherer group.

92

Michigan prehistoric vs. historic period: mele hemeri
As the top Of Table C7 demonstrates, for the Michigan male humeri

 

sample at the distal location (Figures 026-028), there exists an overall pattern
Of positive change (i.e., larger cross-sectional areas and stronger diaphyses)
during the historic period. TA (Figure 026), lmin, and I, (Figure 027) are
significantly larger, indicating an increase in total cross-sectional size and
increased resistance tO bending forces about the M-L axis.

The same overall pattern Of change is observable at midshaft (top Of
Table C8), with TA (Figure 029), I, (M-L bending strength-Figure 030), and l,/ly
(the area moment of inertia index) showing significant increases in the historic
period. At the proximal humeri scan position (Table C9), once again the
historic period male sample is larger and stronger above the elbow, however,

none Of the differences reach statistical significance.

Michigan prehistoric vs. historic period: female humeri
The results for the Michigan female humeri sample are presented at the

bottom Of Table C7 for the distal scan location (Figures 026-028). It can be
seen that the overall pattern Of change mirrors that for the males, with only
PCA and the lJl, strength index showing a decrease in the historic period. In
the female sample, TA is also Significantly larger in the historic period, as is J
(Figure 028), a measure Of torsional or twisting strength.

Compared to the distal humeral site, there are more significant

differences at midshaft (bottom Of Table C8). At the latter location, MA and TA

93
(Figure 029) are larger, and law, I, (Figure 030), J, and the M, index are all

significantly stronger in the historic period. Proximally (bottom Of Table 09)
once again the historic period female humeri sample is larger and stronger
compared to the prehistoric period sample, yet only with the lmazflmin strength

index significantly so.

New York hunter;gatherer vs. agricultural groups: female humeri
At all three cross-sectional locations along the humeral diaphysis for the

New York females, the agriculturalists have generally smaller cortices and
weaker measures of humeral strength (T ables C10-C12). The notable
exception to this pattern can be seen with respect to medullary areas at the

‘ distal (Figure 031) and midshaft locations (Tables C10 and C11, respectively),
where there are significant increases in the agricultural group (and significant
decreases in PCA). None of the other biomechanical properties (save for CA

at midshaft), although smaller in the agricultural group, are significantly so.

Testing of Hypotheses
To reiterate, the primary research goal Of the present study was to

examine postcranial skeletal structural adaptations in bioarchaeological
populations from Michigan and western New York. It accomplished this by
utilizing two separate case studies. First, it examined differences in long bone
structure that may have resulted from concomitant changes in physical activity

between the prehistoric and historic periods in Michigan. Second, it examined

94

the long bone structural changes accompanying the transition from a hunting-
gathering to an agricultural subsistence economy in western New York State.

In order to test the hypotheses as outlined in Chapter 3, two separate
yet complimentary data sets, both of which measure dimensions Of the femur
and humerus, but which are predicated on quite different methodological
approaches, were used. One data set was derived from a whole bone
anthropometric analysis of external bone dimensions, the other was derived
from a biomechanical analysis Of computerized tomographic (CT) scan-
generated diaphyseal cross-sectional Size, shape, and strength characteristics.
The following hypotheses can now be evaluated in light Of the results as

presented above.

Hypothesis 1

There will be no difference between the Michigan prehistoric and historic

period samples with respect to anthropometric measures of long bone

Size.

The anthropometric data for the Michigan sample demonstrate that male
and female femora are significantly shorter, and have generally less robust
diaphyses, in the historic period. Hence, for the femora of both sexes, the null
hypothesis is rejected. Male humeri are also significantly shorter in the historic
period. The remaining measurements present a mixed pattern Of change

demonstrated by the diaphyseal and articular breadth measurements. For the

male humeri sample, then, the hypothesis is also rejected. Although the

95

female humeri are generally smaller in the historic period, the differences do
not reach statistical significance, save for a single diaphyseal measurement,
which is significantly larger in the historic period sample. For the Michigan

female humeri sample, then, the null hypothesis is accepted.

Hypothesis 2

There will be no difference between the Michigan prehistoric and historic
period samples with respect to long bone cross-sectional biomechanical
properties.

The biomechanical data for the Michigan samples indicate that male
femora are generally larger and stronger at the distal site, yet generally smaller
and weaker at the midshaft and subtrochanteric locations in the historic period.
Very few Of the strength measures are statistically different between the two
periods, hence the null hypothesis is provisionally accepted.

At all three locations along the diaphysis, the female femora are larger
in the historic period, and the magnitude Of this increase is generally greater
compared to the male samples. At the distal and subtrochanteric positions,
nearly all Of the measures of size and strength are significantly larger in the
historic period. Hence, the null hypothesis is rejected.

In terms Of the humeri, both males and females Show a general pattern
of larger size and strength in the historic period. This pattern is seen across
the diaphysis (i.e., at scan locations 1, 3 and 5). For all comparisons at all

scan sites, the historic period females show the most significant differences (at

96

midshaft) when compared to their prehistoric counterparts. For both males and
females, the null hypothesis is rejected for the distal and midshaft locations.
However, since none Of the area or strength differences are statistically

Significant proximally, the null hypothesis is accepted for that location.

wile—sis;

There will be no difference between the New York hunter-gatherer and

agricultural samples with respect to anthropometric measures of long

bone size.

For the New York samples, the morphometric data indicate a slight bias
in the direction of longer femora in the agricultural group for both sexes,
although none Of the measures is statistically significant. Also for both sexes,
the diaphyseal dimensions Show a mixed pattern Of change, with some
measures larger in the agriculturalists, while others are smaller. For the New
York male and female femoral data, the null hypothesis is accepted.
Additionally, the expectation (based on previous work-see Chapter 3) that the
agriculturalists would have larger femora compared to the hunter-gatherer
group was not met with this New York data.

. Also in the New York samples, male humeri are longer, but less robust,
in the agricultural group; there is no discernable trend for females between the
two populations. None Of the New York humeral measurements, for either sex,

are statistically significant when the hunter-gatherer and agricultural groups are

compared, hence the null hypothesis is accepted.

97

Hypothesis 4

There will be no difference between the New York hunter-gatherer and
agricultural samples with respect to long bone cross-sectional
biomechanical properties.

The biomechanical data for the male New York samples indicate that for
measures Of size and strength at all three locations along the femoral
diaphysis, agriculturalists are generally smaller and weaker. However, at the
distal and midshaft locations, none of the differences are significant; at the
subtrochanteric location, only l,,,,,,/lmin is significantly different (smaller) in the
agricultural group. Hence, for the New York male sample, the null hypothesis
is accepted.

For the female humeri, the agriculturalists are also generally smaller and
weaker across the diaphysis. At the distal location (scan #1), PCA is
significantly less, and at midshaft CA and PCA are significantly less, in the
agriculturalists. However, at both Of these locations, MA is larger in the
agricultural sample. For these two locations, the null hypothesis is rejected.
For the proximal humerus, the general pattern Of smaller and weaker
diaphyses continues, but none Of the comparisons are significantly different.
Hence, at this location, the null hypothesis is accepted. These results suggest

that the female agricultural sample changed more at the elbow and in the

midshaft region of the arm compared to the shoulder.

98

Table 7
Summary of Changes in Femoral Diaphyseal Dimensions

and Biomechanical Cross-Sectional Measurements

‘ New York2
female male female
VARIABLE

maximum

A-P
midshaft diameter

M-L
midshaft diameter

circumference

A-P
subtroch diameter

M-L
subtroch diameter

CT3
CA
MA
TA

+

J + +

 

‘ Michigan comparisons are for the prehistoric vs. historic period sample;

2 New York comparisons are for the hunter-gatherer vs. agricultural sample;

3 CT scan locations ("1" = distal; "3" = midshaft; "5" = proximal). A "+" and a "-"
indicate a significant increase or decrease, respectively, in the historic period or in the
agricultural sample, as appropriate. A blank box denotes no statistically significant
change in either direction. Shaped area signifies no data is available.

99
Table 8

Summary of Changes in Humeral Diaphyseal Dimensions
and Biomechanical Cross-Sectional Measurements

‘ New York2

female male female

VARIABLE
maximum

maximum
midshaft diameter

minimum
midshaft diameter

circumference
CT“

J

 

See bottom of Table 7 for key.

Chapter 6

DISCUSSION

The results of this research clearly demonstrate that there are numerous
significant differences in the external dimensions Of the femora and humeri
derived from the prehistoric and historic period samples in Michigan. Specifically,
for both males and females, all measures of long bone length, some measures
Of diaphyseal size, and most measures along the proximal and distal articular
surfaces are shorter/smaller in the historic period. The exception tO this is the
pattern Of change seen in the female femoral data, where the head and neck
measurements are larger, but not significantly so, in the historic period.

The biomechanical results reveal a different pattern of change for both
male and female femora. Whereas the anthropometric data reveal a fairly
consistent pattern Of decreased dimensions for the femora and humeri within both
sexes between the prehistoric and historic period samples, the biomechanical
properties for the male femora indicate greater size and strength at the knee, but
general decreases in both' moving along the diaphysis toward the hip.

Alternatively, female femora from the historic period are larger and stronger along

100

101

the entire length of the diaphysis, and especially so near the hip (i.e.,
subtrochanteric).

The direction Of change for the humeral biomechanical sample is even
more consistent than that for the femur in that, with few exceptions, all Of the
measures Of area and strength are larger/stronger for both males and females in
the historic period. This pattern holds across the entire diaphysis and, in terms
Of measures of statistical significance, is most striking distally (i.e., toward the
elbow) and at midshaft.

Quite in contrast to‘the differences documented for the Michigan skeletal
samples, the results for New York demonstrate that there are few size differences,
for either bone or for either sex, when the hunter-gatherer and agricultural groups
are compared. Even within the few significant differences that are evident, there
is little consistency with respect to the direction Of change. In terms of long bone
cross-sectional area and strength, there is also little significant change between
the two groups. In fact, for male femora and female humeri (there are nO female
femoral and male humeral samples), none Of the measures Of diaphyseal strength
per 89 are significantly different.

In studying archaeologically-derived human remains, it is difficult to
establish with certainty parallels between the biomechanical forces acting upon
bone and specific patterns Of physical activity. However, with the aid of the
archaeological record and historical documentation, and with analogy to

comparable studies from other geographic regions, a broad behavioral framework

102

for interpreting the results Of this study can be constructed. It is within this
context that one can most instructively elucidate the patterns Of physical activity,
and potentially the sex-specific behavioral correlates associated with them, that
"link" the biological (i.e., skeletal) and cultural (i.e., behavioral) data examined in
this research.

TO begin, as a general guide for interpreting the results Of the Michigan
case study where prehistoric and historic period skeletal samples are compared,
one can lOOk to the bioarchaeological studies being conducted for the Georgia
coast. As indicated in Chapter 3, in this region research has documented human
long bOne skeletal structural adaptations in prehistoric preagricultural, prehistoric
agricultural, and historic period agricultural populations (Fresia et al., 1990;
Larsen, 1982, 1984, 1987, 1990, 1995, 1997; Larsen and Ruff, 1991; Ruff et al.,
1984). For the present discussion, it is the relationship between the prehistoric
and historic period samples that is Of most interest for comparison with the
Michigan data.

Briefly, the Georgia coast research has shown that femoral structural
strength declines between the precontact preagricultural and precontact
agricultural periods for both males and females. This trend is then reversed for
the contact period agriculturalists, where femoral strength increases, and
especially SO for females (Ruff and Larsen, 1990). Increased long distance travel
in the post-contact (mission) period is one Of the activity-related behaviors

implicated to help explain high strength-related cross-sectional properties in the

103

femoral midshaft of some males. This contention is supported by ethnohistorical
accounts from the region indicating that the Spanish in the area sometimes forced
some males into the repartimiento labor system, and subsequently ordered them
to make long distance trips to St. Augustine and elsewhere (Bushnell, 1981; Ruff
and Larsen, 1990). Other important factors that undoubtedly impacted the activity
patterns (and hence the skeletal structure) of the contact period inhabitants in the
area included epidemics, retaliation by the Spanish following revolts, attacks by
the English-occupied Carolina colony, increased population centralization, and an
overuse Of already marginal soils (Larsen, 1990).

For the Georgia coast humeri sample, the pattern Of decrease, then
increase, in diaphyseal strength between the three periods also holds for males;
for females, the initial decrease in the precontact agricultural period persists, but
does not reverse, and continues to decline, in the postcontact agricultural group.
Here, the suggestion is made that male participation in agricultural responsibilities
may have increased in the contact period, thereby increasing the mechanical
demands on the upper limb.

The results for the Michigan prehistoric vs. historic period comparison are
generally consistent with the prehistoric agricultural and historic period agricultural
results as described for the Georgia coast. In both regions, at least in terms Of
the data for male and female femora and for male humeri, individuals were
experiencing increased biomechanical demands postcontact. The departure

between the two studies can be seen with respect to the female humeri, where

104

on the Georgia coast there is a decline in strength in the historic period,
compared to Michigan, where there continues to be an increase. Although the
directions Of change for the Georgia coast and Michigan are broadly similar, the
reasons, as outlined above, that have been postulated to explain the Georgia
coast results are potentially quite different from those that might explain the
results from Michigan. Accordingly, it is useful to turn to the archaeological and
historical records in order to "tease out" those aspects Of human activity
(especially as relate to subsistence pursuits) that are consistent with the skeletal
results for the prehistoric and historic period comparison in Michigan.

