.. 1
am Fan .

.. .
Ewfima
. r . ,3
:2. , EA...

hr.

 

. . . . . x . )
. . is 1.. .
n24. . . . Emwafinaqqu
2. 1. .

it. , v...
.. I.’ a .1»

mmhmw

9.. a.

, h... .4... . ..
ihfifim. r.
$2.3...

fl.

1 . r“ .
. 2.... 1n.......,\

,. . . . . . . a. . ... l

.. . . ...fiww~..»tn»)n r. .1.

. 3 .i...‘ .... Ii «.1-

 

.2.2. , ..._:......f... ..3.......1.;.... . . .. . , . 1
355223.22 1.. . . . . . . ., . . .. . , . . . , . , . .. mum... hwwwxguawm

 

3.

 

. .
5k}... ..{2

This is to certify that the

dissertation entitled

THE EFFECT OF ENVIRONMENTAL CHANGE AND ECONOMIC POWER ON
THE DIET OF TARASCAN ELITES

presented by
LAURA CAHUE

has been accepted towards fulfillment
of the requirements for

Ph .D . degree in Anthropology

fl‘] Major professor

  

Date March 30, 2001

 

MSU i: an Affirmaliw Action /Equal Opportunity Institution 0- 12771

 

 

LIBRARY

Michigan State
University

 

 

 

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

 

DATE DUE DATE DUE ' DATE DUE

 

.A.; n A. dnn'!

@525 £26

 

HAN D 'J LUU’

 

éifp {277 90217

 

 

SEP 2 3 2001’

 

“58102008

 

T

SEP I/V‘ugu}
1"; )L‘Jl. EU
L ' '

l
.L

 

 

 

 

 

 

 

 

6/01 c-JCIRC/DateDuepGS—pJS

THE EFFECT OF ENVIRONMENTAL CHANGE AND ECONOMIC POWER ON
THE DIET OF TARASCAN ELITES

By

Laura Cahue

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Anthropology

2001

ABSTRACT

THE EFFECT OF ENVIRONMENTAL CHANGE AND ECONOMIC POWER ON
THE DIET OF TARASCAN ELITES

By

Laura Cahue

The Tarascan state emerged in the Lake Patzcuaro basin (LPB) during the
Postclassic period (AD. 900 - AD. 1520), when the water level of Lake Patzcuaro was
fluctuating. As these environmental changes were taking place and the food resources
shifted, the population in the basin was undergoing rapid growth, reaching an estimated
80,000 by the time the Spaniards arrived in AD. 1520 (Pollard, 1982; Pollard, 1993). By
comparing population size and shifting food resources in the LPB, Gorenstein and
Pollard (1983) showed that by the Late Postclassic (AD. 1400 to A.D.1520), the basin
was unable to produce enough maize to feed this population. How could the Tarascan
state, the second largest in Mesoamerica, and one that dominated western Mexico at the
time of European contact, emerge under such dietary constraints?

Ethnohistoric documents indicate that the Tarascan state was not an economically
viable unit, and that it existed, and even thrived, only through the establishment of
differential mechanisms of economic exchange (Pollard, 1982; Pollard, 1993). Non-
elites obtained goods (including maize and beans) through the intensification of local
and regional markets or subsistence activities, while the elites obtained most of their
goods by outright ownership of local production or by tribute secured through militarism

(Pollard, 1993). This study hypothesized that this differential adaptation of elites and

non-elites (which excluded agricultural intensification and hydraulic works) served to
buffer the elite population in the LPB from dietary change as the Tarascan state emerged.

To test this hypothesis the diets of Pre-Tarascan and Tarascan elites were
compared. Stable carbon and nitrogen isotopes (813C and 6'5N values respectively) of
bone collagen were used to assess temporal and regional variation in diet.

The results show that the diet of elites in the Lake Patzcuaro Basin did not change
over time. Among Tarascan administrative centers, elites at those administrative centers
with a higher rank showed diets that were higher in relative proportions of maize than
those of lower rank. The diet of elites at the capital city of Tzintzuntzan was significantly
different fiom that of elites elsewhere. Elites in the Sayula Basin had diets closely
resembling the diet of Tarascan elites from Tzintzuntzan, supporting the ethnohistoric
and archaeological evidence for the presence of Tarascan administrators sent from
Tzintzuntzan. Differences between males and females were not systematically examined
due to inadequate samples, but a preliminary comparison of the data does not show a
difference in diet between the two groups.

This dissertation is the first bioarchaeological study of palaeodiet in the Lake
Patzcuaro Basin, and it is the first step of a long term research agenda designed to
understand human adaptation strategies to highland lake environments in west Mexico.
Future research will aim to improve overall sample size, improve the female to male ratio

in the samples, and to improve the regional distribution of samples.

To my mother, Emestina J. Manrique de Cahue; my late grandmother Luz Alvarez de
Manrique, and my friend Jane E. (Douglas) Kwilecki. At various points in my life, these
women forced me to be strong when I felt afraid, loved me when I could not love myself,

and trusted me when I did not know what I was doing.

ACKNOWLEDGMENTS

“Nothing defines humans better than their willingness to do irrational things in the
pursuit of phenomenally unlikely payoffs.”
-- Scott Adams --

This dissertation and the graduate studies that lead to it, were completed through
the efforts of many people. My guidance committee, friends and family, all supported
me in many different ways, and all encouraged me to persist.

Dr. Norman J. Sauer, served as my advisor and chair of my guidance committee.
To him, I owe thanks for allowing me to pursue my academic interests with relative
independence, and for trusting me to do the right thing. More than that, Dr. Sauer
inspired me to do my very best as a student, a physical anthropologist and an educator. I
conclude my graduate studies under his tutelage with gratitude for his patience and with
the promise of working together as colleagues in the years to come.

I thank Dr. Helen P. Pollard for introducing me to Tarascan prehistory. She
mentored me and provided many opportunities for me to develop as a scholar and form
my own academic relationships with Mexican and North American colleagues. Dr.
Pollard included me as a full participant in her projects as a field worker and allowed me
the freedom and time to pursue and collect my own data during these field seasons.
Beyond this, Dr. Pollard extended me every courtesy due a colleague, and in this way,
she lifted my spirits and my confidence as I made my way through the challenges of
graduate training.

Dr. Ann V. Millard engaged me in stimulating and challenging conversations

about theoretical issues in biological anthropology. She never let me forget that first, and

foremost, I was an anthropologist. During the early years of graduate study, I benefited
from conversations with other faculty, in particular with Dr. Scott Whiteford, Director at
the Center for Latin American and Caribbean Studies. Through these conversations, Dr.
Whiteford challenged me to explain how skeletal analyses of archaeological populations
could contribute to our understanding of contemporary populations. I bring up these
conversations now because they were an important topic of discussion between Dr.
Millard and myself. It was in the context of this ongoing debate that I came to receive
from Dr. Millard, manuscripts of the papers presented at the Wenner—Gren International
Symposium no. 115. “Political-Economic Perspectives in Biological Anthropology:
Building a Biocultural Synthesis,” held in Cabo San Lucas, Baja California Sur, Mexico
in November of 1992. These readings have shaped my theoretical perspective and have
defined my views of biorachaeology. Theoretical perspectives, however, must be
amenable to scientific testing, and no one was more helpful to me during the preparation
of my dissertation proposal than Dr. William A. Lovis.

Dr. Lovis was generous with his time and demanding of my performance. During
my talks with Dr. Lovis I experienced the joy of discovering answers to my questions and
the satisfaction of working through a problem on my own. I will remember with
fondness, the genuine interest he took in the analytical problems I brought to his door.

Finally, but certainly not least, I am indebted to Dr. Mark R. Schurr, fiom the
Department of Anthropology at the University of Notre Dame for making his lab and
himself available to me during the final stages of my research. This project would not

have been possible without Dr. Schurr’s assistance in reading drafts and generously

vi

giving of his time to discuss biogeochemistry with me, filling an intellectual void that
would prove to be the key to a successful project and conclusion of my graduate studies.

I am forever indebted to these five remarkable individuals. Their respect for my
work and ideas, and the generous way in which they shared their work with me stand as
testimony to their commitment to graduate education.

A special note of appreciation to Dr. Thomas Vogel of the Department of
Geological Sciences is due. Dr. Vogel agreed to read my dissertation on a short notice
and served as the College of Social Science’s outside reader. For his friendship, support
and counsel, I am most grateful.

Lic. Rubén Cabrera Castro, archaeologist and curator, Teotihuacan, INAH, Dr.
Gabriela Urufiuela, Chair, Department of Anthropology, Universidad de Las Americas,
and Lic. Salvador Pulido Mendez, archaeologist, INAH Salvamento Arqueolégico
provided samples for this study. I would also like to extend my appreciation to Lie.
Enrique Serrano, Director of the Department of Physical Anthropology, INAH, for
assisting me in securing the samples for this study and in obtaining the proper export
permits.

Ms. Maria Teresa Tersigni and Ms. Martha LaClave assisted in sample
processing. Dr. Jay Siege] from the Department of Criminal Justice and his students
provided a place to work in the Forensic Chemistry Laboratory. Mr. Dennis Birdsell
from the University of Notre Dame provided invaluable assistance with mass
spectrometry. I would like to express my deepest appreciation for the hard work of all of

Dr. Pollard’s students and field workers, in particular, Dr. Christopher T. Fisher.

vii

Intellectually, I have benefited from discussions with Aida Castillej a, Dr. Ann
Grauer, Dr. Denice Hodges, Dr. Bernice Kaplan, Dr. Gabriel Lasker, Dr. Leonard
Lieberman, Dr. Gregory Pereira, Dr. Shirley Schermer, Dr. Margaret Schoeninger, Dr.
Maria 0. Smith, Dr. Joseph Spielberg, Dr. Christine White, and Dr. Scott Whiteford.

I am also grateful to my family. My mother supported me even when it was not
clear to her what, exactly, I was doing. My brothers Alex, and Tony, and my sister Linda
never asked the question “when are you going to get done?” My brother Arturo, built the
amazing ArturoMegatron computer for me, so that I could write my dissertation.

My aunts and uncles in Mexico played a crucial role in the completion of this
project. When I was in Mexico City, they opened their homes to me and made countless
trips to bus stations all around the city to welcome me, or see me off. My aunt Luz Maria
Manrique Alvarez housed and fed me for several weeks at a time, and saw me come and
go in fashions to which she was not accustomed. She worried about me, to be sure, but
put her worry aside and kept a place for me at the dinner table. My Late grandfather José
Refugio Manrique Alvarez loved to listen to my stories about our adventures in the field.
We often had conversations about politics and the revolution in Michoacan. He did not
live to see me graduate, but he had the fortitude of spirit to sit with me several months
before his death and say goodbye. It was not easy for either of us, but we set our hearts
to rest in the knowledge that all would be well. I thank him for that gift.

To my many friends, Gail Barricklow, Laura Delind and her family, Katie Egan,
Isidore Flores, Vandana Goswami, Margaret Graham, Emily Holly, Tae-Sun Kim , Jung-
Hyun Lee, Daurice Lewis, Brent and Eileen McCusker, Colin G. McKinnon, Russell

Skowronek, Beverely Smith, and Heather van Wonner I owe so much. Many of them

viii

helped me financially and opened their hearts and homes to me when I did not have a
place to stay.

To Peter Smith, Aida Castilleja and their family I am grateful for their friendship
their stimulating conversation, and for sharing their knowledge on the Lake Patzcuaro
Basin. To the Rosenthal family for opening the door to their home and making us feel
welcome. Finally, I am grateful to the people of Urichu, and in particular our friends and
colleagues in the field: Alberto, Agustin, Anselmo, Arcelia, Arturo, Delfina, Esperanza,
F rancisca, Guadalupe, J osé, Lourdes, Nacho, Pedro, Salvador, Victor and their families.
This dissertation would not have been possible without their trust, work, and dedication

to our research.

TABLE OF CONTENTS

LIST OF TABLES ............................................................................... ix
LIST OF FIGURES ............................................................................... x
LIST OF ABBREVIATIONS ................................................................. xiii
CHAPTER 1: Introduction .................................................................... 1
CHAPTER 2: Palaeodiet Reconstruction ..................................................... 11
CHAPTER 3: Food Availability, Population Growth and the Development
of the Tarascan State ........................................................... 28
CHAPTER 4: Hypotheses, Methods and Materials ........................................ 54
CHAPTER 5: Results. ............................... ‘ ......................................... 88
CHAPTER 6: Discussion, Conclusions and Future Research ............................. 97
APPENDICES .................................................................................. 110
Appendix A: ........................................................................... 111
Appendix B: ............................................................................ 1 16
Appendix C: ............................................................................ 127
BIBLIOGRAPHY .............................................................................. 137

LIST OF TABLES

TEXT

Table 1. Stable Carbon and Nitrogen Isotope Values for Food Items in the Tarascan
Diet

Table 2 . Lower Calorie Diet

Table 3 Frequency of Individuals Affected with Skeletal and Dental Indicators of
Health

Table 4 Mean 8‘3 C and SISN for Males and Females in Pre-Tarascan and Tarascan
Sites.

APPENDICES

Table A1 Samples Included in the Study

Table Bl Example of Mann-Whitney U Test Worksheet

Table B2 Example of Mann-Whitney U Test Worksheet - Z Approximation

Table B3 Summary of Mann-Whitney U Tests for Mean 8'3Ccou Values (Two-
Tailed, P=0.05)

Table B4 Summary of Mann-Whitney U Tests for Mean S‘SNco" Values (Two-
Tailed, P=0.05)

Table C1 Collagen Extraction Yield, C/N and Stable Carbon and Nitrogen Isotope
Values

Table C2 Replicate C/N and Stable Carbon and Nitrogen IsotOpe Values

xi

Figure 1

Figure 2.

Figure 3.

Figure 4.

Figure 5.

Figure 6.

Figure 7.

Figure 8.

Figure 9.

Figure 10.

Figure 11

Figure 12.
Figure 13.
Figure 14.

Figure 15.

LIST OF FIGURES

Map of the Tarascan state territory showing location of sites mentioned in
the text (modified from Pollard, 1993).

Map of Lake Patzcuaro showing the amount of Class I and 11 land exposed
when the lake level is low (modified from Pollard, 1982).

Map of Lake Patzcuaro showing the amount of Class I and 11 land under
water when the lake level is high (modified from Pollard, 1982).

Map of Lake Patzcuaro Basin showing location of Tzintzuntzan, Urichu
and Tocuaro (modified from Pollard and Gorenstein, 1983).

Main platform in the archaeological zone of Tzintzuntzan, and excavated
structures. Most burials were excavated from block Y3/PNW (modified
fi'om Pollard, 1993).

Excavation block 1 (Y 3/PNW) with intact burials recovered during the
10'h field season.

Map of the site of Urichu showing areas where the intact burials included
in this study were recovered (modified from Pollard and Cahue, 1999).

Map of Urichu, Area 1 showing location of burials (modified from Pollard
and Cahue, 1999).

Map of Urichu, Area 5 showing location of burials (modified from Pollard
and Cahue, 1999).

Tocuaro site. Burial area showing the location of burials in the study
(modified from Araiza and Dominguez with permission from Salvador
Pulido, Salvamento Arqueolégico, INAH).

Collagen extraction yields and C/N values for all treated samples in the
study.

5” C and 8'5N values for Pre-Tarascan and Tarascan elites.
813C and S'SN values for elites in Pre-Tarascan and Tarascan sites.
Mean 513C and 815N values for Pre-Tarascan and Tarascan females.

Mean 613C and 815N values for Pre-Tarascan and Tarascan males.

xii

LIST OF ABBREVIATIONS

%o ............................................................................................................................. per mil
613C ..................................................... Stable isotope ratio of 13C/ 12C relative to a standard
5'5N .................................................... Stable isotope ratio of lSN/MN relative to a standard
A.D. ................................................................................................................. Anno Domini
AMA .................................................................. Amacueca sample from the site of Atoyac
B.C. .................................................................................................................. Before Christ
M ................................................................................................................................... Male
F ................................................................................................................................. Female
SAY .......................................................................... Sayula sample from the site of Atoyac
TOC .......................................................................................................................... Tocuaro
TZIN ................................................................................................................. Tzintzuntzan
TZIN-Os .............................................................................................. Tzintzuntzan ossuary

xiii

Chapter 1

INTRODUCTION AND THEORETICAL BACKGROUND

This study examines the effects of environmental change and sociopolitical power
on the diet of elites in the Tarascan state. The Tarascan state emerged during the
Postclassic (AD. 900 — AD. 1522) in the Lake Patzcuaro Basin (Figure 1); a time when
lake level fluctuations resulted in shifts in the availability of food resources. A
comparison of food availability, population estimates and consumption needs indicate
that by the Late Postclassic (AD. 1350 -— AD. 1522), the basin was unable to produce .
enough food to feed its population (Pollard, 1993). How did the Tarascan Empire, the
second largest in Mesoamerica, and one that dominated western Mexico at the time of
European contact, develop under these constraints?

This dissertation proposes that an emergent, socially stratified elite class in the
Lake Patzcuaro basin used its sociopolitical and economic power to obtain maize from
outside the basin, and protected themselves against the effects of dietary change and the
potential stress of calorie-energy malnutrition. This study measures dietary change and
uses it as an indicator of the ability of elites to use their economic and political power to
cope with fluctuations in food supplies.

Food, Power and Diet in Anthropology

The variety and range of human dietary patterns stand as testimony to the power
of human societies to use culture to overcome environmental and social constraints on
food supplies. Humans adapt to low energy or caloric intakes by conserving energy

through a reduction in body weight, reduced activity levels, and lowering of metabolic

.Amaa .eazom ace
commend 33m £5 E vows—2: moxm 08 mo dogwoo— ofi 336nm 88m 5088B 2: .«o EBEB 05 me 922 A onE

 

2.85:0 cmmoO nausea

050mm 2; .8. _ o.
_

 

3km ”3.0m

   

\I

    
 

 

N £9253.“
ZVUWVEQ
eseawuek 33 \Easoeh
gauging—NH. IV Q00 Ar! .592:
e343.» e3 egee-3.2.3. .
S
/i ._ .. x ._ L
‘{ sash. 23 009.2
gee) «35.5 .33
7%
«0
063.5530

\

\\

 

 

 

rates (Waterlow, 1990). When food surpluses are available and their future availability is
uncertain, food can be consumed and stored in the body as excess fat, for example
(Garine and Harrison, 1988; Little et al., 1990).

As effective as biology is in averting the consequences of food insufficiency,
culture and technology play equally important roles (Douglas, 1975). Freezing, drying,
canning and chilling technologies prevent the spoilage and contamination of foods, and
help alleviate the effects of fluctuating food supplies. Once food is preserved, it can be
stored for future use during periods of shortage. Raised fields, irrigation canals, sprinkler
systems, fertilizers, and more recently, genetic manipulation, are all examples of
technologies developed for increasing food production. When food supplies are
uncertain, many human societies cope by altering their behavior related to food
distribution and consumption. During the dry season, the Ngisonyoka Turkana adults
will undergo voluntary fasting for days in order to ensure that children are well fed (Little
et at., 1990).

Biological and technological coping strategies are important determinants of diet,
but studies of human adaptation that ignore the social, political and economic context in
which it takes place can be misleading in their findings (de Garine and Harrison, 1988).
Furthermore, such studies fail to consider the social relations of power that are involved
in food production and distribution, and consequently, their effect on human adaptation
(Caplan, 1997). Anthropological studies with a focus on food as a source of power and
prestige (Goody, 1982; Mintz, 1996) have shown that from prehistoric times to the
present, food has played a central role in the development of socially complex and

stratified societies. Due to the importance of food for survival and its social, political and

economic role, changes in dietary patterns throughout the course of human evolution are
critical to the study of human adaptation (Schoeninger and Moore, 1992).

Archaeology and bioarchaeology have an extraordinary capacity for evaluating
human dietary patterns over large units of space and time (Saitta, 1998), and over the last
twenty years, dietary reconstruction has become an important part of archaeological
investigations. Knowledge of dietary intakes has helped in the evaluation of the nutritive
effects of increased consumption of maize in the Nashville Basin (Buikstra et al., 1988)
and the relationship between social complexity and agricultural intensification in the
Ohio valley (Schurr and Schoeninger, 1995). Also examined has been the restricted
access to foods to members of different social groups (sex, age, class, occupation
religious status, etc.) in socially complex societies (Huelsbeck, 1988; White and
Schwarcz, 1989).

Differential access to food will have an effect on the individual’s nutritional status
and ability to meet dietary needs, compromising his or her capacity to cope with
infectious diseases, and ultimately affecting health (Larsen, 1997). Undoubtedly, this
relationship is equally true for contemporary and ancient human populations, and it is
examined through direct observation of human behavior or the study of human skeletal
remains. The study of health in prehistory using human skeletal remains is the best
means available for identifying patterns of differential access to resources (for example,
food) between and within human groups (Saitta, 1998).

As long as the analytical focus is on issues that are amenable to archaeological
investigation, archaeology can provide a long-term perspective on the political dimension

of food, diet and health (Dietler, 1996; Saitta, 1998). An archaeological investigation of

the role of food in the articulation and manipulation of status and power can move our
understanding of ancient societies beyond the traditional studies of what people ate
(Dietler, 1996; Hayden, 1996). While human diets must provide adequate nutrition for a
population to survive, what people eat is more often a function of what they are allowed
to eat, or what they are required to eat to ensure that the economic and political priorities
of ruling classes and corporate elites are met (Harris and Ross, 1987).

This study uses a political economic approach to bioarchaeology because of its
ability to examine the social and power relationships that constrain individual strategies
to gain access to resources, including food (Goodman and Leathennan, 1999). The
dissertation largely follows the theoretical framework outlined by the work of those
anthropologists who convened for the Wenner-Gren International Symposium no. 115,
“Political economic Perspectives in Biological Anthropology: Building a Biocultural
Synthesis,” held in Cabo San Lucas, Baja California Sur, Mexico in November of 1992
(Goodman and Leatherman, 1998a).

Political Economy in Bioarchaeology

Bioarchaeology, primarily concerned with how skeletal and dental tissues from
archaeological contexts reveal life histories of individuals and groups, has its theoretical
roots in the theoretical developments of its parent discipline, archaeology (Larsen, 1997).
Early studies of skeletal remains focused on descriptions of skeletal morphology and
pathologies. These studies, diagnostic in nature, were mostly concerned with
documenting case studies of individual diseases (Ortner and Putschar, 1981). Until the
19703, when larger skeletal collections became available for study, patterns in disease

prevalence and morphology were not investigated (Larsen, 1997). These large-scale

projects were conducted at a time when the discipline of anthropology in the United
States was driven by cultural-ecological approaches to the way people adapted to
changing environments (Little, 1995; Thomas, 1998). The concept of adaptation has
considerable power as a theoretical tool to explain spatial and temporal variation in
human biology and behavior. In spite of its extraordinary capacity for evaluating
evolutionary change, it is not the only explanatory paradigm (Gould and Lewontin,
1979). Also problematic is the continued disagreement among evolutionary biologists
and anthropologists over what constitutes the unit of adaptation (i.e., is it individuals,
families, societies, populations, or ecosystems?) (Little, 1995). Its limitations
notwithstanding, the concept of adaptation played a central role in the development of
cultural ecology. This ecological perspective became a unifying force for
anthropologists, as they sought to understand humans in their biological and cultural
contexts (Baker, 1996). Studies with an ecological perspective examined the way in
which humans manipulated their cultural and biological worlds to extract resources from
their environments. The studies were, without a doubt, materialist and ecological, but
their biocultural approach linked them to larger theoretical issues (Goodman and
Leathennan, 1998). While ecological perspectives can evaluate how well people respond
to adverse conditions, and what people do when their primary response systems are
disrupted, they fail to examine why those adverse conditions exist, what are the processes
in which they originate, and what dynamics keep them in place (Thomas, 1998).

In reaction to the cultural ecology of the 1960s, evolutionary biologists and
sociocultural anthropologists addressed several important theoretical concerns.

Evolutionary biologists criticized ecological approaches for their underlying assumption

that adaptation was purposeful and progressive (Gould and Lewontin, 1979), suggesting
instead, that evolution was the random product of historical contingencies (Gould, 1991).
Within sociocultural anthropology the use of cultural ecology and modernization theory
as theoretical frameworks were also under scrutiny. While cultural ecology was
criticized because of its functionalist undertones and its assumption that human societies
were closed, homeostatic and self-regulating systems (Ortner, 1984; Wolf, 1982),
modernization theory was seen as progressive evolutionism (Goodman and Leatherman,
1998; Ortner, 1984). Thus, throughout the 1970s anthropologically informed political
economic theories emerged in ersponse to approaches that failed to consider the centrality
of human agents in social systems (Giddens, 1987).

Anthropological political economy is concerned with how micro- and macro-level
systems and histories intersect to create the social context in which human action takes
place. This approach is concerned with the social relations and institutions that control
fundamental resources (including labor), and it views this control as an expression of
power (Goodman and Leathennan, 1998; Roseberry, 1988). Power relations are formed
within specific social fields (Roseberry, 1998). These social fields of power are web-like
relationships that structure the appropriateness, availability and permissibility of
responses to environmental (socio-political, economic or biological) change. The content
of the relationships that form social fields is “power: over who owns what, who works
where, for how long, and with what return” (Roseberry, 1998). Political economic
perspectives consider the context and the individual’s position within these social webs as
factors that influence social action, providing an explanatory paradigm for human bio-

behavioral variation (Roseberry, 1998).