The Michigan prehistoric samples consist of the Juntunen and Riviera aux
Vase sites. The Juntunen site has served as a type site for late prehistoric
settlement-subsistence systems in the northern upper Great Lakes (Fitting and
Cleland, 1969). Cleland (1966) has characterized the Juntunen site as a primarily
fishing village occupied by individuals who were (perhaps) only marginally
agricultural. The site is situated well north Of the limits for effective agriculture in
the Great Lakes region (Cleland, 1983), where at least 140 frost-free days are
needed for reliable corn production (Yamell, 1964). However, the results of the
faunal analysis from the site indicate that fish bone (mainly sturgeon) account for
only 66% Of the meat in the refuse from the Site (Cleland, 1966). The other 34%
consists Of a variety Of local fauna (McPherron, 1967) including beaver, dog,
porcupine, moose, and bear. The inhabitants of the Riviera aux Vase site were

primarily agricultural-based (Krakker, 1983), however, they too significantly

105

supplemented their diet with a broad range Of subsistence-related activities that
included hunting, gathering and some fishing (Bender, 1979; Wilkinson and Van
Wagenen, 1993).

The historic period Michigan samples are derived from the Lasanen and
Fletcher sites. Both are primarily agricultural but, similar to the situation
described above, here too the primary subsistence activities were likely Often-
times supplemented with hunting, gathering, and fishing. The point to emphasize
here is that in the Great Lakes region none Of major subsistence patterns (e.g.,
agriculture, fishing, or hunting-gathering) were entirely exclusive Of one another
(Tanner, 1987). In fact, Cleland (1983) has commented that the Great Lakes area
is characterized by a large diversity Of subsistence-settlement strategies. Hence,
the picture that emerges, during both the prehistoric and historic periods, is one
where a broad range Of physical activities related to food procurement were
carried out.

Related to this discussion Of food procurement is the important issue of the
extent to which European contact may have altered the subsistence pursuits (and
hence the activity patterns) of individuals in the historic period. There have been
many attempts to examine the impact Of Euro-Indian culture contact in the upper
Great Lakes; most of these have centered on the influence that the fur trade had
on aboriginal lifeways. Gerrnane to the present research is the work Of Fitting
(1976), who has addressed the phenomenon Of acculturation in the Straits of

Mackinac region by examining whether the introduction Of European trade goods,

106

particularly axes, kettles, knives, and guns, altered the subsistence patterns Of the
area. TO do this, he compares the faunal remains from the Juntunen site to those
from the early historic period Tionontate site in St. Ignace, which may well
represent the village associated with the Lasanen site burials. When adjusted for
differences in human population size, the faunal collections from the Juntunen site
and the Tionontate village are not only found to be comparable in size, but the
same species are dominant in both assemblages. Moreover, the trends in
species utilization at the Juntunen site are amplified at the Tionontate site in that
fish account for 99% Of the entire food refuse.

Accordingly, Fitting (1976) argues that if the introduction Of European trade
goods had any effect at all on the prehistoric subsistence strategy in the region,
it was to amelijy the trends already present. Hence he rejects the hypothesis that
European trade goods drastically altered the subsistence base of the pre-contact
peoples in the Straits Of Mackinac. This general thesis is supported by historic
period literature indicating that postcontact period Great Lakes Indians continued
their traditional methods Of food procurement, and that despite the introduction of
European-made items related to subsistence such as guns, traps, and metal
fishing gear, the pressures Of contact only intensified the precontact principal
subsistence regimes (Tanner, 1987).

The results Of the biomechanical analyses are consistent with these
archaeological and documentary data, and it is argued that this idea Of

"amplifying" subsistence-related (and hence activity-related) trends that were

107

already present precontact may help explain why postcontact individuals are
significantly stronger compared with those from the prehistoric period. In addition,
as has been previously described, the results Of the skeletal analyses also
indicate that the mechanical demands on females, even more so than for males,
increased after contact. The biomechanical data documenting large increases in
female femoral strength, especially at the articular ends Of the bone, are
consistent with female-related activities that emphasize bending and/or squatting,
such as corn hoeing and corn pounding. If the Fitting/Tanner theses are-correct,
it would follow that the amount Of time spent by females performing these type Of
activities would have significantly increased in the historic period. An independent
line Of evidence supporting this changing (in terms Of total work effort spent on
agricultural-related activities) role for historic period females is presented below.

As noted in Chapter 3, Bridges (1985, 1989a,b, 1991) has clearly
documented greater biomechanical demand in female Mississippian
agriculturalists compared to Archaic hunter-gatherers form northwestern Alabama,
indicating that subsistence agriculture is not less physically demanding compared
with hunting-gathering. In fact, it has long been documented by anthropologists
that an agricultural way Of life is more labor intensive and time-consuming than
is hunting-gathering (see, for example, Lee and DeVore, 1968 and Sahlins, 1972).

The humeral biomechanical data, especially at midshaft, also argues for an
increased workload for postcontact females. Compared to the forces generated

at the major joint surfaces, it is difficult to reconstruct, in a behavioral sense, the

108

types Of specific activities that might place increased mechanical stress on the
humeral midshaft. Nevertheless, we can speculate that a broad range Of lifting
behaviors, likely in conjunction with the subsistence-related activities described
above (e.g., corn production and/or the lifting Of heavy fish-laden nets associated
with the inland shore fishing subsistence adaptation well-documented for the
straits Of Mackinac region (Cleland, 1982)), is consistent with the increased
humeral strength results.

Other tantalizing evidence that may shed light on the skeletal results can
be found in Anderson (1994), who has recently examined the flow of European
(and especially French) made-trade goods into the western Great Lakes region
between A0. 1715 and AD. 1760. To do this, he uses data from the Montreal
Merchants’ Records (MMR), which consist Of the business records and invoices
Of numerous Montreal merchants. These merchants sold trade goods and
supplies to fur traders about to embark on their journey to trading outposts (eight
were examined) in the interior; hence the MMR represent an important record Of
the sale Of goods by European merchants to European traders. If the traders
purchased their items in accordance with their understanding Of Indian interests
in the trade goods, Anderson has argued that the MMR should accurately reflect
the types and quantities Of items most desired by the Indians. In order to
designate an item on the invoice as a trade good, archaeological and
documentary evidence was used such that if the item has been found on a

French-period Indian site, or if the item appeared on French-period trade lists in

109

documentary sources, it was designated a trade good. lmportantly,
archaeological evidence came from the Lasanen and Fletcher sites, the two
historic period sites examined in this research.

When all trade goods were ranked by functional category (e.g., cooking and
eating, hunting, clothing, weapons, amusements, adornment, etc.) it was found
that there was a very strong demand for clothing, which accounted for nearly two-
thirds Of the expenditure for all Of the goods shipped to all eight Of the trading
outposts. This finding has potentially important implications for the present study
in that it may help to illuminate the changing role for females in the historic period.
Specifically, Anderson argues that access to ready-made clothing, along with pins,
needles, and thread may have reduced the amount of time and labor that women
spent in the actual production of clothing. Perhaps this savings in time was
"reinvested" in physical activities related to the production Of trade items important
to the fur trade, for example garden produce.

Evidence in support Of this suggestion can be found in numerous Great
Lakes historical references that document the importance Of providing Europeans
with Native American-generated agricultural products. Among the Ottawa, for
example, while hunting and fishing was done by males, women planted and
tending fields Of corn, beans, and squash, gathered and dried wild berries,
harvested and dried cultivated crops, and made clothing and trade goods, such
as rush mats, baskets, birch bark boxes, and leather bags (Clifton et al., 1986).

These items not only added to the diversity Of the native foods, but they were also

110

traded in order to provision the French at Mackinac and Detroit (Feast and Feast,
1978). The Ottawa were also able to create wealth (and in turn prestige) in the
eyes Of the French by growing corn and by doing other jobs that the French did
not do themselves (Clifton et al., 1986). Additionally, McClurkin (1988) has
documented that the Ottawa were able to compensate. for lost revenue due to
their diminished role as middlemen in trade by selling provisions to French traders
passing trough the Straits of Mackinac area on their way westward. Whatever the
case, these lines Of evidence support the direction Of biomechanical Change seen
in long bone strength and architecture comparing individuals between the pre- and
postcontact periods. Taken together, they also place a heightened emphasis on
the already-important role that females played in a wide range of activities within

prehistoric and historic period Native American society.

Summapy and Conclusions

' Interpreted within the context Of physical activity, the increases in the cross-
sectional area and strength measures provide evidence for increased
biomechanical demand, for both sexes, in the Michigan historic period. These
biomechanical differences are found across the femoral and humeral diaphyses,
and point toward a generalized increased workload on both the arms and legs.
It is also worth noting that the magnitudeof the femoral cross-sectional size and
strength differences between the prehistoric and historic periods is generally

greater for females than for males. This finding suggests that the level Of activity

111

Changed more dramatically for females than for males between the two periods.

Alternatively, for the New York hunter-gatherer/agricultural comparison, the
results Of the biomechanical analyses do not argue for any significant increase (or
decrease) in workload in the latter group. These results are not in keeping with
those predicted from previous research on the hunter-gatherer agricultural
subsistence transition (Bridges, 1985, 1989a,b, 19991); as such much more work
is needed to identify the reasons for this unanticipated departure.

This research has attempted to document postcranial skeletal structural
adaptations in Native American populations from Michigan and New York and tO
relate the observed variability in long bone size and strength to the activity-related
behaviors Of these past peoples. The analyses attempted here for the prehistoric
vs. historic period comparison in Michigan, and for the‘ hunter-gatherer vs.
agricultural comparison in western New York, Of course, in no way constitute a
definitive study Of the skeletal structure-function relationship within each Of these
skeletal samples. It is hoped, however, that in some small way the results Of this
research have helped lay the groundwork for future studies aimed at elucidating
the biological and cultural changes that may have coincided with the contact
period in Michigan and the hunter-gatherer subsistence transition in western New
York. From this perspective, the results Of the current research, in some
substantive way, will add to what is known about the history of each region.

Not unexpectedly, this research has raised many questions that warrant

further investigation. It is hoped that the results of this study provide a basis for

112

interpreting the relationship between skeletal structure and behavior in a wider
regional context. As such, this study will help to broaden the perspective from
which anthropologists are able to address the complex synergistic relationship
between human activity, biocultural evolution, and associated modifications in

skeletal form and function.

Wr Fugue Research
( 1) Further interpretation Of biomechanical properties

Although the current research has gone far in documenting patterns Of
skeletal change in the two case studies, the most logical next step is to unravel
the functional significance of more Of the biomechanical properties. For example,
in comparative studies, the polar second moment Of inertia (J) has been used as
a measure Of overall long bone strength. A decline in the Irma/Imin ratio, a measure
of cross-sectional circularity, has been interpreted to indicate a decline in activity
level. The I,,IIy ratio represents a measure of relative bending strength in the
anteroposterior direction. Bending strengths about this plane, in the distal
humerus for example, are generated by flexion and extension at the elbow. In
future research efforts, it will be Of interest to see if the ethnohistorical record can
shed additional light on any male/female differences in the types of activities that

might reflect sexually dimorphic tendencies in these measurements.

113

(2) Bone mineral density

Although an analysis of bone mineral density (BMD) was beyond the scope
of this research, BMD data was collected at the time of CT scanning. This was
accomplished by including a phantom in the path of every CT scan taken.
Although absorptiometric modalities have more traditionally been used for
measuring BMD, CT scanners can also measure BMD if the appropriate
accommodations are made to the scanning protocol. The details of just such a
methodology are discussed in Chapter 4 under "The role of the phantom in CT
scanning."

BMD is a material property of bone that is influenced by nutrition, metabolic
factors and mechanical stimuli. In tandem with the architectural properties
examined here, BMD is an important component of bone strength, although in
bioanthropological contexts it is more often used to evaluate dietary-based
hypotheses related to sex-specific patterns of cortical bone loss with aging. In
this respect, BMD data would potentially complement the physical activity data by

addressing questions that are more directly related to diet and nutrition.

(3) Histomorphometry, mechanical loading and diet

This suggestion is perhaps more "intellectual" than practical. The
anthropometric and biomechanical analyses performed in the current research
measure skeletal adaptation at the macroscopic level. Anthropometry is done on

the external surfaces of bone, and CT allows the internal structure of bone to be

114

examined for biomechanical properties without materially compromising the
physical integrity of the bone. However, the dynamic nature of skeletal tissue, as
seen through its ability to continually modify itself throughout the course of an
individual’s lifetime, ultimately manifests itself at the cellular level. This involves
a delicate balance between the action of bone deposition, accomplished by
osteoblasts, and bone resorption, accomplished by osteoclasts.

The microscopic structures associated with this remodeling process, for
example osteons and Haversian systems, can be quantified, revealing important
information about the health of archaeological populations. However, these
structures can only be seen by a microscopic examination of bone thin sections,
which involves an invasive process. Stout (1983, 1989) has previously examined
skeletal remodeling in hunter-gatherers and agriculturalists from archaeological
sites in the midwestem United States, and his results demonstrate greater
remodeling rates in the latter. It is possible that these differences reflect
responses to both activity-related mechanical stimuli as well as to specific
nutritional components (Martin et al., 1985) characteristic of different subsistence
regimes (for example, the low calcium, high phosphorous ratios of maize-based
diets) (Larsen, 1995). It would be of interest to reveal the relative contribution
that each of these factors makes with respect to the skeletal populations

examined in this research.

APPENDICES

APPENDIX A

Femoral and Humeral Anthropometries
Measurements, Instrumentation and Explanations

APPENDIX A

FEMORAL AND HUMERAL ANTHROPOMETRICS

Femoral Anthropometrics

1. Maximum length
Instrument: osteometric board

Explanation: The posterior surface of the bone is placed facing the
osteometric board. The measurement represents the greatest distance from
the most distal projection of the medial condyle, resting against the fixed
upright, to the farthest distance obtainable while maneuvering the proximal
end of the bone from side to side along the board’s surface.