The problem of food acquisition and consumption, and the ability to cope with it,
are constrained by the availability of environmental and social resources (Goodman and
Leatherman, 1998). A political economic perspective is particularly useful to biological
anthropologists because it directs our attention to these problems. Themes that emerge
from political economic bioarchaeology include the macro-level systems of natural and
social resources, and the historical contingency of the constant reshaping of those
enviromnents to which humans must adapt (Goodman and Leathennan, 1998; Roseberry,
1989)

The emergence of state societies is associated with the emergence of social,
political, and economic differences within and between social groups. Regional
differential access to power is the organizing force around which individual actors must
negotiate their social positions. A political economic bioarchaeology is uniquely suited
to examine the biological consequences of this differential access to power (Martin,
1998; Saitta, 1998).

The difficulty of inferring social inequality from archaeological evidence,
however, has prompted archaeologists to gather evidence from a variety of disciplines.
Multidisciplinary approaches to studies of ancient societies have the ability to address the
complexity of interactions between various lines of evidence, and facilitate the
examination of food production and distribution, and its role in the acquisition of status
and power. This dissertation uses a multidisciplinary approach, drawing information
primarily from human skeletal biology, ethnohistory, archaeology, and biogeochemistry.

Chapter Two reviews the literature on palaeodiet reconstruction techniques

available to archaeologists and physical anthropologists to reconstruct the diets of ancient

populations. The focus of the chapter is on the method of stable isotope mass
spectrometry. It makes the argument that this method is best suited for fragmented and
incomplete collections such as those found throughout west Mexico. The chapter
highlights the ability of the method to provide dietary information at the individual level
of analysis in studies of food and power.

Chapter Three is a review of Tarascan archaeology and ethnohistory. This
chapter focuses on recent developments in our understanding of Tarascan prehistory and
the cultural developments that lead to the formation of the Tarascan State. The primary
purpose of the chapter is to provide a context for the archeological problem under
investigation. To this end, the chapter outlines the various lines of evidence, drawn from
a number of disciplines, used to organize what we know about the Tarascans and
conceptualize what we still wish to learn.

Chapter Four has four sections: expectations, materials, hypotheses, and
methods. The expectations section presents a brief discussion of the arguments used to
construct each hypothesis tested in the study. The materials section presents an overview
of the skeletal collections used in the study. A description of those collections included
in this study includes: mortuary context, preservation, and completeness. Age and sex
determinations are also discussed. The hypothesis section presents formal null and
alternative hypothetical statements and the criteria used to reject or accept the null
hypotheses.

The methods section includes complete descriptions of sampling criteria, and
laboratory protocols for collagen extraction, assessment of indigeneity and mass

spectrometric measurements. Included in this section is a discussion of the statistical

methods used to analyze the data. This section makes an argument for the suitability of
non-parametric statistical techniques to the samples in the study.

Chapter Five presents aggregated and disaggregated data, along with the results
of the statistical tests.

Chapter Six discusses the results and offers plausible explanations for the
patterns of variation and distribution displayed by the data. The chapter concludes the
dissertation with a discussion of the limitations and strengths of the study. It also
proposes future research areas to address the limitations of the study, and discusses its
contribution to anthropology in general and to west Mexican archaeology, Mesoamerican

studies, and Mexican physical anthropology in particular.

10

Chapter 2

PALAEODIETARY RECONSTRUCTION

Various methods and techniques from a wide range of disciplines are available to
archaeologists for the reconstruction of ancient diets. These diverse methods provide
information on a large or small scale; and the units of analysis range in scale from groups
such as populations and species to individuals (Schoeninger and Moore, 1992; Schwarcz
and Schoeninger, 1991).

Optimal foraging models, site catchment analysis, and the analysis of floral
(I-Iastorf and Popper, 1988), faunal (Jackson and Scott, 1995), ceramic and stone tool
(Blitz, 1993; Jones, 1996) remains provide information about groups. These analytical
approaches have proven invaluable to archaeologists because they help identify the food
items in the diet, although they are unsuitable for the assessment of the relative
quantitative importance of individual food items in the diet (Schwarcz and Schoeninger,
1991). Foods incorporated into the archaeological record, whether deliberately or
unintentionally, do not often directly reflect the systematic use of foods. Middens, for
example, are imperfect samples of ancient human behavior because plant and animal
remains undergo different taphonomic processes that result in their biased representation
in the archaeological record (Schoeninger and Moore, 1992). In spite of these problems,
assemblages of plant remains have been useful in elucidating patterns of change in plant
use (J ohannsen, 1988). Cut marks on animal bones recovered from middens have

provided information on butchering and processing practices (Marshall, 1986), as well as

11

the preferential utilization of specific parts of the animal with regard to nutritional
consequences (Speth, 1983).

Tools tell us more about the processing of foods than they do about the
consumption of foods. For example, food residues on the internal surfaces of cooking
vessels may help in the identification of the foods that were processed, but cannot help
determine whether they were consumed. F urtherrnore, the analysis of many vessels
would be necessary to recognize the dietary importance of different foods, since
individual ceramic sherds represent a very small sample of behavior (Schoeninger and
Moore, 1992).

Dietary analysis aims to identify the foods and their relative proportions in the
diet. Qualitative reconstructions of diet reach their greatest power for understanding
prehistoric subsistence systems when combined with quantitative methods that provide
dietary information for individuals rather than groups (Schoeninger and Moore, 1992;
Schwarcz and Schoeninger, 1991).

Analytical methods that provide information about individuals include studies of
stomach content, fecal remains, and human skeletal remains. Skeletal remains studied by
physical anthropologists can provide information about diet and its effect on health; these
studies are useful in determining the variety of foods that were available for consumption
to people from different socio-economic strata, and the differential effects that their
consumption had on the health of individuals. The analysis of microwear patterns on
teeth, palaeopathology, and histomorphology of bone tissue gleans dietary information
fi'om the human skeleton (Larsen, 1997). Microwear analysis of teeth helps identify

foods processed in the mouth over several months preceding death. For example, the

12

disappearance of the high polish on teeth, left by vegetable fibers, is associated with the
introduction of maize agriculture (Rose and Harmon, 1986).

Physical anthropologists have detailed increases in the frequency of carious
lesions in prehistoric hunter-gatherers and maize agriculturalists. Overall, there is a clear
tendency for prehistoric maize agriculturalists to have higher frequencies of carious
lesions than do prehistoric foragers (Larsen, et al., 1991). Among maize agriculturalists,
however, the association between increased reliance on maize and carious lesions is
variable. Furthermore, it appears that relatively small changes in diet and food
processing techniques can result in large differences in the prevalence of carious lesions
(Larsen, et al., 1991).

Palaeopathology, developmental defects in bone and enamel, and
micromorphology of bone cross sections help detemrine the nutritional adequacy of diet
(Martin, et al., 1985). For example, White and Armelagos (1997) examined the
micromorphology of femoral transverse sections of Nubian mummies to evaluate the
relationship between osteopaenia and diet. They found that osteoporotic individuals had
elevated nitrogen isotope ratios, suggesting water stress in an arid environment, or protein
stress.

The combination of these methods and techniques plays a crucial role in the
determination of what individuals ate, and the effect it had on their health. Studies of the
actual foods consumed (and the relative amounts in which they were eaten), however,
were not possible until the mid-1970s when methods for the chemical analysis of bone

were first applied to anthropological problems. These new technological developments

13

made it possible for anthropologists to measure diets and evaluate the relationship
between diet, health and social status.

The application of stable isotope analysis to the reconstruction of prehistoric diets
is based on the premise that the isotopic composition of an animal's tissues is similar to,
or deviates by a consistent amount from, that of its diet (AmbroSe, 1993; Chishohn, 1989;
Katzenberg, 1997; Schwarcz and Schoeninger, 1991; van der Merwe, 1982; Vogel and
van der Merwe, 1977). This assumption has been validated in controlled feeding
experiment with laboratory animals (DeNiro and Epstein 1978). Today, the variation in
the carbon and nitrogen stable isotope ratios of bone collagen aids in the reconstruction of
prehistoric human diets and a variety of anthropological problems (Schwarcz and
Schoeninger, 1991; Katzenberg, 1992). Various studies have provided information on
the type of foods consumed, the proportions of different food items in the diet, and

variation in diet through time and space (Schwarcz, et a1. 1985; Schwarcz, 1991).

Principles of Stable Isotope Analysis

IsotOpes are atoms of chemical elements with the same number of protons but
different number of neutrons, resulting in different atomic masses. The isotopes of an
element have the same chemical properties, but differ in their rates of reaction, with the
lighter isotopes generally reacting more readily than the heavier isotope (Ambrose, 1993;
Katzenberg, 1992; Schwarcz and Schoeninger, 1991). Isotopes that decay through time
into another element are radioactive and can be used for radiometric dating; stable
isotopes do not transmute into other elements. The stable isotopes of carbon, nitrogen

and oxygen are the most utilized in palaeoclimatic and palaeodietary studies. Stable

14

carbon and nitrogen isotope analysis involves the measurement of the ratio between two
isotopes-(”C/IZC and lsN/MN, for example). Because the concentrations of '3C and 15N
are very low in the biosphere, the ratios '3 C/ 12C and lsN/MN cannot be measured directly.
Rather, the ratios are expressed as:

5% or 6‘5N = [(Rm../R,W) - 1] x 1000
where R is 13C/ 12C or 15N/MN for 613C or 615N, respectively. The internationally
recognized standard for carbon is the marine carbonate PeeDee Belemnite (PDB) and for

nitrogen the international standard is atmospheric Nair.

Carbon

Smith and Epstein (1971) documented a difference in 613C values between C3
plants (those utilizing the Calvin-Benson photosynthetic pathway) and C4 plants (those
that follow the Hatch-Slack photosynthetic pathway). Maize belongs to the C4 plant
group, which has 813C values in the range of -12.5%o to -l3.1%o for modern maize (Smith
and Epstein, 1971; Boutton, 1991; Tieszen 1991). The average 513 C value for modern C3
plants is -26.5%o (Smith and Epstein, 1971; Boutton, 1991; Tieszen 1991). Plants derive
their carbon from atmospheric C02, and any changes in the isotopic composition of
atmospheric CO2 will affect the isotopic composition of plants. The burning of fossil
fuels (coal, petroletun, and natural gas) has resulted in a decrease 813 C value of

atmospheric CO2 of approximately 1.4%0 (van der Merwe, 1989).

Fossil fuels are strongly depleted in ‘3 C because they are, primarily, the products
of decomposition of C3 plants. The 6‘3 C values of coal range between —20.0 and —

13.0%0; petroletun ranges between —27.0 and —30.0%o. The range of 513C values of

15

natural gas depends on the molecular composition and origin of the hydrocarbon gases.
Natural gas is mostly formed of methane (CH4) (F aure, 1998). The 613 C values of
methane formed through anaerobic decomposition of cellulose range to about —80.0%o;
and methane formed by the breakdown of larger hydrocarbon molecules ranges in 613 C
value from -—25.0 to —50.0%o (Faure, 1998). Ethane (C2H6), propane (C3Hg) and butane
(CrHlo), also found in natural gas, are less depleted in '3 C than methane (F aure, 1998).

Because of the downward shift in 513C value of atmospheric CO2, prehistoric
plants are more positive than modem plants (Keeling et al., 1979). Measurements of
prehistoric maize show an average 613 C of -9.0%o (Schwarcz, 1991).

Consumers will reflect the isotope value of their diet. Carbon isotopic values of
the diet are recorded in the tissues of terrestrial and marine herbivores, with a secondary
metabolic fractionation that depends on the tissue examined. Feeding experiments on
small animals show diet-tissue differences between 1.0 and 3.0%o; and for larger animals
between 3.0 and 5.0%0 (Schoeninger and Moore, 1992). Although there is considerable
disagreement in the magnitude of the diet-tissue differences measured, the estimate for
the difference in the 813C value of bone collagen and diet is generally accepted to be
approximately 3 - 5%o (Schwarcz and Schoeninger, 1991). That means. for example, that

an animal consuming a diet consisting of 100% C3 plants, with an average 513 C value of

—26%o, will have a 8‘3 C value of bone collagen approximately —20%o.

Nitrogen
Some plants incorporate nitrogen from soil in the form of nitrate or ammonium,

while others do so directly from the atmosphere via nitrogen fixing symbionts

16

(Schoeninger and Moore, 1992). Legumes retain endosymbionts and have 8'5N values
similar to that of atmospheric N2 (096a) (Schwarcz and Schoeninger, 1991). Most non-
legumes rely on soil ammonium or nitrate that typically have values greater than that of
atmospheric N2 (Hoefs, 1987). Thus, legumes have 8'5N values that are typically lower
than non-legumes (Schwarcz and Schoeninger, 1991).

As in the case of carbon, the magnitude of difference between the 8'5N of bone
collagen and diet is somewhat uncertain. Animal feeding experiments indicate that the
8'5N of bone collagen is approximately 3960 higher than that of diet (DeNiro and Epstein,
1981; Hare et al., 1991). Researchers have reported S'SN values of bone collagen that are
higher than expected, however, indicating that 6'5N values may be affected by factors
other than diet (Schoeninger and Moore, 1992). Water stress, protein restriction, and
high protein diets, for example, have all been implicated in elevated 5'5N values of bone
collagen (Ambrose and DeNiro, 1986; Heaton et al., 1986; Sealy et al., 1987).

In addition to the diet-tissue spacing in 615N, an upward shift of 3%o in SISN
values also occurs with each trophic level (DeNiro and Epstein, 1981). Trophic level
shifts, therefore, must be accounted for when the relative contribution of different dietary

sources to a consumer are estimated (Schwarcz, 1991; Little and Schoeninger, 1995).

Isotopic routing efi'ects on bone collagen and apatite carbonate
Diet-tissue fractionation factors are not constant, and can vary as a firnction of
proportion and quality of protein in the diet (Ambrose, 1993; Schwarcz and Schoeninger,

1991; Parkington, 1991). Some dietary macronutrients may be preferentially “routed”

17

into particular tissue without homogenizing with the rest of the body’s atoms (carbon and
nitrogen atoms, for example) (Schwarcz, 1991).

Controlled-diet studies on animals indicate that carbon from carbohydrates and
lipids is preferentially routed to bone apatite carbonate, while that from protein seems to
be incorporated into bone collagen (Ambrose, 1993; Ambrose and Norr, 1993; Krueger
and Sullivan, 1984; Lee-Thorpe, et al., 1991; Parkington, 1991). This suggests that 513C
from bone apatite carbonate measures the isotopic signature of the total diet, while that of
collagen measures the isotopic signature of the dietary protein sources (Ambrose, 1993;
Schwarcz, 1991). A comparison of the isotopic composition of collagen to that of bone
apatite carbonate can detect this type of routing for carbon (Lee-Thorp and van der
Merwe, 1988).

In spite of the convincing evidence for routing of amino acids to protein synthesis,
the assumption that the body utilizes dietary protein preferentially during collagen
synthesis may be untenable under conditions of low protein intake (Ambrose and Norr,
1993; Tieszen and F agre, 1993). Schwarcz (in press) has shown that the results of these
studies reflect an inhibition of endogenous synthesis of amino acids when the amino acid
level in the extra-cellular fluid is already at, or above, a critical point. This inhibition is
more likely present in animals (or humans) with adequate amounts of protein in their
diets. The protein is broken down into its respective amino acids in the liver. These
amino acids are made available to the cell for protein synthesis (Schwarcz, in press).

Data from isotopic palaeodiet studies of maize consumers seem to confirm this
point. In a study of maize-consuming prehistoric populations in Eastern North America,

Buikstra and colleagues (1988) found high 8'3 C bone collagen values, indicating that

18

most of the carbon in the collagen came from maize. To produce these high 6'3 C values,
collagen must have incorporated carbon from carbohydrates because maize is relatively
low in protein and lacks sufficient levels of some amino acids (Ambrose, 1993). At
Maya sites from Belize, some individuals have 6'3C bone collagen values > -10%o. If a
linear mixing model (all atoms mix together to form a common pool from which the body
can then draw necessary atoms for tissue synthesis) is assumed, these values represent a
diet with up to 70% maize (Schwarcz, 1991). About 80% of carbon atoms come fi'om
dietary protein and not from carbohydrates (Ambrose and Norr, 1993). Since some of the
dietary protein in the diet of Belize Maya was derived from the flesh of herbivores like
peccaries and turkeys, the calculation of the percent C4 (PC4) based on a linear mixing
model, would underestimate the proportion of maize in the diet. In contrast, a routing
model would estimate that a collagen with a 8 '3 C = ~10%o represents a diet composed of
over 100% maize (Schwarcz, in press). At Maya sites in Belize, the PC4 (70%), based on
a linear mixing model, is already close to the upper limit of nutritional acceptability (no
values greater than 75% maize in the diet have been recorded). Therefore, for
populations with a high dependence of maize, a routing model is unacceptable
(Schwarcz, in press).

A large portion of the animals consumed by the Maya (for example dogs and
turkeys) ate diets consisting of some C4 foods (maize) (Cannon, et al., 1999).
Mesoamerican populations undoubtedly consumed C4-based meat and there is evidence
that they also consumed some C3-based meat, suggesting that a routing model may

overestimate C4 consumption levels (Schwarcz, in press).

19

Still, in diets with adequate levels of animal protein for collagen synthesis, carbon
from carbohydrates and lipids are under-represented in collagen (Ambrose, 1993).
Because endogenous amino acid synthesis is suppressed (inhibited) in the presence of
relatively high concentrations of the respective amino acid in the extra-cellular fluid, we
might expect a strong inhibition effect (routing) in populations which consumed high
levels of animal protein (Schwarcz, in press). In contrast, populations with dietary
protein generally in low supply or largely derived from plants would approach a linear
mixing model because the carbon atoms of all foods (except lipids) would be randomly
assigned to the synthesis of all tissues (Schwarcz, in press).

The biochemistry of collagen synthesis notwithstanding, the isotopic fractionation
between apatite and total diet is somewhat variable (Ambrose, 1993); making it difficult
to use 5'3 C of apatite as an index of total dietary 813C values (Schwarcz, in press).
Furthermore, biogenic 5'3 C signals of archaeological bone, in particular apatite, may be
altered by exchange of carbonate (C03 '2 ) with ground water or soil water after burial, or
by dissolution and recrystallization of bone mineral. In spite of the susceptibility to
diagenetic alteration of bone apatite, few researchers systematically evaluate the apatite
composition prior to 8'3C studies (Wright and Schwarcz, 1996). Infrared and isotopic
evidence at the Maya site of Dos Pilas in Guatemala were used to evaluate the diagenesis
of bone apatite (Wright and Schwarcz, 1996). The results of this study show that even
relatively recent bone may be diagenetically altered. Diagenetically altered bone can be
treated with acetic acid to remove calcium carbonate from bone interstices, but the
removal of carbonate contamination within the apatite mineral remains a serious problem

(Koch et al., 1990; Sillen, 1986).

20

Where diagenetic alteration has not been a factor, the isotopic analysis of bone
apatite has not been of significant use. In a study of five New England populations,
Bourque and Krueger (1994) compared dietary reconstructions based on faunal remains
and historical documents to those using carbon and nitrogen isotopes of collagen and
carbon isotopes in bone apatite carbonate. The results show a high correlation between
513C values of collagen and apatite carbonate, indicating no gain in dietary information
hour the analysis of apatite (Katzenberg and Harrison, 1997).

The difficulties of detecting diagenetic changes in bone apatite and our limited
knowledge of the role that biochemical processes play in protein synthesis make bone
collagen the bone fraction of preference for stable isotope analysis. In light of these
findings and the fact that the present study is concerned with relative amounts of maize

(more, less, the same), the analysis in this dissertation will be limited to bone collagen.

Assessing the Reliability of Stable Isotopic Ratios in Bone Collagen

Archaeological bone may undergo complex post-mortem diagenetic changes that
result in the chemical alteration of proteins and apatite carbonates (Ambrose, 1991;
DeNiro, 1985; Schwarcz and Schoeninger, 1991). Diagenetic alteration of bone, that is,
the sum of the physical, chemical and biological processes that occur in the postmortem
environment, may result in the addition or removal of organic materials to bone (DeNiro,
1985; Hedges and Law, 1989).

The elemental and isotopic composition of collagen and apatite carbonate may be
altered depending on how much exogenous organic material is added or removed; and on

how different the C/N, l3C/‘ZC and 15N/MN ratios of the added or removed organic

21

material are fi'om those that characterize the indigenous material (DeNiro, 1985).
Exogenous organic materials such as humic acids and firngi can be mechanically or
chemically removed prior to isotopic analysis to protect the fidelity of the dietary signal.
Diagenetic changes have been shown to alter the dietary signal in bone collagen,
therefore, it is essential that diagentic changes be evaluated (DeNiro, 1985).

For collagen extracts, values of the ratio between carbon and nitrogen (C/N)
between 2.7 and 3.6 are consistent with retention of the original isotopic value (DeNiro,
1985; Schoeninger and Moore, 1992). Even well preserved bone will have lost some of
its collagen to heat and hydrolysis. Because of this, collagen yields are a good measure
of the degree of preservation of this protein in bone (DeNiro and Weiner, 1988).
Ambrose (1990) has demonstrated that samples that yield collagen weights of at least 2%

or greater than those of dry bone weights have well preserved isotopic signatures.

The “Menu ” and the “Meal the Reconstruction of Diet

Given our understanding of the S'SN and 813C values of dietary items, maize should
differ from all other food sources based on 8'3 C values. Legumes should have lower
BISN values than other dietary items. Fish, deer, rabbit, turkey, duck are C3 plant
consumers and should have higher nitrogen isotope values than plants, and lower carbon
isotope values than maize (even when a protion of the diet includes maize or other C;
plants). Consequently, legumes, maize, and animal protein sources (fish or meat) should
differ in their isotope values. Accurate knowledge of local isotopic composition of food
items is necessary for precise dietary reconstructions (Ambrose, 1993; Schwarcz, 1991;

Schwarcz and Schoeninger, 1991). A pilot study was conducted to determine isotopic

22

values among food sources known to be important to the Tarascans. Food items included
skeletal remains of fish (Goodea Iuitpoldi), rabbit (Sylvilagusfloridamus), deer
(Odocoileus virginianus), peccary (T ayassu tajacu), turkey (Melagris gallopavo), and
dog (Cannisfamiliaris). All dietary items were recovered fi'om the site of Urichu in the
Lake Patzcuaro Basin, and date to the Pre-Tarascan period (Early Postclassic ~ AD. 900)
(Pollard, 1995). Table 1 shows the 8'3C and SISN values for these dietary items. These
values compare to those obtained for faunal remains from Southern Ontario by
Katzenberg (1989) where dogs had 6'5N values of 96%, possibly due to feeding on
human diet leftovers; turkeys had SISN values of 6.9%, and freshwater fish a value of
3.6% for a bottom feeder. Carnivorous (picivorous) fish in Southern Ontario have a
higher average 515N value (8.8%o) than carnivorous fish from the Lake Patzcuaro Basin
(65960). With the exception of an archaeologically recovered maize cob from Urichu (as

yet unanalyzed) no plant dietary items are available for isotopic analysis.

Previous Applications of Stable Isotope Analysis

Stable isotope data have been applied to a diversity of anthropological
questions. The introduction of agriculture in the New World was marked by the arrival
of the C4 plant maize (Zea mays). The intensification of maize cultivation and its
increased use as a staple food item alter its domestication in North America, have been
subjects of intense anthropological investigation. The relationship of maize agriculture
intensification and the development of chiefdoms have been of particular interest to '
archaeologists (Schurr and Schoeninger, 1995). Likewise, physical anthropologists have

paid considerable attention to this relationship and its effect on the health of North

23

TABLE 1

Stable Carbon and Nitrogen Isotope values for Food Items in the Tarascan Diet

 

I’m-Tarascan species Bone Collagen
(AD. 350 - 1100)

 

 

FAUNA A

Rabbit -19.7 2.6
(Sylvilagusfloridamus)

Deer -20.5 5.1
(Odocoileus virginianus)

Fish -12.9 6.5
(Goodea luitpoldr)

Turkey -12.1 6.7
(Melagris gallopavo)

Peccary -12.5 8.0
(T ayassu tajacu)

Dog -10.0 9.8
(Cannisfamiliaris)

C4 PLANTS

Maize 3 -90 8.7
(Zea maize)

Amaranth C -12.5 11.1
(Amaranthus sp.)

c3 PLANTS

Beans C -26.0 0.0
(Phaseolus vulgaris)

Chenopodium C -26.0 7.4
(Chenopodium sp.)

 

A Measured from Archaeological Bone
13 Schwarcz et al., 1985
c Schoeninger, 1995

24

American prehistoric populations (Buikstra and Milner, Bumsted, 1984; Cohen and
Armelagos, 1984; Katzenberg, et al., 1985; Spielman, et al., 1990).

Isotopic analyses of Mesoamerican and South American populations show a
highly variable pattern of maize consumption. Maize consumption in the Tehuancan
Valley appears to occur far earlier than in other regions of North America, with a shift to
maize by 4000 BC. (DeNiro and Epstein, 1981). In lower Central America, Maize
consumption in Panama between 3000 and 5000 BC. is minimal; and in Costa Rica the
adoption of maize is dated to ca. AD. 300 — 1550 (Norr, 1991).

By far, the Lowland Maya region is the most extensive, isotopically studied
region of Mesomerica (White and Schwarcz, 1993; White, et al., 1993, 1994; Wright and
Schwarcz, 1996; Wright and White, 1996). Stable isotope studies in the region have
changed the picture of Maya dietary ecology. Evidence from Larnanai and Pacbitun in
Belize indicate that the Maya were less reliant on maize than were other Mesoamerican
populations (White, et al., 1993) showing a nonlinear temporal change in the emphasis on
maize (Wright and Schwarcz, 1989).