2. Bicondylar length
Instrument: osteometric board

Explanation: Also known as oblique or physiological length. The bone is
positioned posterior side down, with both condyles placed against the fixed
upright, and the movable block is positioned against the head.

3. Biomechanical length (FBL)
Instruments: osteometric board and first segment of anthropometer

Explanation: This measurement, used to calculate the locations of the five
CT scans, is referred to as femoral biomechanical length (FBL) for the
purposes of this study. The landmarks used to define this measurement are
identical to those described in detail in Ruff and Hayes (1983a). With the
posterior side of the bone positioned away from the observer, FBL is
represented by the average distance between the most distally-projecting
contact points of the medial and lateral condyles and the superior surface of
the femoral neck, just medial to where the neck merges joins with the
greater trochanter. This proximal endpoint is located near the insertion of
the Obturator intemus muscle. Gray (1976, p. 501) describes the Obturator
intemus muscle as being "inserted into the anterior part of the medial
surface of the greater trochanter proximal to the trochanteric fossa."

117

118

4. Maximum Trochanteric Length
Instrument: osteometric boa rd

Explanation: This measurement is also known as trochanter-condylar length.
Wrth the posterior side of the bone facing down, the greatest distance from
the most sUperior projecting aspect of the greater trochanter to the most
distal projection of the medial condyle is obtained. The measurement is
most readily taken by placing the medial condyle against the fixed upright of
the osteometric board and then aligning the movable block against the
highest point on the greater trochanter. The distance is then read off
millimeter graph paper affixed to the osteometric board.

5. Anteroposterior (A-P) midshaft diameter
Instrument: sliding caliper
Explanation: This measurement is based on a maximum length-derived
midpoint (as opposed to bicondylar or FBL length) and is taken in the
sagittal plane, at a right angle to the bone’s anterior surface, and
perpendicular to the mediolateral midshaft diameter (see below). The Iinea
aspera is not avoided when taking this measurement.
6. Mediolateral (M-L) midshaft diameter
Instrument: sliding caliper
Explanation: This measurement is also based on a maximum length-derived
femoral midpoint. It is taken in a transverse plane, at a right angle to the
sagittal plane, and perpendicular to the anteroposterior midshaft diameter.
7. Midshaft circumference
Instrument: flexible tape
Explanation: This measurement is also based on maximum femoral length.
It is taken along a plane corresponding to both the A-P and ML midshaft

diameters, and it represents the greatest distance around the exterior
surface of the femoral shaft.

8. Anteroposterior (A-P) subtrochanteric diameter

Instrument: sliding caliper

119

Explanation: This measurement is taken in the sagittal plane, distal to the
lesser trochanter. It is not represented by a well-defined, fixed, anatomical
landmark. In taking the measurement, the gluteal tuberosity (the lateral
ridge of the linea aspera) is avoided. It is taken three centimeters inferior to
the center of the prominence of the lesser trochanter, along the pectineal
line, which is represented by the intermediate ridge of the linea aspera. A-P
subtrochanteric diameter also demarcates the location, marked temporarily
with pencil on the lateral side of the bone, needed to "level" the femur in the
longitudinal, or A-P plane, for the purposes of computerized tomography
scanning orientation.

9. Mediolateral (M-L) subtrochanteric diameter
Instrument: sliding caliper
Explanation: This measurement is taken in the transverse plane, distal to the
lesser trochanter. It is recorded in the same location, but perpendicular to,
the A-P subtrochanteric diameter, which is located as described above.

10. Maximum (Vertical) head diameter

Instrument: sliding caliper

Explanation: This measurement represents the greatest distance obtainable
along the articular surface of the femoral head.

11. Horizontal (Transverse) head diameter
Instrument: sliding caliper

Explanation: This measurement represents the greatest distance obtainable
across the articular surface of the femoral head. It is most readily taken
when oriented with respect to an anteroposterior plane (i.e., perpendicular to
the plane established for the vertical head diameter measurement). The
bone is oriented such that the measurement can be taken while viewing the
fovea capitis.

1 2. Head girth
Instrument: flexible tape

Explanation: This measurement represents the maximum distance around
the articular surface of the head.

120

13. Minimum neck height
Instrument: sliding caliper

Explanation: This is a measurement of the minimum distance, taken in the
vertical plane, across the femoral neck.

14. Horizontal neck breadth
Instrument: sliding caliper
Explanation: Also a measurement of minimum femoral neck distance, taken
in a parasagittal plane, at the same location, but perpendicular to, the
orientation used to determine minimum neck height.

15. Neck girth
Instrument: flexible tape
Explanation: This measurement describes the minimum distance around the
exterior surface of the femoral neck. When taking this measurement, care is
taken to pull the tape taught across the entire outer neck surface.

16. Anteroposterior (A-P) medial condylar length
Instrument: sliding caliper
Explanation: This measurement, representing the greatest distance across
the medial condylar surface, is taken with the femur oriented in a horizontal
plane relative to the observer. The measurement is defined as the distance
from the posterior edge of the tibial articular surface to the most forwardly
(i.e., highest) projecting aspect of the condyle.

17. Anteroposterior (A-P) lateral condylar length

Instrument: sliding caliper

Explanation: This measurement is taken with the femur oriented as
described above, but for the lateral condyle.

18. Proximal epiphyseal breadth

Instrument: sliding caliper

121

Explanation: This measurement represents the maximum distance between
the lateral edge of the greater trochanter and the articular surface on the
femoral head. It is taken by holding one end of the caliper fixed against the
medial edge of the greater trochanter and pivoting the bone mediolaterally
until the maximum distance on the (medial) articular surface is obtained.

19. Biepicondylar breadth
Instrument: sliding caliper

Explanation: This measurement represents the greatest distance obtainable
between the outer most projections of the medial and lateral epicondyles.

Humeral Anthropometrics

1. Maximum length
Instrument: osteometric board

Explanation: This measurement is taken from the most superior portion of
the humeral head to the most distal aspect on the medial edge of the
trochlea. The head is placed against the fixed vertical, and the bone is
rotated in all directions until the maximum length is obtained.

2. Biomechanical length (HBL)
Instrument: first segment of anthropometer

Explanation: This measurement, used to calculate the locations of the five
CT scans, is referred to as humeral biomechanical length (HBL) for the
purposes of this study. The anatomical landmarks associated with this
measurement, described in Ruff and Larsen (1990, p. 97), are the proximal
surface of the humeral head and the distal edge of the lateral lip of the
trochlea.

3. Lesser tubercle-trochlear length
Instrument: first segment of anthropometer
Explanation:th the bone lying posterior side down, the maximum distance

from the highest point on the lesser tubercle to the most distal projection of
the trochlea, generally lying along its medial lip, is obtained.

122

. Greater tubercle-medial epicondylar length

Instrument: first segment of anthropometer

Explanation: This measurement represents the greatest distance from the
most superior aspect of the greater tubercle to the inferior edge of the
medial epicondyle.

. Head-medial epicondylar length

Instrument: first segment of anthropometer

Explanation: With the bone oriented posterior side down, this measurement
represents the greatest distance from the most superior aspect of the
humeral head to the inferior margin of the medial epicondyle.

. Maximum midshaft diameter

Instruments: sliding caliper and osteometric board

Explanation: This measurement is based on maximum humeral length. The
midshaft location is marked at the time the maximum length measurement
was taken. The measurement is obtained by rotating the shaft in an
anteromedial direction (Bass, 1987), until the maximum diameter is obtained.
. Minimum midshaft diameter

Instrument: sliding caliper

Explanation: After establishing the midshaft position according to the method
described above, this measurement is taken at 90° to the maximum
midshaft diameter.

. Midehaft circumference

Instrument: flexible tape

Explanation: Again with the midshaft position located as described above,
this measurement represents the greatest distance around the external
surface of the bone.

. Maximum (Vertical) head diameter

Instrument: sliding caliper

123

Explanation: The measurement represents the greatest distance obtainable
along the articular surface of the humeral head.

10. Horizontal (Transverse) head diameter
Instrument: sliding caliper
Explanation: This measurement represents the greatest distance obtainable
across the articular surface of the humeral head, viewed in an A-P plane,

that is, perpendicular to the plane established for the vertical head diameter
measurement.

11. Head girth
Instrument: flexible tape

Explanation: This measurement represents the maximum distance around
the articular surface of the head.

12. Anterior trochlear breadth
Instrument: sliding caliper
Explanation: Positioned posterior side down with the distal end of the bone
facing the observer, this measurement represents the greatest distance
between the medial and lateral trochlear margins.

13. Posterior trochlear breadth
Instrument: sliding caliper
Explanation: This measurement is taken with the bone positioned posterior
side down with the distal end of the bone facing the observer. The
measurement represents the greatest distance between the medial and
lateral trochlear margins.

14. Biepicondylar breadth
Instrument: sliding caliper
Explanation: The humerus is positioned posterior side down with the distal

end facing the observer. The measurement represents the greatest distance
between the medial and lateral epicondyles.

APPENDIX B

Results for Femoral and Humeral Anthropometric Data
Michigan and New York Samples

Tables 81-88

APPENDIX B

Table B1

Femoral Anthropometrics: Michigan Males

 

Length

 

460.3 1 20.9 438.4 1 24.6 -21.9 *
455.7 1 21.2 434.6 1 26.5 -21.1 *
438.2 1 21.1 416.9 1 25.1 -21.3 *
450.8 1 22.2 431.6 1 24.1 -19.2 *

Maximum

 

Bicondylar
P

 

I Biomechanical

 

Maximum
Trochanteric

Midshaft Diameter

Anteroposterior II 29.8 1 2.7 29.8 1 2.8 0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Mediolateral 26.6 1 2.2 27.0 1 2.4 0.4
Circumference 88.7 1 6.8 88.5 1 7.6 -0.2
Subtrochanteric
Diameter
Anteroposterior 27.1 1 2.6 29.3 1 2.2 2.2 *
Mediolateral 32.0 1 3.0 29.4 1 3.4 -2.6 *
Head Diameter
Maximum 46.9 1 2.4 46.1 1 3.6 ~0.8
Horizontal 45.9 1 2.2 44.8 1 3.0 -1.1
Girth 147.6 1 7.1 142.9 1 9.9 -4.7
Neck
Minimum Height 32.0 1 2.5 31.2 1 2.1 -0.8
Horizontal Breadth 26.8 1 2.3 26.4 1 2.2 -0.4

 

 

 

 

 

 

 

126
Table B1 (continued)

 
   

 

Variable‘

 

 

 

 

 

 

 

 

  

 

 

 

 

 
 

I Condylar Length -
[I Medial 63.9 1 4.3 I 62.9 1 5.3 -1.0
“ Lateral 64.4 1 3.5 I 62.4 1 2.8 -2.0
Breadth
Proximal 98 4 1 4 9 93.9 1 7 5 -4 5
Epiphyseal .
Biepicondylar 81.6 1 3.9

 

‘ All measurements are in millimeters (mean 1 SD).

2 Juntunen and Riviere aux Vase sites.

3 Lasanen and Fletcher sites.

‘ Difference and direction of change ("+" or "-") between sample means.
5 Significance level for two-sided t-tests at p<0.05.

 

 

 

Femoral Anthropometrics: Michigan Females

“PW—”Changer—
Variable_‘_ I

Table 32

    

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Length ,
II Maximum 436.3 1 22.7 414.8 1 18.4 -21.5 A”
Bicondylar 430.5 1 22.9 410.4 1 17.3 -20.1
Biomechanical 413.2 1 21.5 395.0 1 15.7 -18.2 ll
Maximum 424.5 1 22.7 409.2 1 16.3 -15.3 ll
Trochanteric
Midshaft Diameter %l
Anteroposterior 26.8 1 2.9 25.5 1 1.8 -1.3
Mediolateral 25.0 1 1.7 25.1 1 1.9 0.1
Circumference 81.8 1 6.0 80.1 1 4.8 -1.7
Subtrochanteric
Diameter
Anteroposterior 25.0 1 2.1 25.3 1 2.5 0.5
Mediolateral 29.2 1 3.0 29.0 1 2.6 -0.2
Head Diameter
Maximum 41.8 1 2.7 43.1 1 2.9 1.3
Horizontal 40.9 1 2.8 42.4 1 2.9 1.5
Girth 130.0 1 8.5 134.8 1 8.2 4.8
Neck
Minimum Height 28.1 1 2.1 29.1 1 2.1 1.0
Horizontal Breadth 23.3 1 2.3 24.2 1 1.8 0.9
Girth 85.3 1 6.0 89.3 1 5.1 4.0
Condylar Length
Medial 56.9 1 4.5 57.3 1 3.0 0.4

 

 

58312.2

 

 

 

128
Table B2 (continued)

 

 

 

Prehistoric’ Historic3 Change4 p'5
Variable‘
Breadth
Proximal ' 88.4 1 5.3 86.6 1 4.1 -1.8
Epiphyseal

 

 

 

 

 

 

See bottom of Table B1 for key.