No stable isotopic studies of ancient diet have been conducted previously in the
West Mexican region. In spite of recent ethnohistoric and archaeological research in
West Mexico, bioarchaeological studies in the region are still few. Recent archaeological
research in west Mexico has challenged culture-historical paradigms that link west
Mexico to other regions, and has produced new data for the interpretation of the human
ecology of highland lakes (Amauld, 1996; Macias-Goytia and Vackimes Serret, 1990;
Pollard, 1997; Valdez , et al., 1996). Unfortunately, bioarchaeological research in west

Mexico has not followed the same trend. With few exceptions (Cahue, 1999; Urur'tuela,

25

1998), the analyses of skeletal material in the region are mostly focused on studies of
patterns of mortuary treatment, cranial and dental alterations, population genetic
histories, and taphonomy (Cahue and Pollard, 1999; Macias Goytia, 1989; Pereira, 1995,
2000; Pickering, et al. 1985). One reason for this pattern is that physical anthropologists
working in the region are not directly interested in human adaptation and biocultural
aspects of human health and behavior. Another reason may be that skeletal materials in
the region are often incomplete and not well preserved, impeding the use of traditional
methods of analysis of skeletal indicators of health. To fill these theoretical and
analytical gaps in bioarchaeological research in west Mexico, a research program to
understand human adaptation to highland lake environments is under development. The
research program will have a strong biocultural focus and will address the interaction
between food availability, infectious diseases, population size, sociopolitical complexity
and economic power, and the effect this interaction has on the health of human
populations in the region. This dissertation is the first contribution toward the
development of this program. It uses carbon and nitrogen isotopes to examine regional
and temporal variations in elite diets during the development of the Tarascan state in the
Lake Patzcuaro Basin.

To summarize, the diet of ancient populations is an important component of
bioarchaeological research. A variety of analytical methods have been developed to
understand subsistence strategies across time and space. These methods have expanded
our knowledge about the types of foods produced, processed and consumed by ancient

populations. Analytical methods for the study of skeletal remains (human and non-

26

human) have produced a significant body of literature that elucidates the various ways in
which dietary practices affect skeletal biology.

This dissertation is concerned with determining the relative amounts of maize
actually consumed by elites in the Lake Patzcuaro Basin and how diet was affected by
environmental change and economic power. This study uses skeletal collections that are
fi'agmentary and incomplete. Stable isotope analysis was selected for this study because
of its ability to glean dietary information from collections consisting of skeletons that
have undergone some diagenetic alteration, are fragmented and incomplete.

In the following chapter the archaeology, ethnohistory and Bioarchaeology of the
Tarascan state are presented as background to the anthropological problem under

investigation in this study.

27

Chapter 3

FOOD AVAILABILITY, POPULATION GROWTH AND THE DEVELOPMENT OF
THE TARASCAN STATE

Two great civilizations emerged in Mesoarnerica during the Postclassic (A. D.
900-1500). These two polities were the Aztecs, who settled in the Basin of Mexico, and
the Tarascans who settled in the Lake Patzcuaro Basin in the modern state of Michoacan,
in west México (Figure 1). At the time of Spanish contact, the Tarascans held political
and economic control of west Mexico, and although they maintained an active military
frontier with the Aztecs, the Tarascans successfully resisted their conquest attempts. The
Tarascans, known among the Aztecs for their military prowess, also impressed the
Spaniards with their economic administration and strong political control over their
territory (Gorenstein and Pollard, 1983).

The process of state expansion resulted in the incorporation of increasingly
diverse communities. This diversity challenged the state’s economic and political power.
To ensure economic exploitation of populations and resources, and to secure the integrity
of its frontiers, the state had to integrate these communities (Pollard, 1993). The sole
purpose of ethnic integration was to serve the interests of the Lake Patzcuaro Basin, in
particular the elite at Tzintzuntzan. The state was divided into three ethnic zones: an
ethnic heartland, an area of active ethnic assimilation, and a zone of ethnic segregation.

The first zone was the heartland (mostly the Lake Patzcuaro Basin), where several
cultural/linguistic groups lived before the formation of the Tarascan state, was
transformed into a zone where Tarascan ethnicity characterized the cultural identity and

Tarascan became the dominant language. Local leaders of administrative centers within

28

this region were appointed by the Tarascan king, and dealt directly with the Tzintzuntzan
elite administrators and were assumed to be loyal to the state (Pollard, 1993).

The second zone, active assimilation, was added to the state after 1440. Many of
the resources used by elites in the Lake Patzcuaro Basin to mark their status came from
this zone. Because of this, elites became increasingly more dependent on this population,
and its loyalty to the state was of great concern to Tzintzuntzan. Loyalty was ensured by
a process of gradual ethnic assimilation; a process that required local populations to
perceive benefits for themselves. The Tarascan state provided local leaders with prestige
and power from direct association and identification with the state’s social structure. In
return for their loyalty, local leaders were elevated to local state administrators
(angdmecha, he who wears a lip plug), marking their position with the ceremonial
insertion of a lip plug during the feast of sicuindiro (RM, 1980). Local claims to
resources and social status, were not legitimized.

The third zone was one of ethnic segregation, and it extended out to the frontiers
of the state. In this area, loyalty to the Tarascan elite in Tzintzuntzan was ensured
through the provision of military protection. Administrators in this zone were elites sent
in from the state’s geopolitical core; ambassadors of sorts, that articulated with the local
population to secure resources and make sure the community remained loyal. A set of
burials at the site of Atoyac, in the Sayula Basin, which included Tarascan burial goods
(polychrome ceramics, bronze metal objects, obsidian lip plugs) seems to represent
Tarascan administrators sent by the state.

Imports used for the elite came not only from within the Tarascan state, but also

from beyond. Turquoise, for example, came from the northwest, marine shell from the

29

Pacific, and jade and jadeite from Oaxaca and the Motagua Valley in Honduras (Pollard,
1999). The channels of acquisition of these exotic precious items were less in number as
the distance of their sources increased, making the use of the items more restricted
(Pollard, 1993). These luxury goods served largely to maintain status differences
between the elite and the rest of the population.

Within the Lake Patzcuaro Basin in the heartland, the Tarascan state maintained
administrative control of its economic networks by establishing administrative units.
These administrative centers were directly under the control of the Tarascan capital, and
governed by members of an administrative elite class, designated by the Tarascan king
(Pollard, 1993). Gorenstein and Pollard (1983) used Thiessen polygons to locate the
administrative units in the Lake Patzcuaro Basin and rank them in a hierarchy of five
levels.

As the administrative center for the state territory, Tzintzuntzan was the only
Rank 1 center; and as the administrative center for the Basin, it was also designated as a
Rank 2 center. Below Tzintzuntzan there were eight administrative centers with Rank 3.
Rank 3 centers were those centers subject to Rank 2 settlements, and in administrative
control of several settlements (Rank 4 and 5). The Rank 3 administrative centers were all
equal in power and status within the state. Links between same-rank centers were
minimal if existent at all, and decision-making power flowed directly from Tzintzuntzan
(Gorenstein and Pollard, 1993).

Documentary evidence for the origin and expansion of the Tarascan state is
primarily known from the Relacio'n de las ceremonias y ritos y poblacién y gobernacio'n

de los indios de la provincia de Michoaca'n. This document, recorded between 153 8-

30

1541, was translated and transcribed by the Franciscan priest Fray J eronimo de Alcala
(Warren, 1971). Several editions of the Relacio’n de Michoaca'n (RM), as it is commonly
known, have been published (1956, 1980), including translations into English and French
(Pollard, 1993). Several other documents written within the Tarascan territory during the
early colonial period supplement the RM. These include the Relaciones Geogrdficas de
Michoaca'n (RG) of 1579—1580 (1958, 1985, 1987), and the Visitas de Carvajal of 1523-
1524 among others. Also important are sixteenth-century dictionaries and grammars of
the Tarascan language. These include a grammar written by Maturino Gilberti in 1558
and two dictionaries, one also by Gilberti in 1559 and another by Juan Baptista de
Lagunas in 1574 (Warren, 1983).

In all likelihood, episodes of lake level fluctuation influenced the development of
the Tarascan state. Lake level fluctuations directly affected the availability of irrigable
land, which was very important in supporting maize production (Pollard, 1982, 1993).
When the lake level was low, it exposed the most fertile land in the basin. The exposure
of fertile land may have attracted distant populations to the basin, contributing to an
increase in population density. Conversely, when the lake level was high, the availability
of exposed lacustrine soils along the lakeshore was reduced, reducing the agrarian
potential of the basin. A reduction in agrarian potential, in particular that for the
cultivation of maize, at a time when the population was increasing rapidly, would have
undoubtedly placed enormous pressure on local communities to compete for access to,
and control of, available irrigable land.

Pollard (1982) calculated that the basin could only produce enough maize to

support a population of 45,000, clearly not enough to feed the 80,000 people estimated to

31

have lived in the Basin by the end of the Late Postclassic ( A.D. 1350-1522). A
comparison of available resources, population size and consumption needs suggests that
the core of the Tarascan state in 1520, a time of elevated lakelevel, was not a viable
economic unit (Pollard, 1993; 1999).
The Development of the Tarascan State: Shifting Resources, Population Growth
and Consumption Needs

Lake Patzcuaro is a tropical, high-altitude, fi'eshwater lake in the Central Mexican
Plateau, located at 2035 m asl and approximately 360 km northwest of Mexico City. The
C-shaped lake is in a small closed basin that drains a catchment of 929 kmz; the surface
area covers 130 m2 (Chacon Torres, 1993; O’Hara, 1993). Lake Patzcuaro is an
amplifier lake, and is highly susceptible to minor climatic variations (O’Hara, 1993;
Street-Perrot and Harrison, 1985). A simultaneous small decrease in overall rainfall and
increase in temperature and evapo-transpiration rates can result in an annual 2 m lake

level difference (West, 1948).

Environmental Change

Minor climatic fluctuations in Lake Patzcuaro’s lake level have been documented
by several research projects in the region (Fisher et al., 1999; McCosh et al.,1999;
O’Hara, 1993; O’Hara et al., 1994). O’Hara (1991) examined the temporal and spatial
variation in rates of erosion in the Lake Patzcuaro Basin by analyzing the sediment
chemistry record from 20 short cores. The magnetic susceptibility records of the cores
provide evidence for a period of severe droughts during the Early Postclassic (AD. 900-

1000/1100) that resulted in a 4-5 m drop in lake level. This dry period was followed by a

32

wetter period that resulted in a rise in lake level between A.D. 1100-1300 (O’Hara,
1991), confirming the ethnohistoric record in the RM (1980), which provides evidence
for a rise in lake level to an elevation of 2039 m asl by AD. 1380 ( O’Hara, 1993).

A geomorphological study of the Lake Patzcuaro Basin produced a tightly
distributed sequence of AMS dates (AD. 120 — 850) from a mechanically excavated
trench in the southwest portion of the basin (Fisher et al., 1999). This dated sequence
records rapid and continual lake level fluctuations during most of the Classic period
(AD. 120 — 775), with an apparently long-term regression of the lake after AD. 775.
The formation of the clayey soil associated with this long-term drop in lake level is
buried under marsh, indicating a rise in lake level, and confirming the ethnohistoric
record (Fisher et al., 1999).

Settlement pattern data from a recent survey of the southwest portion of the Basin
show that from the Preclassic through the Epiclassic, settlements were generally widely
scattered, small and lacustrine in orientation (McCosh et al., 1999). The location of
recently exposed sites indicates that between A.D. 1000/1100 and AD. 1300 the lake
level was below 2032 m asl (Pollard, 2000). These data also suggest that many of the
lacustrine sites occupied between A.D. 1000\1100-1350, were abandoned after AD.
1350, as more communities moved away from the lakeshore (Pollard, 2000). The
location of earlier sites near the lakeshore and later sites at higher elevations indicate that
the lake level rose to 2040-2045 m asl (Pollard, 2000), corroborating O’Hara’s (1993)

estimate of 2039 m asl by AD. 1380.

33

Shifts in Resource Availability

Regressions in lake level affected the available resources in the Basin by
diminishing lacustrine resources and exposing rich, fertile lacustrine soils (Gorenstein
and Pollard, 1983; Pollard, 1982). Conversely, when the lake level was high, it covered
the most fertile land in the Basin. Pollard (1982, 1993) estimates that a rise in lake level
would have submerged 5 percent of the basin, resulting in a 60 to 70 percent reduction of
inigable land. Consequent to this shift in water level, lacustrine resources (especially
fish and waterfowl) would have increased while the potential to produce maize and beans
would have decreased (Pollard, 1982, 1993). The settlement pattern data and
geomorphological studies of soil erosion (Fisher et al., 1999) suggest that this was a time
of clearance of less fertile and more marginal lands. Concurrently, the availability of
deer and rabbit may have decreased as settlements moved to higher ground and cleared

marginal lands, displacing local fauna (Pollard, 1982, 1993).

Population Size and Density

This shift in available resources took place at a time of dramatic increases in
population size and density. The shift in settlement locations is concurrent with a marked
increase in the number and size of the settlements. During the Preclassic, Classic and
Epiclassic periods, the settlement of the Basin is by small communities that are widely
scattered and generally lacustrine in orientation (Pollard, 1999). During the Early
Postclassic, the number of sites increases (5 to 10) and the hectares occupied doubles (52

to 107), more than doubling during the Middle Postclassic (32 sites and 472 ha) (Pollard,

2000). During the Late Postclassic there is a slight drop in the number of sites (23) while
the area occupied is doubled again (851 ha) (Pollard, 2000), suggesting larger and denser

settlements.

Shifts in Socio-Political Organization

The absence of regional authority with decision-making power at a time of rapid
population growth led to competition between local elites for control and access to
diminishing resources. This competition resulted in the formation of highly nucleated
populations (Pollard, 1993). The RM (1980) indicates that this competition was the
cause for the succession of wars that concentrated economic and political power in the
hands of the uacusecha elites.

Major shifts in socio-political organization begin to take place in the Classic
Period. Between AD. 500 and 700 there were elites in centers around Central
Michoacén with a broadly shared tradition from the Late Preclassic Chupicuaro culture
(Pollard, 1996; 1999). The mortuary tradition of this period is dominated by primary
extended burials, but by the Late Classic Period, individuals were buried in group tombs,
used over multiple generations. Burials were associated with preciosities imported as
finished goods of raw materials derived from many parts of Mesoamerica. Cranial
deformation, dental mutilation and exotic grave goods served to distinguish the elites at
these centers from the surrounding populations (Pollard, 1996). Access to exotic goods
through participation in the macroregional exchange system marked local elite status.

The importation of these exotic finished goods indicates that elites derived their social

35

and economic power through linkages to other parts of western Mesoamerica (Pollard,
1996; 1999). Settlements have ball courts and mound groups with sunken plazas.

The breakdown of Classic period exchange networks and the re-organization of
the Mexican Highlands between the Early and Middle Postclassic (A.D. 900-1300)
brought about another shift in socio-political organization (Pollard, 1996). Large
nucleated upland sites appear, grow and expand, while smaller, less defensible lakeshore
sites are abandoned (Pollard, 1996). Centralization, social stratification and economic
and political integration begin between A.D. 1100-1350, leading to the formation of the
Tarascan state (Pollard, 1996).

In the Late Postclassic (A.D. 1350-1525) sites appear in less defensible locations
near the lakeshore. New areas of the uplands (malpais) are settled. There is a dramatic
growth in size and complexity of large urban centers with new monumental construction.
The Lake Patzcuaro Basin becomes the cosmological center in the state religion. Elites
no longer mark their status with imported exotic finished goods. Instead, they mark their
status with goods that link them to the Tarascan elites and to Tarascan state religion

(Pollard and Cahue, 1999).

Dietary Needs and Food Availability

By comparing productivity estimates and dietary needs Pollard (1982) showed
that the Lake Patzcuaro Basin produced insufficient maize to feed a population
consuming a hypothetical Tarascan diet. From ethnohistoric documents, we know that
the Tarascans utilized a range of foods that included more than fourteen genera of

domesticated plants, four genera of local fish, various local waterfowl, deer, small

36

mammals, domestic turkey, condiments (cacao, honey, salt) and lime for maize
preparation (Gorenstein and Pollard, 1983). Because there is little direct evidence for the
use of these foods by the Lake Patzcuaro Basin populations, Pollard (1982) proposed a
hypothetical diet based on foods consumed in the basin in the early sixteenth century and
on ethnographic data. In all probability, this diet reflected the consumption patterns of
Tarascan elites. While it is likely that the non-elite population consumed a different diet,
this difference is not well documented. Therefore, this diet will be referred to as the
Tarascan diet.

Theoretically, this elite diet would have provided approximately 2400 M
calories. Pollard’s (1982) estimates are based on a one third loss in nutrients due to
cooking procedures. The caloric content of the diet after taking this loss into account is
derived from 600g of maize and amaranth combined (1950 cal, 81.3%), 88g of beans
(176 cal, 7.3%), 100g fish (67 cal, 2.8%), and fruits and vegetables (200 cal annual
average, 8.3%) a day (Gorenstein and Pollard 1983; Pollard, 1982). Maize/amaranth,
beans and fish, combined, constitute approximately 90% of the total calories in this diet,
while maize/amaranth (C4 foods) alone constitute approximately 71% of total diet by
weight. The proportion of maize/amaranth in this diet is comparable to that in diets
estimated by Brand (1951) for ethnographic populations in the Lake Patzcuaro Basin,
Ivanhoe (1978) for the protohistoric Texcoco population, and Ortiz de Montellano (1990)
and Whitmore (1992) for the Aztec in the Basin of Mexico.

For 80,000 people to consume this hypothetical diet, the following annual
amounts of food would be needed: 17,520,000 kg maize/amaranth; 2,560,000 kg of

beans; 2,080,000 kg of fish; 832,000 kg of meat (primarily deer, rabbit, duck, and turkey)

37

(Pollard, 1982). The estimated productivity figures for these food items are not sufficient
to meet dietary needs, as there were only 9,821,200 kg of maize/amaranth, 2,331,050 kg
of beans, 4,732,800 kg of fish, and an average of 410,000 kg of meat (deer, rabbit, duck
and turkey combined) produced (Pollard, 1982). Of these items, only fish was available
at a surplus.
Meeting Dietary Needs: Dietary Change, Agricultural Intensification, or Economic
Exchange?

The rapidly growing population in the Lake Patzcuaro Basin may have responded
to declining maize availability by changing their diet, intensifying agricultural
production, or participating in economic exchange networks to obtain the needed maize

(and other resources).

Dietary Change

It is unlikely that a major shift in diet took place because maize most likely
provided 80% of the caloric dietary intake and the available alternative foods, while
abundant, could not have replaced these energy requirements (Ortis de Montellano, 1990;
Pollard, 1993; Whitmore, 1992; Whitmore and Williams, 1998). Assuming equal
consumption levels for all members of the population, regardless of age, sex, or social
status, each individual would have a diet that is calorie-energy deficient, but high in
protein (Table 2). Protein levels that are this high have been to hypercalciuria,
osteopaenia, and osteoporosis (Linkswiler, etal., 1974). Additionally, a calorie-deficient

diet would have serious health and social consequences.

38

TABLE 2

 

 

 

 

 

 

 

 

 

a Cravioto et al., 1945; Flores et al., 1960
b Gorenstein and Pollard, 1983

Lower Calorie Diet
Amount Calories (1) Protein (3)
g/day cal/day g/day
Maize 336 1202.88 28.22
358cal/100g 8.4g/100g
Beans 79 270.97 17.93
343ca1/100g 22.7g/100g
Fish 162 108.54 100.28
67cal/100g 9’ 61.9g/100g
SUBTOTAL 1582.39 146.43
Substitute
Deer 14 21.7 4.13
155ca1/100g 29.5g/100g
1604.09 150.56
Duck 14 16.1 3.26
115ca1/100g 23.3g/100g
1598.49 149.69
Rabbit 14 13.86 2.57
99cal/100g 18.4g/100g
1596.25 149
Turkey 14 37.52 2.81
268ca1/100g 20.1g/100g
1619.91 149.24
Requirements ° 2200 45
Males ‘ 2475 so
Females " 1920 70

° Ortiz de Montellano (1990) after FAQ-WHO (1985)

d Whitmore and Williams (1998)

39

A population consuming a calorie-deficient diet would have lacked the necessary
energy to engage successfully in military raids, activities associated with the development
and expansion of the state, and more importantly, food production. Changing the diet to
consume what was available would compromise the nutritional status of the population
and increase the risk of acquiring infectious diseases. Calorie-energy deficiency affects
the work capacity, fertility, mortality and morbidity of a population; disrupting the social,
political and economic structure of the community (Allen, 1984). It is unlikely that a
nutritionally compromised population would have had the work capacity necessary for
the construction and expansion evidenced in the large civic-ceremonial centers that
appear during this time. F urtherrnore, fertility rates do not seem to have been a problem,

as evidenced by the dramatic increase in population size and density.

Agricultural intensification

An alternative to dietary change could have been to intensify the cultivation of
available irrigable land, and to clear and cultivate less fertile and more marginal lands.
Geomorphological and archaeological data showthat intensive agriculture, while not
extensive, was practiced on terraces constructed on hill slopes between Tzintzuntzan and
Ihuatzio; and on non-lacustrine, irrigable lands (Polard, 1999). There is no evidence for
chinampa (raised fields) construction (Pollard, 1999), but agricultural canals ressembling
those described by West (1947) have been documented.

The radiocarbon dates associated with the agricultural features (canals) found in
the Lake Patzcuaro Basin indicate that they were constructed during the period of low

lake levels when population densities were relatively low, and before the formation of the

40

state (Fisher et al., 1999). These dated canals provide the earliest evidence of agricultural
features in the Lake Patzcuaro Basin, and represent the emergence of agricultural

intensification in the region.

Economic Exchange

In contrast, a more likely response seems to have been the acquisition of needed
foods from outside the basin given that maize was not only of nutritional value, but also
of symbolic significance to the Tarascans (as it was to other Mesoamerican societies)
(Ortiz de Montellano, 1990; Pollard, 1993). Given that the Tarascan state was not a
viable economic unit, it could have only existed through the establishment of
mechanisms of economic exchange (Pollard, 1982; Pollard; 1993).

Evidence for economic exchange in the Tarascan state can be found in the
ethnohistoric record. Market exchange is not amply recorded in the Relacién de
Michoaca'n (1980) or in the Relaciones Geogrdficas, but the RM (1980) shows a drawing
of the Asajo market. Pollard (1993) cites several terms from the Gilberti (1559)
dictionary, related to market exchange. These are merchant (mayapeti), marketplace
(mayepeto), trade (mayapecua), and marketing (mayapenit). The RM (1980) indicates
that three major markets serviced the Lake Patzcuaro Basin population during the Late
Postclassic (A.D. 1350-1520). Two of these were located within state administrative
centers at Tzintzuntzan and Pareo; the third, Asajo, was located northwest of the basin
(Gorenstein and Pollard, 1983).

Documented goods and services flowing through the markets include maize,

beans, chile peppers, amaranth, local fruit, ducks, local birds and feathers, fish, cotton

41

cloth and clothing, slaves, prepared food, medicinal plants, and household services
(grinding maize, carrying water, etc.) (Pollard, 1993). Undocumented goods believed to
have flowed through market networks include squash, nopal cactus, dogs, maguey fiber,
wood products, chert, opal, copal, salt, lime, rabbit, and turkey among others (Pollard,
1993). The availability of deer was limited in the Basin by restrictions to their natural
habitat, and although this resource is not documented in market network context, its
general use by the population indicates it may have flowed through the markets (Pollard,
1993).

While non-elites obtained their goods through the markets, elites obtained goods
through state-controlled institutions (tribute, state long-distance merchants, and although
poorly documented, local tribute through outright ownership to production) (Pollard,
1982). The Tarascan state had a vast, centralized and hierarchically organized tribute
network (Pollard, 1993). The bulk of goods from various regions of the state passed
through various levels up to the capital, Tzintzuntzan, where they were stored in central
warehouses (Pollard, 1993). The stored items were used for the Tarascan nobility at
Tzintzuntzan, foreign emissaries, and religious functionaries. In addition, these goods
served as emergency stores for the local population and during periods of war,
maintained the large armies (Pollard, 1993). The most common items on tribute lists
were maize and cotton cloth and clothing. Also on the lists were slaves, sacrificial
victims, household services, metal objects, cacao, gourds, gold, silver and copper. Salt,
beans, chile peppers, rabbit, turkey, honey, maguey wine, feathers and ceramic vessels

appeared in both market and tribute contexts (Pollard, 1993).

42

The state claimed ownership of the most productive lands in the Basin, and used
them to produce the food directly consumed by the royal household and Tarascan
nobility. The products grown on these lands included maize, beans, chile peppers,
squash, amaranth, fi'uit and tobacco. The Tarascan nobility may have also held exclusive
rights to firewood, lumber, deer and rabbit from local forests (Pollard, 1993).

Although not well documented, local tribute was probably a source of
manufactured goods and services. Since much of the Basin’s best land supported the
elite, a large portion of the foods forming the majority of the non-elite diet (maize and
beans) must have been imported through market networks (Pollard, 1999). In settlements
with immediate access to prime agricultural lands, marshes, or fishing zones people could
exchange surpluses for needed items (Pollard, 1993). In exchange for items flowing into
the Basin, the Tarascans exported some goods and services. Fish was the primary food
commodity exported fi'om the Basin in exchange for maize, amaranth, beans, and chili
peppers through the market network, and is associated with lakeshore populations
devoted to the harvesting of fish. In addition to fish, only manufactured goods seem to
have flowed outside the Basin.