1 29
Table 33

Femoral Morphometrics: New York Males

Hunt-Gach 191611116163 Change
_Vari_able___

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Length
LMaximum 456.1 1 12.1 458.0 1 24.5 1.9
Bicondylar 450.8 1 13.1 453.0 1 26.0 2.2
Biomechanical 430.3 1 13.6 431.9 1 24.1 1.6
II Maximum _ 442.9 1 15.7 439.0 1 23.8 -3.9
Trochanteric
II Midshaft Diameter
Anteroposterior 28.6 1 1.7 30.0 1 2.2 1.4
Mediolateral 26.6 1 1.6 25.5 1 1.8 -1.1
Circumference 86.9 1 4.6 87.1 1 5.2 0.2
Subtrochanteric
Diameter

 

29.7122 I 25212.1 I -4.5 I *
30.3125 I 32311.9 I 2.0 I *

Anteroposterior

 

Mediolateral

 

Head Diameter

Maximum 46.4 1 2.1 46.2 1 2.7 -0.2
Horizontal 45.6 1 1.8 45.4 1 2.9 -0.2
Girth 145.9 1 6.0 144.5 1 8.6 -1.4

 

 

 

 

 

 

 

 

 

ll
Neck
Minimum Height

 

 

 

 

 

 

 

 

31.9 1 1.4 32.2 1 2.8 0.3
Horizontal Breadth 26.8 1 2.1 27.4 1 2.5 0.6
Girth 97.8 1 4.6 97.8 1 8.3 0
Condylar Length -
' Medial I 61.6 1 3.4 62.2 1 2.7 0.6
I , Lafe'a' I 642 i___3 54;} 2_2 __ __ _ ___1

 

 

#_ _ ._*_7, ____.

130
Table B3 (continued)

Femoral Morphometrics: New York Males

 

-_.—_____.1

 

 

 

 

 

 

Hunt-Gath2 . Agriculture” Change‘ p5
Variable‘
II Breadth
Proximal . 98.0 1 2.9 96.2 1 6.6 -1.8
Epiphyseal I
Biepicondylar ‘ 79.6 1 2.8 79.7 1 3.7 0.1

 

‘ All measurements are in millimeters (mean 1 SD)

2 Frontenac Island site.

3 Tram and Harscher sites.
‘ Difference and direction of change ("+" or "-") between sample means.

5 Significance level for two-sided t-tests at p<0.05.

131
Table B4

Femoral Anthropometrics: New York Females

Variable‘

    

 

 

      
  
 

    
   

 

        
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I Length
I Maximum 423.4 1 16.5 425.7 1 19.1 2.3
I Bicondylar 419.0 1 19.4 420.5 1 19.1 1.5
I Biomechanical 400.3 1 18.0 405.4 1 15.6 5.1 II
Maximum 408.4 1 16.9 481.4 1 14.3 10.0
Trochanteric
Midshaft Diameter
I» Anteroposterior [ 25.3 1 2.1 27.3 1 1.8 2.0 *
Mediolateral 25.0 1 1.7 24.2 1 1.6 -0.8
Circumference 78.2 1 5.4 80.8 1 4.6 2.6
Subtrochanteric
Diameter
Anteroposterior ll 25.6 1 1.1 I 23.8 1 1.3 I -1.8 * 1
Mediolateral 27.8 1 3.9 I 30.9 1 3.1 I 3.1
Head Diameter 7 I
Maximum 42.1 1 1.3 42.1 1 2.1 0
Horizontal 41.7 1 1.5 41.5 1 2.1 -0.2
Girth 132.2 1 4.4 132.1 1 6.3 -0.1
Neck
Minimum Height 28.7 1 1.2 29.2 1 1.9 0.5
Horizontal Breadth 24.0 1 1.1 24.9 1 2.1 0.9
Girth 86.5 1 3.4 89.2 1 5.4 2.7
Condylar Length
Medial 55.0 1 4.3 57.5 1 3.5 2.5 II
Lateral

 

 

 

132
Table B4 (continued)

 

 

 

 

Hunt-Gath2 Agriculture“
Variable‘
[Breadth
Proximal 90.0 1 4.7 85.7 1 5.1 -4.3
Epiphyseal
Biepicondylar 68.0 1 0.0 4.4

 

 

 

 

See bottom of Table 83 for key.

1 33
Table B5

Humeral Anthropometrics: Michigan Males

Variable‘ I

   

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Length
Maximum 333.7 1 16.8 314.2 1 16.0 -19.5 * II
Biomechanical 330.4 1 16.5 307.9 1 15.3 -22.5 * II
Lesser Tubercle- 320.4 1 15.8 297.1 1 12.9 -23.3 * II
Trochlear
Greater Tubercle- . 322.3 1 16.5 298.7 1 13.9 -23.6 * II
Medial Epicondylar
Head-Medial 325.2 1 16.5 304.7 1 15.3 -20.5 *
Epicondylar

Midshaft Diameter
Maximum 23.4 1 1.7 23.7 1 1.4 0.3 II
Minimum 16.9 1 1.3 17.2 1 1.6 0.3 II
Circumference 67.5 1 4.3 68.6 1 4.6 1.1 II

Head Diameter
Maximum 45.2 1 2.2 44.1 1 2.8 -1.1
Horizontal 42.3 1 2.7 40.8 1 2.0 -1.5

I Girth 136.5 1 5.4 129.6 1 6.5 -6.9 *

 

 

 

 

    
 

. Breadth

Anterior Trochlear

 

44.4 1 2.1 43.5 1 1.9 -0.9
24.0 1 1.4 23.9 1 1.9 -0.1

    
 

 

Posterior Trochlear

      

 

  

 

 

 

Biepicondylar

 

See bottom of Table B1 for key.

1 34
Table 86

Humeral Anthropometrics: Michigan Females

l— _—_ m h I
L____.W__ _______-____ __ - _______________ ____ . ______ __ _- ___I

' Length

       

 

     

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

    
   
 

 

 

Maximum 305.3 1 17.3 296.7 1 10.6 -8.6

IL Biomechanical 302.9 1 16.5 292.7 1 10.1 -10.2 in

IL Lesser Tubercle- 294.9 1 16.5 285.4 1 9.7 -9.5
Trochlear
GreaterTubercle- ' 296.8 1 17.1 288.6 1 9.3 -8.2
Medial Epicondylar
Head-Medial 299.6 1 17.4 291.6 1 9.3 -8.0
Epicondylar

Midshaft Diameter
Maximum Y 20.2 1 1.7 21.2 1 1.7 1.0 *
Minimum 14.3 1 1.6 14.8 1 1.6 0.5
Circumference 58.3 1 5.0 60.7 1 4.9 2.4
Head Diameter

Maximum 40.7 1 2.4 40.3 1 1.4 -0.4
Horizontal 37.6 1 2.2 37.3 1 1.6 -0.3
Girth , 121.4 1 7.5 120.8 1 3.5 -0.6

II Breadth
Anterior Trochlear 39.5 1 2.6 40.7 1 2.2 1.2
Posterior Trochlear 22.2 1 1.8 22.3 1 1.6 0.1
Biepicondylar 56.0 1 4.2 54.1 1 4.0

 

 

 

 

 

     

See bottom of Table B1 for key.

1 35
Table B7

Humeral Anthropometrics: New York Males

 
 

 

 

 

 

 

 

 

 

 

Variable‘

Length
Maximum 312.5 1 30.7 318.4 1 15.3 5.9
Biomechanical 308.3 1 30.4 314.2 1 15.1 5.9
Lesser Tubercle- 296.0 1 28.6 307.5 1 14.9 11.5
Trochlear
Greater Tuberclev ' 301.8 1 30.3 . ‘ 304.9 1 14.9 3.1
Medial Epicondylar
Head-Medial 305.0 1 30.1 ' 309.1 1 15.0 4.1
Epicondylar

 

 

 

 

Midshaft Diameter

 

Maximum 22.8 1 3.2 21.7 1 1.5 -1.1
16.8 11.9 15.7 11.4 -1.1

66.2 1 9.3 62.2 1 4.6 4.0

 

Minimum

 

Circumference

 

 

 

 

Head Diameter

Maximum El 43.8 1 3.2 43.1 1 3.2 -0.7

 

 

 

 

 

 

 

 

 

 

   

Horizontal 40.8 1 4.0 40.6 1 2.7 -0.2

Girth 128.8 1 11.3 131.2 1 7.6 2.4
Breadth

Anterior Trochlear 43.8 1 4.2 43.2 1 2.1 -0.6

Posterior Trochlear 25.7 1 15.4 22.9 1 1.2 -2.8

Biepicondylar 60.0 1 5.8 59.4 1 3.5 -0 6

 

 

 

 

 

_h-—=§==J_

 

See bottom of Table B3 for key.

1 36
Table BS

Humeral Anthropometrics: New York Females

Hunt-Gath2 Agriculture"

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Maximum 306.8 1 19.6 303.8 1 14.0 -3.0
Biomechanical 302.7 1 19.0 299.9 1 13.5 -2.8
Lesser Tubercle- 291.7 1 16.3 293.8 1 12.3 2.1
Trochlear
Greater Tubercle- 300.8 1 17.2 292.4 1 12.8 8.4
Medial Epicondylar
Head-Medial 304.2 1 16.2 295.3 1 13.4 8.9 jl
Epicondylar
Midshaft Diameter
Maximum ‘ 20.5 1 1.7 20.2 1 1.1 -o.3
Minimum 14.1 1 2.0 14.4 1 1.2 0.3
Circumference 57.9 1 5.4 57.9 1 3.4 0
Head Diameter
Maximum 42.0 1 3.3 40.1 1 2.2 -1.9
Horizontal 38.8 1 3.8 36.8 1 1.7 -2.0
Girth 123.8 1 10.0 119.1 1 5.9 -4.7
Breadth
Anterior Trochlear 39.3 1 2.5 40.1 1 2.3
Posterior Trochlear 21.8 1 1.1 22.2 1 1.6
Biepicondylar 57.1 1 5.4 54.9 1 3.2

 

 

 

   

See bottom of Table B3 for key.

APPENDIX C .

Results for Femoral and Humeral Biomechanical Data
Michigan and New York Samples

Tables C1-C12

APPENDIX C

Table C1

Biomechanical Data: Scan #11
Michigan-Male and Female Femoral Samples

_.——_.__.—__.
l
l
I

Cross-Sectional Change5
Properties2

 

A._.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

  

 

 

 

  

 

  

 

  

 

  

 

  

 

Males

CA 154.0 1 26.2 145.0 1 31.9 90

MA 299.6 1 48.9 368.1 1 64.3 68.5

n TA ‘ . 453.6 1 41.11 . 513.1 1 68.7 59.5
lm . 112.6 1 24.3 126.5 1 41.9 13.9

Imin 77.9 1 12.9 83.4 1 18.4 5.5

Ix 79.6 1 13.3 85.5 1 18.1 5.9

ly 111.0 1 24.5 124.4 1 42.1 ~ 13.4

J 190.5 1 35.2 209.9 1 58.4 19.4

PCA 34.2 1 6.4 28.5 1 5.9 -5.7
Imllm 1.44 1 0.14 1.51 1 0.27 0.07

If LII, 0.73 1 0.12 0.72 1 0.13 -0.01

I Females

n CA 136.4 1 17.5 133.6 1 18.8 -2.8
247.5 1 44.8 360.0 1 31.2 112.5
383.9 1 45.5 493.7 1 25.5 109.8

86.4 1 17.7 121.3 1 17.1 34.9

56.8 1 13.0 68.9 1 10.5 12.1

58.0 1 13.2 71.0 1 9.5 13.0

85.2 1 18.2 119.2 1 18.0 34.0

143.2 1 28.5 190.2 1 25.3 47.0

i _ 35.9 1 5.5 27.1 1 4.1 -8.8

   

 

138

 

 

 

 

 

 

 

139
Table C1 (continued)

 

   

 

   
 
 

 

 

 

 

 

 

 

 

 

 

Cross-Sectional Prehistoric3 Historic‘ Changes p6
Properties2
Imllm 1.54 1 0.25 1.78 1 0.23 0.24
LII, 0.69 1 0.14 0.60 1 0.06 -0.09
1Scan taken at 20% (i.e., femoral distal location) of biomechanical length (see text for
explanation).

2CA = cortical area; MA = medullary area; TA = total subperiosteal area; Imax and 'm =
maximum and minimum second moments of area, respectively; Ix and I, = second
moments of area about the x and y axes, respectively; J = polar second moment of
area; PCA = percent cortical area (CA/T A) x 100; lelm and lle represent cross-
sectional shape indices. All area measurements are in mm’; all second moments of
area measurements are in mm‘. For norrrializatibn, areas (CA. MA and TA) are
divided by (biomechanical length)2 and then multiplied by 105; second moments of
area (IN, I”, Ix, l,, J) are divided by (biomechanical length)‘ and then multiplied by
10 .

3Juntunen and Riviere aux Vase sites.

‘Lasanen and Fletcher sites.

5Difference and direction of change ("+" or "-") between sample means.
6Significance level for two-sided t-tests at p<0.05.