In return for tribute and state control of land, forests, and mineral resources, the
elite exported “elite culture,” state religion, and administrative organization; maintaining
and defending the state’s borders (Pollard, 1993). The flow of services and goods within
the Tarascan state served to create and maintain uneven relationships between the Lake
Patzcuaro Basin and other regions of the state; and between Tzintzuntzan and other

administrative centers within the Basin (Pollard, 1993).

43

The elite appear to have used economic and political power as a mechanism to
cope with environmental change and food insufficiency, while the rest of the population
adapted by specializing in fish production and manufactured goods for exchange in

regional markets (Pollard, 1982; 1993; 1999).

Assessing the Nutritional Adequacy of the Hypothetical Tarascan Diet

A significant portion of the foods imported for the elite included sources of meat
protein, such as deer and rabbit, indicating that the elite had access to adequate sources of
protein. The nutritional value of the Tarascan diet proposed by Pollard (1982) was
reevaluated using the nutritional standards and requirements used by Ortiz de Montellano
(1992). This reevaluation shows that the Tarascan diet would have provided
approximately 67g of animal protein (132.27g of total protein) and 2516 cal when the
source of animal protein is fish. When fish is substituted with 100 g of other sources of
animal protein, the total protein in the diet averages 93g, ranging from 88.77g with rabbit
to 99.87 with deer. The caloric value averages 2609 cal, ranging from 2548.84 cal with
rabbit to 2717.84 cal with turkey. This diet is clearly above the safe levels of protein and
energy intakea defined by the Food and Agriculture Organization (FAO/WHO/UNU,
1985) of 2200 cal and 45g of protein. The adequacy of the protein in many maize staple
diets, however, may be low in the absence of complementing legumes, or lime treatment
during preparation; countering the general assumption that if energy needs are met, the
need for other nutrients will also be met (Ulijaszek, 1995).

Maize protein is deficient in lysine and tryptophan, but has fair amounts of

sulphur-containing amino acids (methionine and cystine). The convertion of tryptophan

to niacin is negatively affected by high proportions of leusine, raletive to isoleucine, in
diet heavily dependent on maize. Alkaline cooking techniques (such as lime treatment)
ameliorates the production of niacin by increasing the amount of isoleucine in maize
(Katz, et al., 1974). Alkaline cooking increases the amount of niacin and improves the
quality of the maize protein, but it also decreases its overall nutrient content (Araya et al.,
1981; Bye, 1981; Katz et al., 1974; Ortiz de Montellano, 1990).

In addition to alkaline cooking of maize, dietary tryptophan and niacin can
increased by consuming a maize-bean dietary complement. Protein in legumes is a
relatively rich source of tryptophan and niacin, but is low in sulfur-containing anrino
acids (Bye, 1981). A maize-bean dietary complement, however, is deficient in ascorbic
' acid, and riboflavin (Araya et al., 1981; Bye, 1981). Leafy vegetables such as amaranth
(“quelites”) complement the maize-bean diet in these nutrients (Bye, 1981).

The consrunption of whole fish adequately fulfills calcium requirements,
increases the total protein and improves the quality of dietary protein (Mayer, 1962).
While fish was clearly the primary source of animal protein, the Tarascan diet included
meat sources of protein (deer, rabbit, turkey, and duck) as evidenced in the RM (1980).
Pollard proposed two non-fish meats per week (Pollard, 1982). Squash fi'uits contribute
vitamins A and C, and minerals to the diet (Araya et al., 1981).

The nutritional composition of the Tarascan diet indicates that it adequately met
nutritional requirements, and was likely, not a significant source of nutritional illnesses.
Because of its complexity, and its heavy dependence on a relatively small number of
foods (maize, beans, fish), however, the Tarascan diet would have been very susceptible

to changes in the availability of these foods, regardless of magnitude. If the proportions

45

of foods in the Tarascan diet shifted enough to alter their nutritional complementarity,
individuals would have been more susceptible to infections due to the synergistic effect
between nutrition and infectious illnesses. If the population of the Lake Patzcuaro Basin
was unable to secure enough food to maintain the Tarascan diet, it may have experienced
an increase in infectious diseases and developmental defects secondary to nutritional

deficiencies.

Health in the T arascan State

Evidence for health in the Tarascan state is available fiom the ethnohistoric
documents and from a very limited number of skeletal studies. Ethnohistoric documents
were used in a study of Tarascan medicine and medical belief systems (Sepulveda y H,
1988). There were more than twenty terms in the Gilberti (1975) dictionary describing
accidents or trauma as a cause of illness, four terms for emotional ailments, more than
forty terms for pain, blindness and cataracts (“clouded eyes”). There were more than
forty-five terms for infectious or contagious diseases, although many of the terms used
describe symptoms of diseases imported by Europeans. Other conditions listed include
deliberate abortion, animal bites (snake and insects) and burns, but none referring to
hunger or malnourishment.

Osteological evidence of health within the Tarascan state is available fi'om only
two studies; one conducted by Urufluela (1996) at the site of Atoyac in the Lake Sayula
Basin, and one currently under way at the site of Urichu in the Lake Patzcuaro Basin

(Cahue, 1998).

46

TABLE 3

Frequency of Affected Indiviudals with Skeletal and Dental Indicators of Health

 

 

ATOYAC 13.72% 7.7 73.33% 51.66% 2
n= 51 n= 104 n= 30 n= 60

URICHU 12.50%
n= 16

90.00%
n= 30

61.66%
n= 60

 

 

 

 

 

 

 

 

 

 

 

The preliminary data in Table 3 show that in Atoyac, the frequency of individuals
affected with carious lesions decreased over time, while at Urichu, it increased. The
higher frequency of affected individuals in the Tarascan Urichu may reflect the age
distribution of the individuals in the sample. In the Pre-Tarascan sample, 62.5% of
individuals in young adulthood (20-34 yrs) and under, including adolescents and
children. In contrast, the percentage of individuals in these age brackets in the Tarascan
sample at Urichu is only 21% with only two children in the sample. Sixty eight percent
of the Tarascan sample is middle adulthood, with seven individuals between 35-49 years,
and four over fifty years.

Enamel hypoplastic defects show a marked decline over time in the Atoyac
sample, suggesting an improvement in the ability of children to overcome physiological
insults and survive into adulthood. Compared to the Atoyac populations, the Urichu
populations show lower frequencies of individuals with enamel hypoplastic defects, and

little difference over time, suggesting that the Urichu population was better able to cope

47

with physiological insults. None of the children in the Pre-Tarascan Urichu sample have
enamel hypoplastic defects

The frequencies of individuals with periosteal lesions show a temporal decline at
both sites, and the differences between the two sites for both time periods are negligible.
The periosteal lesions identified at these two sites provide evidence of non-specific
infectious processes. The temporal decline in periostitis prevalence in the Urichu case is
of interest because it is accompanied by an increase in population size and settlement.
This pattern highlights the argument, so eloquently made by Ortrrer (1992) regarding the
interpretation of skeletal infectious lesions. Very rarely do acute infectious illnesses and
viral infections affect the skeleton. Almost all infectious conditions that are evident in
the skeleton are chronic diseases and of bacterial etiology (Ortner, 1992). The higher
prevalence of periostitis in the Pre-Tarascan samples may indicate the presence of
chronic bacterial infectious diseases in individuals who managed not to be overwhelmed
by the infectious agent, but were unable to produce an immune response that would
effectively eliminate this agent from the body. As populations grew larger and denser,
chronic bacterial infections could have become more acute, killing their hosts before
there was time for skeletal lesions to develop.

Given the interaction between nutritional status and infectious diseases, infectious
disease should be more prevalent among populations whose diets fail to meet nutritional
requirements. Nutritional status, however, is not the only factor affecting the prevalence
of infectious diseases; the individual’s immune response, the age of onset, the biology of
the infectious agents and the social conditions contributing to its transmission are also

important (Ortner, 1992).

The preliminary data presented here, although limited, seem to indicate that the
Tarascan samples experienced more acute infectious diseases. Given this observation, a
study of infectious diseases among these populations would be incomplete without an
understanding of their nutritional status. An evaluation of the Tarascan diet will be the
first step toward this goal.

The overriding hypothesis in this dissertation is that the Tarascan elite were able
to use their political and economic power to obtain food from outside the Lake Patzcuaro
. Basin, supplementing goods extracted from the labor of local populations as they
produced food to meet tribute obligations.

To evaluate the effect of environmental change, food availability and economic
power on the diet of Tarascan elites, two analytical temporal units were determined: Pre-
Tarascan and Tarascan. The Pre-Tarascan period is temporally defined from the
beginning of the Classic period (AD. 350) up to the time of political and economic
unification of the Lake Patzcuaro Basin (ca 1300). The Tarascan period is defired from
the time of political and economic unification of the Lake Patzcuaro Basin to the time of
contact with Europeans (AD. 1525).

In summary, the available data for the Lake Patzcuaro Basin show that these two
analytical units differ in environmental conditions, agrarian potential, settlement patterns,
population size and density, economic exchange and socio-political organization.

The Pre-Tarascan period (A.D. 350-1300) (Figure 2), is associated with a low
lake level, high agrarian potential, and low lacustrine resources. During this time,
population size and density are relatively low, and elites are deriving their power fiom

links to outside elites, marking their status with exotic luxury items form outside the

49

Basin. The presence of two agricultural intensification features (canals) at a time of
relatively low population density indicates that in the Lake Patzcuaro Basin, the
emergence of agricultural intensification is not associated with demographic pressure as
proposed by Boserup (1965). Instead, agricultural intensification seems to signal the
production of surpluses, which may have been used by elites as tools to acquire exotic
goods as they attempted to consolidate their power (Fisher et al., 1999; Hayden 1991).
In contrast, the Tarascan period (A.D. 1300-1525) (Figure 3) is associated with
rising lake levels, reduced agrarian potential and high population densities. The
populations of the Basin are politically and economically unified, and elites are
hierarchically organized. Elites reside in administrative centers established throughout
the Basin, their power is derived fiom ties to Tzintzuntzan, and they mark their status
with locally crafted items that resemble those used and worn by the Tarascan nobility.
The preliminary data available on the health of Pre-Tarascan and Tarascan
populations in the region suggest a pattern of spatial and temporal variation in the
frequency of affected individuals for dental carious lesions, enamel hypoplastic defects,
and periosteal lesions. The decline in periosteal lesions coupled with a significant
demographic explosion in the region was most certainly affected by the nutritional status
of the population. Although a decline in nutritional status renders a population
susceptible to infectious diseases, the ethnohistoric and ethnographic data do not support
this proposition. To determine whether or not the nutritional status of Tarascan elites
declined, this dissertation used stable isotope analysis of bone collagen to detect spatial
and temporal dietary patterns, with the expectation that the diet of elites did not change

through time and that the diet of elites at various administrative centers would be the

50

same. Two pooled samples were formed, and differences in their isotopic values were
compared. Nonparametric statistics were used to evaluate the significance of the
findings. Chapter four describes the sampling criteria, the scientific hypotheses
formulated for this study, and the analytical methods used to evaluate the rejection of the

null hypotheses.

51

.853 .onaao can REEL 89a vommvoEv 32 mm _o>2 0x3
05 non? @0890 98— : emu—O can H 820 mo “55% 05 menses? 08:05am 815.? an: .N 059m

 

 

 

 

 

macaw—no.5—

U=2m~ a
2:5 : 820 I
2:3 _ 9.20 I

.955 3:5

 

 

 

 

_o>oA claim Bei—
imam
ehasouwem axed

 

 

52

282 .258 eea 238 See 8585 ewe. a 3:; 8E
05 sons» .833 .595 9:2 3 320 38 H 320 we Enema 05 9:39? ohmsofiem 3:3 me an; .m 0.5me

 

 

 

 

2325:5—
U

932 z

unai— = mas—U

155 — mun—U I

.951— 3::

 

 

 

 

 

 

 

 

_e>eq 8.3 .35
imam
92255.3 334

 

 

 

53

Chapter 4

HYPOTHESES, METHODS AND MATERIALS

The Tarascan state developed at a time when the population was growing,
settlements were becoming denser and the lake level was rising. Under these conditions,
the Lake Patzcuaro Basin as unable to produce enough maize to feed its population.
Considering the number of words in the Tarascan language for infectious illnesses and
associated symptoms (Sepulveda y H., 1988), it is entirely possible that these conditions
also resulted in an increase in infectious diseases.

If the Tarascan elite used their political and economic power to obtain foods and
resources that were not available in the Lake Patzcuaro Basin, their dietary intake of
maize should be close to that in the proposed T arascan diet. As discussed in Chapter 3,
this diet adequately fulfills energy and protein requirements (Pollard, 1982). Therefore,
the nutritional status of the Tarascan elite could not have compromised their health, yet
the limited data available on skeletal markers of health suggest a sift in diet (dental caries
decrease at Atoyac and increase at Urichu) and an increase in the incidence of acute
infectious illnesses (periosteal lesions decrease at both sites). Given the synergistic effect
between nutritional status and infectious illnesses, knowledge of the Tarascan diet and its
nutritional adequacy are essential for understanding the effect of state formation on the
health of individuals.