140
Table CZ

Biomechanical Data: Scan #31

Michigan Male and Female Femoral Samples

 

I
1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

  

 

   

 

   

 

  

 

  

 

  

 

      

 

 

 

   

  

 

 

 

 

 

 

CA 211.5 1 23.1 186.3 1 50.0 -25.2 1|
1. MA 85.8 1 27.4 125.5 1 36.4 39.7 II
TA 297.3 1 42.2 311.8 1 45.3 14.5 II
I“, 78.0 1 23.7 74.8 1 27.6 -3.2 All
I", 57.4 1 13.7 59.1 1 20.3 1.7
I, 69.6 1 16.7 69.3 1 25.1 -0.3 II
Iy 65.7 1 21.5 64.6 1 23.4 -1.1 II
J 135.3 1 36.8 133.9 1 47.5 -1.4 u
PCA 71.6 1 5.7 69.4 1 12.1 -12.2
Inn/IN 1.35 1 0.14 1.27 1 0.13 -0.08
1,11, 1.06 1 0.15 1.07 1 0.18 0.01
Females J
CA 191.3 1 17.9 192.3 1 18.7 1.0
MA 81.2 1 32.6 101.5 1 19.0 20.3 1]
TA 272.5 1 30.1 293.8 1 27.9 21.3
I... 63.1 1 12.6 72.5 1 14.3 9.4
1",, 48.0 1 8.6 53.2 1 7.4 5.2
i, 54.5 1 11.7 56.4 1 8.9 1.9
1y 56.6 1 9.4 69.3 1 12.9
111.1 119.8 125.7 1 21.5
J

141
Table C2 (continued)

 

 

--- ~——————————~————~-——~-— — ——————————————— — _,_ A —— - F=== 1
Cross-Sectional Prehistoric3 Historic4 Change5 p6
Properties2
'mflm 1.31 1 0.20 1.36 1 0.13 0.05

 

 

 

 

 

 

‘ 0.96 1 0.13 0.81 1 0.06 -0.15

_—__ _ .___—______..__A_____ _—___ -—_—_—‘_—__—-w_————. _ . *_ .__

______

1Scan taken at 50% (i.e., femoral midshaft location) of biomechanical length (see text
for explanation).

See bottom of Table C1 for key.

142
Table C3

Biomechanical Data: Scan #5‘
Michigan Male and Female Femoral Samples

    

Cross-Sectional %Change;
Properties

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

    

 

 

     

   

Males
CA 217.4 1 26.2 204.2 1 39.8 -13.2 ll
MA 152.9 1 54.5 165.0 1 36.6 12.1 ll
TA 370.3 1 59.5 369.2 1 40.4 -1.1 II
I“, 123.4 1 28.1 119.5 1 29.0 -3.9
I“, 70.1 1 22.3 65.0 1 17.5 -5.1
I" 93.0 1 25.3 92.1 1 24.8 -0.9
I, 100.5 1 28.1 92.4 1 29.7 -8.1
J 193.5 1 48.8 184.4 1 45.8 -9.1
PCA 58.7 1 9.0 55.3 1 8.9 -3.4
lm/lm 1.76 1 0.40 1.84 1 0.21 0.08
LII, 0.93 1 0.23 1.00 1 0.39 0.07

Females ll
CA 201.9 1 28.0 234.0 1 21.0 32.1 * ll
MA 124.2 1 36.2 140.2 1 41.2 16.0
TA 326.1 1 40.1 374.2 1 53.1 48.1 *
lm 106.2 1 22.5 138.7 1 33.7 32.5 *
I“, 51.4 1 14.1 69.3 1 17.9 17.9 *
Ix 71.0 1 16.6 92.2 1 26.7 21.2 *
I, 86.5 1 22.7 115.7 1 34.7

157.5 1 35.1 208.0 1 51.1

 

 

    

 

 

 

143
Table C3 (continued)

Cross-Sectional Prehistoric3 Historic‘
Properties2

 

2.00 1 0.15

 

 

 

 

 

1Scan taken at 80% (i.e., femoral subtrochanteric location) of biomechanical length
(see text for explanation).

See bottom of Table C1 for key.

144
Table C4

Biomechanical Data: Scan #1‘
New York Male Femoral Sample

 

CrossSectional Hunt-Gath3 Agriculture‘ Change5 I
Properties2 f f __ - - _I

 

 

 

 

 

 

 

 

 

 

 

 

 

Males i
CA I 159.6 1 15.6 145.2 1 21.5 -14.4 I
MA 303.7 1 65.3 309.5 1 77.4 5.8 I
TA , 463.4 1 62.5 454.8 1 82.7 -8.6 II
1m . 120.3 1 31.8 108.2 1 34.1 -12.1
1",, I 81.6 1 9.9 75.8 1 21.1 -5.8 I
i" 83.4 1 9.2 79.2 1 23.2 -4.2 I
I, 118.6 1 32.1 104.9 1 32.7 -13.7
J I 202.0 1 40.5 184.1 1 54.1 -179
PCA 34.4 1 6.0 31.9 1 5.6 -2.5
rum/1,1,, 1.47 1 0.23 1.43 1 0.18 0.04

0.70 1 0.11 0.76 1 0.12

 

 

 

 

 

 

1Scan taken at 20% (i.e., femoral distal location) of biomechanical length (see text for

explanation).

2See bottom of Table C1 for key.

3Frontenac Island site.
‘Tram and Harscher sites.

5Difference and direction of change ("+" or "-") between sample means.
“Significance level for two-sided t-tests at p<0.05.

145
Table CS

Biomechanical Data: Scan #3‘
New York Male Femoral Sample

|

    

I

     

I Cross-Sectional

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. Properties’
Males
CA 217.2 1 17.5 205.5 :1: 24.4 -11.7
MA 74.1 1 22.4 79.5 1 20.8 5.4
TA 1 291.3 1 25.7 285.1 1 27.8 -6.2
I“, 72.3 1 13.4 72.4 1 13.1 0.1
L... 57.3 1 9.4 53.0 1 11.5 -4.3
Ix 64.2 1 9.5 69.6 1 14.6 5.4
I, 65.3 1 12.9 55.7 1 10.3 -9.6
J 129.6 1 21.5 125.3 1 23.5 -4.3
PCA 74.6 :1: 6.3 72.1 1 6.3 -2.5
1,111“, 1.26 1 0.16 1.37 1 0.16 0.11
0.98 1 0.11 1.25 1 0.18

 

 

 

1Scan taken at 50% (i.e., femoral midshaft location) of biomechanical length (see text
for explanation).

See bottom of Table C4 for key.

146
Table C6

Biomechanical Data: Scan #51
New York Male Femoral Sample

   

Cross-Sectional Hunt-Gath3 Agriculture‘ Change5
Properties2

 

    

233.9 1 22.7 211.1 1 28.4

   

 

 

 

 

 

 

 

 

 

 

 

 

 

II MA 114.2 1 29.3 136.3 1 38.4 22.1
If TA 348.1 1 38.9 347.5 1 34.2 ’ -0.6
II I... 128.4 1 25.9 108.6 1 22.1 -19.8
1",, I 59.5 1 12.1 62.7 1 12.7 3.2
I, I 89.9 1 18.2 83.9 1 18.9 -6.0 II
I, I 98.0 1 20.1 87.4 1 16.6 -10.6
J I 187.9 1 37.2 171.3 1 30.3 -16.6
, PCA I 67.2 1 5.6 60.7 1 8.4 -6.5 II
I lm/lm I 2.16 1 0.19 1.73 1 0.40 -0.43 *
I LII I 0.92 1 0.10 0.96 1 0.21 0.04

‘Scan taken at 80% (i.e., femoral subtrochanteric location) of biomechanical length

(see text for explanation).

See bottom of Table C4 for key.

147
Table C7

Biomechanical Data: Scan #11
Michigan Male and Female Humeral Samples

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Properties
CA 182.2 1 22.3 197.5 1 40.7 15.3
MA 73.1 1 28.0 98.3 1 46.3 25.2
TA 255.3 1 30.3 295.8 1 45.1 40.5
In, 63.4 1 14.0 85.6 1 34.4 22.2
In," 41.6 1 9.2 51.4 1 12.6 9.8
L 42.9 1 9.8 54.7 1 15.3 11.8
I, 62.1 1 13.5 82.3 1 31.8 20.2
J 104.9 1 21.9 137.0 1 46.5 32.1
PCA 71.9 1 8.6 67.4 1 14.1 -4.5
'n-mflrrh 1.54 1 0.21 1.62 1 0.27 0.08
LII, 0.70 1 0.12 0.69 1 0.10 —0.01
Females

CA 153.5 1 32.7 167.2 1 27.0 13.7
MA 73.1 1 33.7 94.6 1 32.3 21.5
TA 226.6 1 32.3 261.8 1 47.8 35.2
IN 48.0 1 13.9 68.7 1 30.1 20.7
I“, 33.0 1 9.5 41.1 1 11.2 8.1
L 34.0 1 9.9 41.9 1 12.2 7.9
L 47.1 1 13.5 68.0 1 29.2 20.9
J 81.1 1 22.4 109.8 1 41.0 28.7
68. 0 1 12.6 = 64.6 1 8.0 -3.4

 

 

 

 

 

 

 

 

148
Table C7 (continued)

   
 

 
 
  

 

 

  
 

 

 

   
 

 

 

 

Cross-Sectional Prehistoric‘ Historic‘ Change5 p6
Propertiesz
lm/Im 1.47 1 0.21 1.62 1 0.31 0.15
0.73 1 0.11 0.65 1 0.12 -0.08

 

 

 

 

1Scan taken at 20% (i.e., humeral distal location) of biomechanical length (see text for
explanation).

See bottom of Table C1 for key.

149
Table C8

Biomechanical Data: Scan #3‘
Michigan Male and Female Humeral Samples

I Cross-Sectional Changes .
I Properties2 _ _, ____ _

   

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I Males
II CA 175.9 1 27.5 183.1 1 35.8 7.2
MA 110.9 1 40.2 134.9 1 41.3 24.0
TA 286.8 1 39.1 318.1 1 23.4 31.3
I“, 76.0 1 19.8 88.0 1 14.5 12.0
'm 44.6 1 10.5 51.1 1 11.1 6.5
. L 55.5 1 14.6 68.4 1 13.1 12.9
I I, 65.1 1 14.5 70.7 1 10.6 5.6
J 120.6 1 28.0 139.1 1 22.6 18.5
PCA 62.0 1 9.8 57.8 1 12.0 -4.2
lm/lmin 1.72 1 0.30 1.76 1 0.29 0.04
IJI, 0.85 1 0.11 0.97 1 0.10 0.12

Females

CA 154.3 1 32.9 158.7 1 24.8 4.4
MA 95.7 1 34.1 131.1 1 45.0 35.4
TA 250.1 1 32.0 289.9 1 49.1 39.8
I“, 59.5 1 14.5 77.4 1 21.0 17.9
Imin 32.1 1 8.4 38.5 1 12.4 6.4
L 43.9 1 11.0 60.3 1 16.9 16.4
I, 47.6 1 12.9 55.5 1 16.5 7.9
J 91.5 1 22.4 115.8 1 33.0 24.3
61.9 1 11.5 55.7 1 10.1 -6.2

 

 

 

 

 

 

1 50
Table C8 (continued)

 

 

 

 

Cross-Sectional Prehistoric3 Historic‘ Change5 p6
Propertiesz
IMO/In." ' 1.87 1 0.19 2.06 1 0.27 0.19
0.95 1 0.16 1.10 1 0.09 .

 

 

 

 

 

1Scan taken at 50% (i.e., humeral midshaft location) of biomechanical length (see text
for explanation).

See bottom of Table C1 for key.

151
Table 09

Biomechanical Data: Scan #5‘
Michigan Male and Female Humeral Samples

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

M CA 139.5 1 20.4 143.7 1 19.5 4.2
MA 240.2 1 46.8 256.4 1 54.5 16.2
TA 379.7 1 55.4 400.1 1 46.0 20.4
IN 80.1 1 20.1 84.5 1 17.6 4.4
I“, 63.5 1 16.9 69.4 1 9.7 5.9
lx 72.6 1 18.5 81.6 1 18.1 9.0
I, 71.0 1 19.1 72.3 1 8.9 1.3
J 143.6 1 36.2 153.8 1 26.4 10.2
PCA 37.1 1 5.2 36.5 1 7.4 -0.6
lmflmin 1.28 1 0.16 1.21 1 0.12 -0.07
lJl, 1.04 1 0.13 1.12 1 0.13 0.08

Females

CA 134.7 1 25.9 135.6 1 25.2 0.9
MA 192.8 1 37.7 223.5 1 59.7 30.7
TA 327.5 1 43.9 359.1 1 71.5 31.6
I“, 65.2 1 16.0 71.8 1 25.7 6.6
lm 51.3 1 14.9 60.0 1 22.6 8.7
I, 57.9 1 15.8 64.3 1 24.7 6.4
I! 58.5 1 15.4 67.4 1 24.9 8.9
J 116.5 1 30.1 131.7 1 48.2 15.2

41.3 1 7.3 38.5 1 7.5 -2.8

 

 

 

 

 

 

 

1 52
Table C9 (continued)

 

    

 

  

Cross-Sectional Historic‘ Changes p‘5
Properties2
lmllmh 1.30 1 0.17 1.20 1 0.07 -0.10 *

 

 

 

 

 

 

   

1,71y

1Scan taken at 80% (i.e., humeral proximal location) of biomechanical length (see text
for explanation).

1.00 1 0.16 0.96 1 0.16 -0.04

 

 

See bottom of Table C1 for key.

1 53
Table C10

Biomechanical Data: Scan 11111
New York Female Humeral Sample

Cross-Sectional Hunt-Gath3 Agriculture‘ Change5 I
Properties’ __ __ _ __ _ - ___ __

   

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Females
CA 172.0 1 28.9 147.2 1 27.0 -24.8
MA 51.4 1 15.0 83.1 1 33.5 31.7 *
TA 230.3 1 29.6 223.4 1 36.5 -6.9
I"m 49.1 1 16.1 48.8 1 12.5 —0.3
I", 34.5 1 10.3 31.9 1 7.3 -2.6
lx 35.2 1 10.7 32.5 1 7.5 -2.7
I, 48.4 1 15.6 48.1 1 12.2 -0.3
J 83.6 1 26.2 80.6 1 19.1 -3.0
PCA 77.1 1 5.1 64.5 1 11.8 -12.6
lmflm 1.42 1 0.11 1.53 1 0.18 0.11
0.73 1 0.05 0.68 1 0.08 -0.05

 

 

 

‘Scan taken at 20% (i.e., humeral distal location) of biomechanical length (see text for
explanation).

See bottom of Table C4 for key.

154
Table C11

Biomechanical Data: Scan #31
New York Female Humeral Sample

   

 

     

Cross-Sectional Hunt-Gath3

        

 

 

 

 
 

 

 

    

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Agriculture‘ Change5 7.]
Properties” T
Females

CA 178.2 1 29.7 141.5 1 30.6 -36.7 *
MA 3 61.9 1 16.6 105.8 1 40.4 43.9 *
TA ’ 240.1 1 33.0 247.3 1 32.1 7.2

I... 64.9 1 16.8 55.7 1 14.1 -9.2

I", 1 30.7 1 8.6 29.3 1 6.9 -1.4

1,. 1 I 46.5 1 13.3 43.4 1 10.8 -3.1

11 49.1 1 15.1 41.6 1 10.3 -7.5

.1 ; 95.5 1 24.7 85.0 1 20.1 -10.5

PCA ' 74.2 1 6.2 57.9 1 12.9 -16.3 *
1mm“, 5 2.13 1 0.27 1.91 1 0.25 -0.22

I 0.98 1 0.26

 

‘Scan taken at 50% (i.e., humeral midshaft location) of biomechanical length (see text

for explanation).

See bottom of Table C4 for key.

155

Table C12

Biomechanical Data: Scan #51

New York Female Humeral Sample

 

 

  

 

   

 

   

 

  

 

  

 

  

 

  

 

  

 

  

 

  

 

 

 

Cross-Sectional Hunt-Gath3 Agriculture‘ Changes
Properties2
Females
CA 139.4 1 13.8 124.8 1 23.1 -14.6
MA 204.0 1 74.9 191.2 1 47.2 -12.8
TA 343.4 1 75.2 316.0 1 50.3 -27.4
lm 71.1 1 20.6 57.9 1 15.1 -13.2
1,“." 55.4 1 18.7 47.4 1 13.2 -8.0
I" 60.0 1 19.0 52.1 1 15.1 -7.9
1y 66.5 1 19.9 53.3 1 13.3 -132
J 126.5 1 38.3 105.4 1 27.7 -21.1
PCA 42.6 1 11.7 40.2 1 8.2 -2.4
lw/Im 1.30 1 0.15 1.24 1 0.16 -0.06
0.90 1 0.09 = 0.97 1 0.14 0.07 ‘

 

 

 

 

 

 

1Scan taken at 80% (i.e., humeral proximal location) of biomechanical length (see text
for explanation).

See bottom of Table C4 for key.

APPENDIX D

Figures to Anthropometric and Biomechanical Results
Michigan and New York Samples

Figures D1-D31

 

 

Maximum Femoral Length (mm)

- Michigan Prehistoric

 

1
Michigan Historic
Males Females
l

 

 

Figure D1-Maximum length of the femur (mm): Michigan

157

158

 

   

Maximum Trochanteric Length (mm)
&
‘6’

409 . .Michigan Prehistoric

400 . Michigan Historic

Males Females

 

 

Figure D2-Maximum trochanteric length of the femur (mm): Michigan

 

159

 

 

c
.n
10..
.5
h

m
P

n

a
.nla
h
.nlv
M

 

AEEV .9955 croucacoobnzm n..<

 

 

Figure D3-Subtrochanteric A-P diameter of the femur (mm): Michigan

160

 

M-L Subtrochanteric Diameter (mm)

- Michigan Prehistoric

 

/ .Michigan Historic
i Males Females

1

Figure D4-Subtrochanteric M-L diameter of the femur (mm): Michigan

 

 

161

 

Neck Girth (mm)

- Michigan Prehistoric

 

, // .Michigan Historic
Males Females

 

 

 

Figure DS-Femoral neck girth (mm): Michigan

162

 

Proximal Epiphyseal Breadth (mm)

- Michigan Prehistoric

Michigan Historic

 

 

 

 

Figure D6-Proximal epiphyseal breadth of the femur (mm): Michigan

 

163

 

// .New York Agriculture

   

Z
a? - New York Hunt-Gath

  

E25 6.265 c.2822 m.<

 

 

Figure D7-Midshaft A—P diameter of the femur (mm): New York

164

 

 

31-

A-P Subtrochanteric Diameter (mm)

- New York Hunt-Gath

 

 

 

, V/// .New York Agriculture
Males Females

 

 

figure D8-Subtrochanteric A-P diameter of the femur (mm): New York

165

 

 

mm
mm
mm
mm
n//////////////////2

AEEV .2955 2.2550955 .32

  

Males Females

 

 

 

Figure D9—Subtrochanteric M-L diameter of the femur (mm): New York

166

 

Maximum Length (mm)

 

Males Females

 

- Michigan Prehistoric

Michigan Historic

 

Figure D1 O-Maximum length of the humerus (mm): Michigan

 

167

 

I 330 -]

w 320«

Lesser Tubercle-Trochlear Length (mm)

 

 

290.
.Michigan Prehistoric
280 .Michigan Historic

 

 

 

 

Figure D11-Lesser tubercle-trochlear lengfl1 of the humerus (mm): Michigan

168

 

330 -

 

Head-Medial Epicondylar Length (mm)

- Michigan Prehistoric

 

Michigan Historic

Males Females

 

L

Figure D12-Head-medial epicondylar length of the humerus (mm): Michigan

169

 

 

 

.m

.m .m

.m m

m h.

D. H

.m. .11.

M M,

I m

m7//////
E

a n n m m. w

AEEV 86:35 c5525. 825885.

  

 

 

Figure D13-Maximum midshaft diameter of the humerus (mm): Michigan

170

 

Head Girth (mm)

- Michigan Prehistoric

 

Michigan Historic

 

Males Females

Figure D14wHumeral head girth (mm):.Michigan

 

171

 

CT Scan #1 (Distal)

360 _ 368 I

- Michigan Prehistoric

Cross-Sectional Medullary Area (mm2)

 

//// Michigan Historic

 

 

L Males Females

 

Figure D15-Cross-sectional medullary area of the femur (mm2): Michigan

172

 

CT Scan #1 (Distal)
540-

Cross-Sectional Total Area (mm2)

- Michigan Prehistoric

 

Michigan Historic

Males Females

 

 

 

Figure D16-Cross-sectional total area of the femur (mm2): Michigan

173

 

CT Scan #1 (Distal)
130-

 

—-l
N
\r

.31
N
.1

”J

\\V

.1
.4
(a)

\\\\\\\\\\\\\\\\\\\ \‘ .

1001

Imax-Maximum Bending Strength (mm4)
8

k\\\\\\\

     

 

 

. - Michigan Prehistoric
1 Bi . . . .
80 .Mrchrgan Hrstonc
Males Females

 

Figure D17-lmax (maximum bending strength of the femur-mm4): Michigan

174

 

CT Scan #1 (Distal)

   
   
   
 

Males Females

I

I 220.

I _

I .2101

/

3‘ 2001 //

I E ‘

1 g 7191

I 5 I ‘
*5, 180
C

I g I
(D I

I 2 160<I

I .9

' 2

I O

I 'T

'3 1401 14

I 120I

I

I

- Michigan Prehistoric

Michigan Historic

 

Figure D18-J (torsional strength of the femur-mm4): Michigan

 

 

 

175

 

CT Scan #1 (Distal)

384

PCA-% Cortical Area (mm2)

 

' .Michigan Prehistoric

Michigan Historic

Males Females

 

Figure D19-PCA (% cortical area of the femur-mm2): Michigan

 

176

 

 

 

 

CT Scan #3 (Midshaft) I
130.
a: I
I E 1204
‘g i
E
1 0.
2’ 1
9
3 I
'3 1001
I 2
I 9
I .9 90‘
. g I
| ml I
8 80I M' . . .
9 I - Ichrgan PrehIStonc
. 0 f
‘ 70 Michigan Historic
I Males Females

 

Figure DZO-Cross-sectional medullary area of the femur (mm2): Michigan

177

 

I CT Scan # 5 (Subtrochanteric)
I 240I

- Michigan Prehistoric

Cross-Sectional Cortical Area (mm2)

 

Michigan Historic

Males Females

 

 

 

Figure D21-Cross-sectional cortical area of the femur (mm2): Michigan

178

 

 

Cross-Sectional Total Area (mm2)

CT Scan #5 (Subtrochanteric)
31101
I

;374‘

3704_
1‘ 370 .

   
   

3304 - Michigan Prehistoric

320 I Michigan Historic

Males Females

 

Figure D22-Cross-sectional total area of the femur (mmZ): Michigan

 

179

CT Scan #5 (Subtrochanteric)

150-I

 

 

- Michigan Prehistoric

 

 

I Males Females

 

 

Figure D23-lmax (maximum bending strength of the femur-mm4): Michigan

180

 

I CT Scan # 5 (Subtrochanteric)

- Michigan Prehistoric

 

lmin-Minimum Bending Strength (mm4)

Michigan Historic

Males Females

 

 

 

Figure D24-Imin (minimum bending strength of the femur-mm4): Michigan

181

 

CT Scan #5 (Subtrochanteric)
220 -

J-Torsional Strength (mm4)

- Michigan Prehistoric

 

Michigan Historic

Males Females

 

Figure DZS-J (torsional strength of the femur-mm4): Michigan

 

182

 

CT Scan #1 (Distal)

320 1

Cross-Sectional Total Area (mm2)

- Michigan Prehistoric

 

Michigan Historic

 

Males Females

 

Figure D26-Cross-sectional total area of the humerus (mm2): Michigan

183

 

CT Scan #1 (Distal)
604

1

Ix (M-L Bending Strength-mm4)

- Michigan Prehistoric

Michigan Historic

 

 

 

 

Figure DZ7-lx (M-L bending strength of the humerus-mm4): Michigan

184

 

CT Scan #1 (Distal)
150I

140.1

I r—~

1301

  

120i

J-Torsional Strength (mm4)
5

- Michigan Prehistoric

Michigan Historic

Males Females

 

 

 

Figure D28-J (torsional strength of the humems-mm4): Michigan

185

 

CT Scan #3 (Midshaft)
3401
I

I

- I
3204 _ I
318 I I

Cross-Sectional Total Area (mm2)

- Michigan Prehistoric

 

Michigan Historic

Males Females

 

 

 

Figure D29-Cross-sectional total area of the humerus (mm2): Michigan

186

 

CT Scan #3 (Midshaft)

70~

{—1

68

60+

 

50:

IX (M-L Bending Strength-mm4)

- Michigan Prehistoric

  

\\ \\\\\\\ V

40 I / .Michigan Historic

Males Females

 

I
I
I
I

 

Figure D30-Ix (M-L bending strength of the humerus-mm4): Michigan

187

 

 

901

70-

Cross-Sectional Medullary Area (mm2)

 

Figure D31-Cross-sectional medullary awe of the humerus (mm2): New York

 

CT Scan #1 (Distal)

- New York Hunt-Gath

New York Agriculture

 

APPENDIX E

Figures to CT Scanning

Figures 1-4

11
D

. mi...-

 

Figure 1 - Technicare 1440 CT Scanner

189

190

 

Figure 2 - Cross-sectional CT images (top to bottom 4 different left humeri)

191

, _J.\..:».v.~0«:sta.smn ‘ .

 

Figure 3 - CT long bone orientation box

192

SCHEHHTIC OF BONE SCANNING BPPQRATUS

tiinrf 2.111!” I‘LH’HN’I'I" 1 If.

1 thllz’“; (‘1 U41

.iLE Iilu‘fl FLnYFIZOfl o

.lvll 11111111541

B—_~—————————-—‘
111111 [1.4.
ivy-1113m—
:g_' . .1 7'er?! ‘.LU
W...

1
54.1mm. 80:: [WEE ..

PI! WL E ELM!” II. [ME

.8 CH FO.‘

 

Figure 4 - CT image of long bone scanning apparatus

LITERATURE CITED

LITERATURE CITED

Albrecht GH, Gelvin BR, and Hartman SE (1993) Ratios as a size adjustment in
morphometrics. Am. J. Phys. Anthropol. 91:441-468.

Ambrose J (1973) Cemputerized transverse axial scanning (tomography): Part 2.
Clinical application. Brit. J. Radiol. 46:1023-1047.

Ambrose J (1975) Proceedings: A brief review of the EMI scanner. Brit. J. Radiol.
49.605.

Anderson DL (1994) The flow of European trade goods into the western Great
Lakes region, 1715-1760. In Brown JSH, Eccles WJ, and Heldman DP (eds.):
The Fur Trade Revisited: Selected papers of the sixth North American fur trade
conference, Mackinac Island, Michigan, 1991. East Lansing: Michigan State
University Press, pp. 93-115.

Bass WM (1987) Human Osteology: Columbia: Miss. Archaeol. Soc.

Beck TJ, Ruff CB, and Bissessur K (1993) Age-related changes in female femoral
neck geometry: Implications for bone strength. Calcif. Tissue Int. Suppl. 1:S41-
S46. ‘

Bender SJ (1979) Paleodemographic analysis of a Late Woodland site in
southeastern Michigan. Midcont. J. Archaeol. 4:183-208.

Boyd DP, and Parker DL (1983) Basic principles of computed tomography. In AA
Moss, G Gamsu, and HK Genant (eds): Computed Tomography of the Body.
Philadelphia: W.B. Saunders Company, pp. 1-21.

Brashler JG, and Holman MB (1985) Late Woodland continuity and change in the
Saginaw Valley of Michigan. Arctic Anthropol. 22:141-152.

Bridges PS (1985) Changes in Long Bone Structure with the Transition to
Agriculture: Implications for Prehistoric Activities. Ph.D. dissertation: The
University of Michigan, Ann Arbor, MI.

Bridges PS (1989a) Bone cortical area in the evaluation of nutrition and activity
levels. Am. J. Human Biol. 1:785-792.

Bridges PS (1989b) Changes in activity with the shift to agriculture in the
southeastern United States. Curr. Anthropol. 30:385-394.

194

195

Bridges PS (1991) Skeletal evidence of changes in subsistence activities between
the Archaic and Mississippian time periods in northwestern Alabama. In ML
Powell, PS Bridges, and AM Wagner-Myers (eds): What Mean These Bones?
Tuscaloosa: University of Alabama Press, pp. 89—101.

Brock SL (1985) Biomechanical adaptation of the lower limb bones through time
in the prehistoric southwest. Ph.D. dissertation: University of New Mexico,
Albuquerque, NM.

Brock SL, and Ruff CB (1988) Diachronic patterns of change in structural
properties of the femur in the prehistoric American Southwest. Am. J. Phys.
Anthropol. 75:113-127.

Burr DB (1980) The relationship among physical, geometrical and mechanical
properties of bone, with a note on the properties of nonhuman primate bone.
Yrbk. Phys. Anthropol. 23:109-146. '

Burr DB, and Piotrowski G (1982) Femoral mechanics in the Lesser Bushbaby
(Galago senegalensis): Structural adaptations to leaping in Primates. Anat. Rec.
202:419-429.

Burr DB, Piotrowski G, and Miller G (1981) Structural strength of the macaque
femur. Am. J. Phys. Anthropol. 54:305-319.

Burstein AH, Currey J, Frankel VH, Heiple KG, Linseth P, and Vessely JC (1972)
Bone strength: The effect of screw holes. J. Bone Jt. Surg. 54:1143.

Burstein AH, Zika JM, Heiple KG, and Klein L (1975) Contribution of collagen and
mineral to the elastic-plastic properties of bone. J. Bone Jt. Surg. 57.956.

Bushnell AT (1981) The King’s coffer: Proprietors of the Spanish Florida Treasury,
1565-1702. Gainsville: University Presses of Florida.

Carter DR, and Spengler DM (1978) Mechanical properties and composition of
cortical bone. Clin. Orthop. Rel. Res. 135:192-217.

Chamay A, and Tschantz P (1972) Mechanical influences in bone remodeling:
Experimental research on Wolff’s Law. J. Biomech. 5:1 73.180.

Cleland CE (1965) Beta activity of the animal burials at the Juntunen site (20MK1)
on Bois Blanc Island. In AJ Jelinek and JE Fitting (eds.): Studies in the Natural
Radioactivity of Prehistoric Materials. Ann Arbor: University of Michigan, Museum
of Anthropology, Anthropol. Pap., No. 25, pp. 56-58.

196

Cleland CE (1966) The prehistoric animal ecology and ethnozoology of the upper
Great Lakes region. Ann Arbor: University of Michigan, Museum of Anthropology,
Anthropol. Pap., No. 29.

Cleland CE (1971) Analysis of burial associations and conclusions. In CE Cleland
(ed.): The Lasanen Site, an Historic Burial Locality in Mackinac County, Michigan.
East Lansing: Mich. State Univ., Anthropol. Ser., Vol. 1, No. 1, pp. 89-95.

Cleland CE (1982) The inland shore fishery of the northern Great Lakes: Its
development and importance in prehistory. Amer. Antiq. 47:761-784.

Cleland CE (1983) Indians in a changing environment. In SL Flader (ed.): The
Great Lakes Forest: An Environmental and Social History. Minneapolis: The
University of Minnesota Press, pp. 83-95.

Clifton, JA, Cornell GL, and McClurken JM (1986) People of the Three Fires: The
Ottawa, Potawatomi and Ojibwa of Michigan. Grand Rapids: Inter-Tribal Council.

Clute RD (1971) The physical anthropology of the Lasanen Site. In CE Cleland
(ed.): The Lasanen Site, an Historic Burial Locality in Mackinac County, Michigan.
East Lansing: Mich. State Univ., Anthropol. Ser., Vol. 1, No. 1, pp. 107-144.

Cody DD, and Flynn MJ (1989) A technique for measuring regional bone density
in human lumbar vertebral bodies. Med. Phys. 16:766-772.

Computed Tomographic Scanning (1977). National Academy of Sciences.
Washington, DC.

Currey JD (1968) The adaptation of bones to stress. J. Theo. Biol. 20.91-106.

Dickie DL, Goldstein SA, Flynn MJ, and Bridges P (1984) Analysis of the
geometric properties of human long bones using computer axial tomography. J.
Biomech. 17:868 (abstract).

Enlow DH (1968) Wolff’s Law and the factor of architectonic circumstance. Am.
J. Orthodont. 54:803-822.

Eyman CE and Bettarel (1965) Physical anthropology of the Juntunen Site,
20MK1, Bois Blanc Island, Michigan. Manuscript of file, University of Michigan,
Museum of Anthropology.

Feest JE, and Feest CF (1978) Ottawa. In BG Trigger (ed.): Handbook of North
American Indians, Volume 15, Northeast. Washington: Smithsonian Institution,
pp. 772-786.

197

Ferretti JL, Capozza RF, Mondelo N, and Zanchetta JR (1993) Interrelationships
between densitometric, geometric, and mechanical properties of rat femora:
lnferences concerning mechanical regulation of bone modeling. J. Bone Miner.
Res. 8:1389-1396.

Fitting JE (1965) Late Woodland cultures of southeastern Michigan. Anthropol.
Pap., Mus. Anthropol., University of Michigan, No. 24.

Fitting JE (1976) Patterns of acculturation at the straits of Mackinac. In CE
Cleland (ed.): Culture Change and Continuity. New York: Academic Press, pp.
321-334.

Fitting JE (n.d.) The Mackinac Bubble: Economic theory and late Period trade in
the Straits of Mackinac. Unpublished manuscript on file at the Michigan History
Division.

Fitting JE and Cleland CE (1969) Late prehistoric settlement patterns in the upper
Great Lakes. Ethnohistory 16:289-302.

Fitting JE and Zurel RL (1976) The Detroit and St. Clair River Area. In DS Brose
(ed.): The Late Prehistory of the Lake Erie Drainage Basin: A 1972 Symposium
Revised. Cleveland: The Cleveland Museum of Natural History, pp. 214-250.

Fleagle JP (1977) Locomotor behavior and skeletal anatomy of sympatric
Malaysian leaf-monkeys (Presbytis obscure and Presbytis melalophos). Yrbk.
Phys. Anthropol. 20:440-453.

Frankel VH, and Burstein AH (1970) Orthopaedic Biomechanics. Philadelphia:
Lea & Febiger.

Fresia AE, Ruff CB, and Larsen CS (1990) Temporal decline in bilateral
asymmetry of the upper limb on the Georgia Coast. In CS Larsen (ed.): The
Archaeology of Mission Santa Catalina Guale: 2. Biocultural Interpretations of a
Population in Transition. New York: Anthropol. Pap. Am. Mus. Nat. Hist., No. 68,
pp. 121-150.

Frost HM (1964) The Laws of Bone Structure. Springfield: Charles C. Thomas.

Frost HM (1967) An Introduction to Biomechanics. Springfield: Charles C.
Thomas.

Frost HM (1973) Bone Modeling and Skeletal Modeling Errors. Springfield:
Charles C. Thomas. .

 

 

198

Frost HM (1995) Perspectives: On a "paradigm shift" developing in skeletal
science. Calcif. Tissue Int. 56:1-4.

Gambarelli J, Guerinel G, Chevrot L, and Mattei M (1977) Computerized Axial
Tomography. New York: Springer-Verlag Berlin Heidelberg, pp. 1-26.

Garn SM (1970) The Earlier Gain and Later Loss of Cortical Bone: Springfield:
Charles C. Thomas.

Genant H, Boyd D, Rosenfeld D, Abols Y, and Cam CE (1981) Computed
tomography. In SH Cohn (ed): Non-Invasive Measurements of Bone Mass and
Their Clinical Application. Boca Raton: CRC Press, Inc., 121-149.

Goodship AE, Lanyon LE, and MacFie H (1979) Functional adaptation of bone to
increased stress. J. Bone Jt. Surg. 61A2539-546.

Gordon R, Herman GT, and Johnson SA (1975) Image reconstruction from
projections. Scient. Am. 233:56-68. .

Gray H (1976) Gray’s Anatomy. Philadelphia: Lea & Febiger.
Greenman EF (1957) Riviere au Vase Site. Mich. Archaeol. 3911.

Hamilton ND, Peterson JB and McPherron A (1984) Perishable fiber industries in
the upper Great Lakes: A late Woodland case study from the Juntunen Site.
Paper presented at the 49th Annual Meeting of the Society for American
Archaeology, Portland, Oregon.

Harper VG (1945) A ceramic study of the Riviere au Vase Site in southeastern
Michigan. Papers of the Mich. Acad. Sci., Arts and Letters, Vol. 30:479-491. Ann
Arbor: University of Michigan Press.

Hayes WC (1986) Bone mechanics: from tissue mechanical properties to an
assessment of structural behavior. In GW Schmid-Schonbein, SL-Y Woo and BW
Zweifach (eds): Frontiers in Biomechanics. New York: Springer-Verlag, pp. 196-
209.

Hayes WC (1986) Basic Biomechanics of the Skeleton. In HK Uhthoff and E
Stahl (eds). Current Concepts of Bone Fragility. New York: Springer-Verlag, pp.
3-18.

Hayes WC, and Carter DR (1979) Biomechanics of Bone. In DJ Simmons and
AS Kunin (eds): Skeletal Research: An Experimental Approach. New York:
Academic Press, pp. 263-300.

199

Hayes WC, and Snyder BD (1981) Toward a quantitative formulation of Wolff’s
Law in trabecular bone. In Symposium on the Mechanical Properties of Bone.
Amer. Soc. Mech. Engineers, Boulder, CO. pp. 43-68.

Hert J, Liskova M, and Landrgot B (1969) Influence of the long-term, continuous
bending on the bone. Folia Morphol. 17:389-399.

Hinsdale WB and Cappannari SC (1941) Distribution of perforated human crania
in the western hemisphere. Papers of the Mich. Acad. Sci., Arts and Letters, Vol.
26:459-462. Ann Arbor: University of Michigan Press.

Hinsdale WB and Greenman EF (1936) Perforated Indian crania in Michigan. Ann
Arbor: University of Michigan, Museum of Anthropology. Occas. Contrib., No. 5.

Houghton F (1927) The migrations of the Seneca Indians. Am. Anthropol. 29:241-
250.

Hoyte DAN, and Enlow DH (1966) Wolff’s Law and the problem of muscle
attachment on resorptive surfaces of bone. Am. J. Phys. Anthropol. 24:205-214.

Hounsfield GN (1973) Computerized Transverse Axial Scanning (Tomography):
Part 1. Description of system. Brit. J. Radiol. 46:1016-1022.

Jones HH, Priest JD, Hayes WC, Tichenor CC, and Nagel DA (1977) Humeral
hypertrophy in response to exercise. J. Bone Jt. Surg. 59:204-208.

Jungers WL, and Minns RJ (1979) Computed tomography and biomechanical
analysis of fossil long bones. Am. J. Phys. Anthropol. 50:285-290.

Kannus P, Sievanen H., and Vuori l (1996) Physical loading, exercise, and bone.
Bone Suppl. 1:1S-3S.

Kerr D, Morton A, Dick l, and Prince R (1996) Exercise effects on bone mass in
postmenopausal women are site-specific and load-dependent. J. Bone Miner.
Res. 11:218-225.

Krakker JJ (1983) Changing sociocultural systems during the late Prehistoric
Period in southeast Michigan. Ph.D. dissertation, University of Michigan, Ann
Arbor, MI.

Lanyon LE (1984) Functional strain as a determinant of bone remodeling. Calcif.
Tissue Int. Suppl. 36:56-61.

200

Lanyon LE (1986) Biomechanical Factors in Adaptation of Bone Structure. In HK
Uhthoff and E Stahl (eds): Current Concepts of Bone Fragility. Berlin: Springer-
Verlag, pp. 19-33.

Lanyon LE (1996) Using functional loading to influence bone mass and
architecture: Objectives, mechanisms, and relationship with estrogen of the
mechanically adaptive process in bone. Suppl. 1:37S-43S.

Lanyon LE, and Baggott DG (1976) Mechanical function as an influence on the
structure and form of bone. J. Bone Jt. Surg. 58:436-443.

Lanyon LE, and Bourn S (1979) The influence of mechanical function on the
development and remodelling of the tibia. J. Bone Jt. Surg. 61:263-273.

Lanyon LE, and Rubin CT (1983) Functional load bearing as a determinant for
bone remodelling. In B Frame and JT Potts Jr. (eds): Clinical disorders of bone
and mineral metabolism. Amsterdam: Excerpta Medica, pp. 183-185.

Lanyon LE, and Rubin CT (1984) Static vs. dynamic loads as an influence on
bone remodelling. J. Biomech. 17:897-905.

Larsen CS (1982) The anthropology of St. Catherines Island. Prehistoric human
biological adaptation. Anthropol. Pap. Am. Mus. Nat. Hist., Vol. 57, Part 3. New
York.

Larsen CS (1984) Health and disease in Prehistoric Georgia: The transition to
agriculture. In MN Cohen and GJ Arrnelagos (eds): Paleopathology at the
Origins of Agriculture. New York: Academic Press, pp., 367-392

Larsen CS (1987) Bioarchaeological interpretations of subsistence economy and
behavior from human skeletal remains. In MB Schiffer (ed.): Advances in
Archaeological Method and Theory. San Diego: Academic Press, Vol. 10., pp.
339-445.

Larsen CS (1990) Biological interpretation and the context for contact. In CS
Larsen (ed.): The Archaeology of Mission Santa Catalina Guale: 2. Biocultural
Interpretations of a Population in Transition. New York: Anthropol. Pap. Am. Mus.
Nat. Hist., No. 68, pp. 11-25.

Larsen CS (1995) Biological changes in human populations with agriculture.
Annu. Rev. Anthropol. 24:185-213. '

Larsen CS (1997) Bioarchaeology: Interpreting Behavior from the Human
Skeleton. Cambridge: Cambridge University Press, pp. 195-225.

 

201

Larsen CS, Kelly RL, Ruff CB, Schoeninger MJ, and Hutchinson DL (1996)
Biobehavioral adaptations in the western Great Basin. In EJ Reitz, LA Newson
and SJ Scudder (eds): Case Studies in Environmental Archaeology. New York:
Plenum Press, pp. 149-174.

Larsen CS, and Ruff CB (1991) Biomechanical adaptation and behavior on the
Prehistoric Georgia Coast. In ML Powell, PS Bridges, and AM Wagner Mires
(eds): What Mean These Bones? Tuscaloosa: University of Alabama Press, pp.
102-113.

Ledley RS, Di Chiro G, Luessenhop AJ, and Twigg HL (1974) Computerized
transaxial x-ray tomography of the human body. Science 186:207-212.

Lee RB, and DeVore l (eds) (1968) Man the Hunter. New York: Aldine
Publishing Company.

Liskova M, and Hert J (1971) Reaction of bone to mechanical stimuli. Folia
Morphol. 19:301-317.

Lovejoy CO, Burstein AH, and Heiple KG (1976) The biomechanical analysis of
bone strength: A method and its implication to platycnemia. Am. J. Phys.
Anthropol. 44:489-506.

Lovis WA (1985) Seasonal settlement dynamics and the role of the Fletcher Site
in the Woodland adaptations of the Saginaw Drainage Basin. Arctic Anthropol.
22:153-170.

Mainfort RC (1977) The Fletcher site cemetery (20BY28) Bay 00., Michigan: A
study in the social dynamics of the contact period. Ph.D. dissertation, Michigan
State University, East Lansing, MI.

Mainfort RC (1979) Indian Social Dynamics in the Period of European Contact:
Fletcher Site Cemetery, Bay County, Michigan. East Lansing: Mich. State Univ.,
Anthropol. Ser., Vol. 1, No. 4.

Mainfort R (1985) Wealth, space, and status in a historic Indian cemetery. Am.
Ant. 50:555-579.

Martin DL, Goodman AH, and Armelagos GJ (1985) Skeletal pathologies as
indicators of quality and quantity of diet. In RI Gilbert Jr and JH Mielke (eds):
The Analysis of Prehistoric Diets. Orlando: Academic Press, pp. 227-279.

Martin RB, and Atkinson PJ (1977) Age and sex-related changes in the structure
and strength of the human femoral shaft. J. Biomech. 10:223-231.

202

Martin RB, and Burr DB (1989) Mechanical adaptation. In RB Martin and DB Burr
(eds): Structure, Function, and Adaptation of Compact Bone. New York: Raven
Press, pp. 143-185.

McClurkin JM (1988) We Wish to be Civilized: Ottawa-American Political Contests
on the Michigan Frontier. Ph.D. dissertation: Michigan State University, East
Lansing, MI.

McPherron A (1967) The Juntunen Site and the Late Woodland Prehistory of the
Upper Great Lakes Area. Ann Arbor: University of Michigan, Museum of
Anthropology, Anthropol. Pap., No. 30.

Morgan CL (1983) Basic Principles of Computed Tomography. Baltimore:
University Park Press.

Nelson DA (1985) An analysis of variations in bone density and cortical loss in
three Native American skeletal populations. Ph.D. dissertation, Michigan State
University, East Lansing, MI.

Nelson DA (n.d.) The Fletcher site cemetery: Ethnohistory, archaeology, and
osteology. Unpublished manuscript on file at Michigan State University, East
Lansing, MI.

Norusis MJ (1990) SPSS Base System User’s Guide. SPSS, Inc. Chicago:
Illinois.

Norusis MJ (1993) The SPSS for Windows Base System User’s Guide. SPSS,
Inc. Chicago: Illinois.

Ortner DJ and Putschar WGJ (1981) Identification of Pathological Conditions in
Human Skeletal Remains. Smithsonian Contributions to Anthropology, No. 28.
Washington: Smithsonian Institution Press.

O’Shea JM (1988) Social Organization and Mortuary Behavior in the Late
Woodland Period in Michigan. In RW Yerkes (ed.): Interpretations of Culture
Change in the Eastern Woodlands During the Late Woodland Period. Occas.
Pap. Anthropol., No 3. Columbus: Ohio State University, pp. 68-85.

Parfitt AM (1988) Bone remodeling. Henry Ford Hosp. Med. J. 36:143-144.
Peny J (1980) A discussion of CT scanner geometries and capabilities. AE van

Voorthuisen (ed): Proceedings of the First European Symposium on Fourth-
Generation CT Scanning, pp. 11-25. Amsterdam: Excerpta Medica.

203

Pfeiffer S (1977) The Skeletal Biology of Archaic Populations of the Great Lakes
Region. Ottawa: Nat. Mus Man Mercury Ser., Archaeol. Sur. Canada, Pap., No.
64.

Piotrowski G, Sullivan M, and Colahan PT (1983) Geometric Properties of Equine
Metacarpi. J. Biomech. 16:129-139.

Policy Implications of the Computed Tomography (CT) Scanner (1978). Office of
Technology Assessment. Washington, DC.

Raemsch CA (1993) Mechanical procedures involved in bone dismemberrnent and
defleshing in prehistoric Michigan. Midcont. J. Archaeol. 18:217-244.

Reilly DT, and Burstein AH (1974) The mechanical properties of cortical bone.
J. Bone Jt. Surg. 56:1001.

Ritchie WA (1944) The Pre-lroquoian occupations of New York state. Rochester
Mus. Arts, Sci., Mem. No.1.

Ritchie WA (1945) An early site in Cayuga County, New York. Rochester Mus.
Arts, Sci., Res. Rec. No. 7.

Ritchie WA (1965) The Archaeology of New York State. New York: The Natural
History Press.

Ruff CB (1981) Structural Changes in the Lower Limb Bones with Aging at Pecos
Pueblo. Ph.D. dissertation, University of Pennsylvania, Philadelphia, PA.

Ruff CB (1984) Allometry between length and cross-sectional dimensions of the
femur and tibia in Homo sapiens sapiens. Am. J. Phys. Anthropol. 65:347-358.

Ruff CB (1989) New approaches to structural evolution of limb bones in primates.
Folia Primatol. 53:142-159.

Ruff CB, and Hayes WC ( 1982) Subperiosteal expansion and cortical remodeling
of the human femur and tibia with aging. Science 217.945-948.

Ruff CB, and Hayes WC (1983a) Cross-sectional geometry of Pecos Pueblo
femora and tibiae-a biomechanical investigation: I. Method and general patterns
of variation. Am. J. Phys. Anthropol. 60:359-381.

Ruff CB, and Hayes WC (1983b) Cross-sectional geometry of Pecos Pueblo
femora and tibiae-a biomechanical investigation: ll. Sex, age and side
differences. Am. J. Phys. Anthropol. 60:383-400.

204

Ruff CB, and Larsen CS (1990) Postcranial biomechanical adaptations to
subsistence strategy changes on the Georgia Coast. In CS Larsen (ed.): The
Archaeology of Mission Santa Catalina Guale: 2. Biocultural Interpretations of a
Population in Transition. New York: Anthropol. Pap. Am. Mus. Nat. Hist, No. 68,
pp. 94-120.

Ruff CB, Larsen CS, and Hayes WC (1984) Structural changes in the femur with
the transition to agriculture on the Georgia Coast. Am. J. Phys. Anthropol.
64:126-136.

Ruff CB, and Leo FP (1986) Use of computed tomography in skeletal structure
research. Yrbk. Phys. Anthropol. 29:181-196.

Ruff CB, Trinkaus E, Walker A, and Larsen CS (1993) Postcranial robusticity in
Homo. l: Temporal trends and mechanical interpretation. Am. J. Phys. Anthropol.
91:21-53.

Ruff CB, Walker A, and Trinkaus E (1994) Postcranial robusticity in Homo: III.
Ontogeny 93:35-54.

Runestad JA, Ruff CB, Nieh JC, Thorington RW, and Teaford MF (1993)
Radiographic estimation of long bone cross-sectional geometric properties. Am.
J. Phys. Anthropol. 90:207-213.

Rybicki EF, Mills EJ, Turner AS, and Simonen FA (1977) In vivo and analytical
studies of forces and moments in equine long bones. J. Biomech. 10:701-705.

Sahlins MD (1972) Stone Age Economics. Chicago: Aldine-Atherton, Inc.

Sauer NJ (1974) An Analysis of the Human Skeletal Material From the Fletcher
Site (20BY28). Ph.D. dissertation, Michigan State University, East Lansing, MI.

Schwartz GT, and Conroy GC (1996) Cross-sectional geometric properties of the
Otavipithecus mandible. Am. J. Phys. Anthropol. 99.613-623.

Seeram E (1982) Computed Tomography Technology. Philadelphia: W. B.
Saunders Company.

Stein I, Stein R0, and Beller ML (1955) Living Bone in Health and Disease.
Philadelphia: J. B. Lippincott Company.

Stout SD (1983) The application of histomorphometric analysis to ancient skeletal
remains. Anthropos 10.60-71.

205

Stout (1989) Histomorphometric analysis of human skeletal remains. In MY Iscan
and KAR Kennedy (eds): Reconstruction of Life from the Skeleton. New York:
Alan R. Liss, pp. 41-52.

Sumner DR (1984) Size, shape, and bone mineral content of the human femur in
growth and aging. Ph.D. dissertation, The University of Arizona, Tucson, AZ.

Sumner DR, Mockbee B, Morse K, Cram T, and Pit M (1985) Computed
tomography and automated image analysis of prehistoric femora. Am. J. Phys.
Anthropol. 68225-232.

Tanner HH (1987) Atlas of Great Lakes Indian History. Norman: University of
Oklahoma Press.

Tate JR, and Cam CE (1982) High-resolution computed tomography for the
comparative study of fossil and extant bone. Am. J. Phys. Anthropol. 58.67-73.

Ter-Pogossian MM (1983) Physical principles and instrumentation. In JKT Lee,
SS Sagel, and RJ Stanley (eds): Computed Body Tomography. New York:
Raven Press, pp. 1-7.

Thwaites RG (1896-1901) The Jesuit relations and allied documents: Travels and
explorations of the Jesuit missionaries in New France 1610-1791. Cleveland: The
Burrows Brothers Co.

Townsley W (1948) The influence of mechanical factors on the development and
structure of bone. Am. J. Phys. Anthropol. 1:25-45.

Trinkhaus E, Churchill SE, and Ruff CB (1994) Postcranial robusticity in Homo:
II. Humeral bilateral asymmetry and bone plasticity. Am. J. Phys. Anthropol. 93:1-
34.

Weinmann JP, and Sicher H (1955) Bone and Bones: Fundamentals of Bone
Biology. Second Edition. St. Louis: The CV. Mosby Company.

White TD (1991) Human Osteology. New York: Academic Press.

Wilkinson RG and Van Wagenen KM (1993) Violence against women: Prehistoric
skeletal evidence from Michigan. Midcont. J. Archaeol. 18:190-216.

Winter J and King W, "I (1983) Basic Principles of Computed Tomography. In M
Greenberg (ed.): Essentials of Body Computed Tomography. Philadelphia: W. B.
Saunders Company, pp. 1-23.

206

Wolff J (1892) Das Gesetz der Transformation der Knochen. Berlin: A.
Hirschwald.

Woo SL-Y, Kuei SC, Amiel D, Hayes WC, White FC, and Akeson WH (1981) The
effect of prolonged physical training on the properties of long bone: A study of
Wolff’s Law. J. Bone Jt. Surg. 63:780-787.

Wray CF, and Cameron DG (1968) The Harscher Site: Burial excavations on a
prehistoric Iroquois site in the town of East Bloomfield, New York. Unpublished
report submitted to Rochester Museum and Science Center.

Wray CF, and Cameron DG (1970) Archaeological excavations on the Tram Site,
Livonia, New York. Unpublished report submitted to Rochester Museum and
Science Center.

Wray CF, Sempowski ML, and Saunders LP (1991) Tram and Cameron: Two
Early Contact Era Seneca Sites. The Charles F. Wray Series in Archaeology, Vol.
II, Rochester Museum and Science Center Research Records No. 21.

Wright H (n.d.) A description and history of investigations of the Juntunen site.
Unpublished manuscript on file at the University of Michigan Museum of
Anthropology. Ann Arbor.

Yarnell RA (1964) Aboriginal relationships between culture and plant life in the
upper Great Lakes region. Ann Arbor: University of Michigan, Museum of
Anthropology, Anthropol. Pap., No. 23.

"‘llllllllllllllllllllf