In this study, dietary patterns among elites were measured using stable isotope
geochemistry. It is important to note that the aim of the study is not to reconstruct the

actual diet, but to record the carbon and nitrogen isotopic signatures of one group relative

to another. An actual dietary reconstruction would essentially test the accuracy of the
proposed hypothetical diet; and although desirable, this test is not yet possible given the
limited information available about the isotope ecology of the region.

Theoretically, an increased dependence on C4 foods (maize and flesh of C4
consumers) leaves a 6'3 C signature in bone collagen that is higher than that for
consumers of C3 foods and the flesh C3 consumers. A diet dependent on legunrinous
plants will leave a 6'5N signature in bone collagen that is relatively low, although current
estimates indicate that legumes (like beans) must provide approximately 17% of the diet
before they produce a clearly discernible isotopic signature, typically, a difference in
5'5N of 2% (Schwarcz et al., 1985).

Differences in 6'5N values may be the result of shifts in trophic levels in the diet.
This phenomenon is best recognized in studies of carnivory versus herbivory, where the
SN differences in bone collagen are expected to be approximately 3%0 (Schoeninger
and Moore, 1992). Most human studies, such as the present one, are concerned with
whether people ate 10 versus 25% meat, for example (Schoeninger and Moore, 1992).
Current estimates suggest that humans could not survive on greater than 40% meat
protein consumption, with few exceptions such as the Eskimos and the Dogrib (Draper,
1977; Swathmary et al., 1987). Most human populations do not consume more than 25%
of their calories as meat protein (N011 and Avery, 1988). If the relationship of 8'5N of
bone collagen and diet were linear, a difference in meat consumption of 15% (hour 10
t015%) would leave 3 SN signature in bone collagen that is less than 0.5%0
(Schoeninger and Moore, 1992). To detect such shifts, samples that are very large and

have very little within group variation are necessary. In spite of these limitations, S‘SN

55

values are usefirl in determining relative dependence on meat (greater or lesser), although
the estimates must be regarded as crude (Schoeninger and Moore, 1992). Bone collagen
5'5N values seem to be more usefirl in determining differences between two ecosystems
providing dietary protein (e.g., marine versus terrestrial, coral reef versus deep ocean, and

lacustrine versus terrestrial).

MATERIALS

A regional approach to the study of the diet of the population living within the
Tarascan territory is justified by the cultural, linguistic and environmental homogeneity
within the region (Pollard, 1993). Ideally, this study would include a temporal
comparison of elite and non-elite individuals’ diets. It is unfortunate, however, that to
date, non-elite burials from this region are not available for study.

Collections from four archaeological sites located within the Tarascan territory
were selected to form two samples; Pre-Tarascan (PT) (AD. 1 - 1300) and Tarascan (T)
(AD. 1300 - 1525). The selection of sites was based on a three month pilot study
designed to conduct archival research to: 1) identify available skeletal materials from the
region, 2) locate excavated skeletal collections, and 3) assess the collections for degree of
skeletal fragmentation and bone preservation.

Several additional collections were surveyed in the study. While many
collections met the criteria for inclusion, they were not available for study at the time,
although future access to these collections is possible. In the Zacapu Basin, several
skeletal collections should be available for future study. From the Classic period, the

sites of Lorna Alta and Guadalupe provide excellent collections of skeletal remains from

56

discrete and commingled contexts. The Lorna Alta collection consists of thirty-one
funerary urns with cremated remains, one firnerary urn with non-cremated remains, six
primary burials and seven secondary burials (Amauld, et al., 1993). The Guadalupe
collection consists of the commingled remains of at least 40 individuals with good
preservation (Gervais, 1989). Also available form the Zacapu Basin are the Postclassic
collections fiom El Palacio and Milpillas. The El Palacio collection , excavated by
Lumholtz and Hrdilcka (1898), is curated in the Department of Anthropology of the
American Museum of Natural History in New York, City, New York. The Milpillas
collection is under the custody of the Centro de Estudios Mexicanos y Centro
Americanos (CEMCA) in Mexico City, Mexico, and it was studied by Olivier Puaux
(1989), as part of his doctoral theses, from the University of Paris 1, in Paris, France.

In the Cuitzeo Basin two large skeletal collections have been recovered from the
Classic to Postclassic site of Tres Cerritos and the Postclassic site of Huandacareo. Both
sites were excavated by Dr. Angelina Macias-Goytia, Researcher at the INAH - Centro
Regional, Michoacan. The Tres Cerritos collection represents the remains of over 50
individuals, from multiple burials in a tomb. The remains are reported to be in poor
condition, with very poor preservation (Macias-Goytia and Vackimes-Serret, 1989). The
site of Huandacareo dates to the Postclassic. Over 90 individuals are represented in this
collection from 88 individual and multiple burials (Macias-Goytia, 1990).

Outside the lake basins, two sites have yielded important skeletal collections,
neither of which is radiometrically dated, but both are assumed to date between the
Classic and the Epiclassic. These sites are Santa Maria on the outskirts of the city of

Morelia, and Tingarnbato (Tinganio). The Santa Maria collection is curated at the INAH-

57

Salvamento Arqeologico warehouses in the State of Mexico, outside of Mexico City,
Mexico, and it may represent more than 35 individuals. The number of individuals
represented in the multiple use tomb at Tingambato has been the subject of debate (Pifia
Chan and Of, 1982). One study estimates the number around 100 individuals (Lagunas
Rodriguez, 1987), but in a recent study Pereira (1997) estimates a minimum number of
individuals (MNI) of 36 based primarily on the calvaria (36), the mandibles (35) and
tehfemora (32 right and 24 left). The materials are housed in the National Museum of
Anthropology and History in Mexico City, Mexico. Requests were made to access all of
these collections, unfortunately, they were not available at the time this study was
conducted.

Four sites were available for study, three in the Lake Patzcuaro Basin (Técuaro,
Tzintzuntzan, and Urichu) (Figure 4), and one in the Sayula Basin (Atoyac) (Figure 1).
All the sites were administrative centers of various ranks under the direct control of the
state (Pollard, 1993).

Tzintzuntzan was the capital of the Tarascan state, and as such it held control over
the Tarascan Empire, placing it at the top (Rank 1) of the state’s administrative hierarchy.
Tzintzunztan was also the administrative center of the Lake Patzcuaro Basin, and as such,
it was designated a Rank 2. Eight other administrative centers were established within
the lake basin. Each one was governed by achaecha (or sefiores ) and reported directly to
the royal dynasty in Tzintzuntzan. These administrative center were: Erongaricuaro,
Urichu, Pechataro, Pareo, Pacandan-Xaracuaro, Itziparamucu, Uayarneo and Patzcuaro.
Each of these centers administered several dependent villages and hamlets, who reported

directly to them and had no direct contact to Tzintzuntzan. Each unit would have had

58

Amma 888050 can Egon 88m panacea 8389 end
32.5 dafigfifinh me 5582 wagon.“ 53m 08:05am 9:3 05 mo 32 v Bum—m

 

    

 

i i . .. d
e 3....
. . ...; 1 a . .
. .... .. .. t . EMN
. x212 ...». . .11, .x. L t
... r. .. .1 ..-. 1,. .
.r (it. . ..
I. 2. . . .. ...
. 3.. u r. r. v
. e . Vi . it... x... . e I
_. r \ ... .l .1. .
. . .. C {Kr .
. ... vii t. f
,. . a l . . .
a .... \ .... rt ... tr... it: t
1.. .. . ..
.11... F l.

w ~ ”wk/f1“.
.. 7:92 02:33.2 MES l,
L 1; .-.... .{1

wavy

 

 

 

 

 

 

 

 

 

 

\L T. ...- t

59

between one and sixteen dependent settlements, representing between 2,500 and 9,800
people. Within the administrative hierarchy of the state these centers were Rank 3
centers. Eighty two settlements were below a Rank 3, and associated with specific
archaeological sites. Unfortunately, many level 4 and 5 communities are not associated

with specific sites, and thus, cannot be distinguished archaeologically.

Tzintzuntzan - Rank 1 and Rank 2

The Instituto Nacional de Antropologia e Historia (INAH) has conducted over ten
field seasons at this Late Postclassic site (Castro Leal, 1986; Cabrera Castro, 1977, 1987).
The first and second seasons were directed by Alfonso Caso, assisted by E. Noguera in
1929, and J. Acosta and D. Rubin de la Borbolla in 1937 and 1938, respectively. (Cabrera
Castro, 1977). Rubin de la Borbolla directed the third (1940), fourth (1942), fifth (1943-
44) and sixth (1946) seasons assisted by H. Moedano, M. Acosta, R. Gali, and R. Pifia
Chan. One brief season (seventh) was conducted in 1956 under the direction of R.
Orellana. Btween 1962 and 1977, R. Pifla Chan directed three field seasons (eighth,
ninth, and tenth), assisted by K. Oi, R. Cabrera Castro and students of the Escuela
Nacional de Antropologia e Historia (ENAH) (Cabrera Castro, 1977).

Forty-one burials were recovered during the first nine field seasons, and date fi'om
the Late Postclassic to the time of contact (AD. 1350 — 1525) (Cabrera Castro, 1987).
Excavations during the tenth field season produced eighteen additional burials (17 at the
base of yacata 3 and l in structure F), bringing the total documented number to fifty-nine
Cabrera Castro, 1987), although this figure has been questioned (Castro Leal, 1987). In
general, male burials were associated with clay pipes and obsidian ear and lip plugs, and

the female burials with large numbers of polychrome ceramics (Castro Leal, 1986). The

60

V burials present a high incidence of dental alteration, types Al and A2 (Romero Molina,
1986), and cranial deformation, a characteristic explicitly referred to as a marker of elite
status in the Relacio'n de Michoacdn (1980: 187).

The burials located during the survey for skeletal remains were mostly those
excavated during the 10“1 field season, and were made available by, the archaeologist,
Lic. Rubén Cabrera Castro, Research Project Director at the archaeological zone of
Teotihuacan, INAH. Figure 5 shows the areas of the site where excavations have been
conducted (structures B, F , G). Not shown are excavation blocks A, C, D, and E.

The samples included in this study were obtained from the materials recovered
during the tenth field season. Although the direct association between sample and burial
location is lost, a map (Figure 6) was constructed utilizing original field notes and
sketches provided by Cabrera Castro indicating the location of burials recovered during
the tenth season. Tentative associations were made by correlating the labels in the boxes
and on the various bags within the boxes with field notes and drawings. This map
illustrates that all samples were taken fiom burials located on the main platform. The
associated goods and cranial deformation (currently under analysis) also indicate that

these were elite individuals.

Tzintzuntzan Ossuary

An ossuary off the northeast side of the main platform (Figure 5) was located
during the third field season (Rubin de la Borbolla, 1941). In 1992, a trench was
accidentally placed though the ossuary by a field worker, revealing the stratigraphy

described by Rubin de la Borbolla almost 50 years earlier. The ossuary contains

61

disarticulated remains, with an interesting stratigraphy. The first 70 cm consist of
fragmented irregular skeletal elements like vertebrae, scapulae, ribs, and small cranial
fiagments. Below this layer, there is a layer of calcined bone fragments containing
obsidian blades twisted by the heat of the fire. Below this layer there is layer of
disarticulated cranial bones and long bones, most of which are relatively complete.

This unfortunate accident also made it possible to examine a sample of the
skeletal remains contained in the ossuary. At the request of Efi'ain Cérdenas, Head
Archaeologist at the INAH Centro-Michoacan, the skeletal material was crudely
inventoried and a gross count of skeletal elements was submitted to him (Cahue,
unpublished data). The skeletal remains are generally more robust than those excavated
from the main platform, and from Urichu and Tocuaro. The preservation of the non-
bumt bone is excellent. There is considerable porotic hyperostosis, cribra orbitalia, and
infectious lesions (periosteal lesions, osteomyelitis).

The differences in morphology and pathology between these remains and those
excavated elsewhere suggest that the ossuary does not contain the remains of Tarascan
elites. Ethnohistoric accounts suggest that the ossuary contains the remains of sacrificial
victims (Rubin de la Borbolla, 1941). Sacrificial victims would have been captured
warriors and non-Tarascan elites captured during war, but Tarascan individuals who
offended the king could have also been killed at Tzintzuntzan, as a matter of justice and

not a sacrificial offering, and deposited in the ossuary.

62

Urichu - Rank 3

Urichu is located in the Lake Patzcuaro Basin (Figure 4), on the northwest portion
of a lava flow (malpais) south of the modern town of San Francisco Urichu (Figure 7).
Urichu is ethnohistorically documented to have been an elite center prior to the formation
of the state, and one of the eight administrative centers directly answering to
Tzintzuntzan after the state emerged (Pollard and Gorenstein, 1982). The site covers
more than 90 ha and contains at least three public zones with pyrarnid-plaza complexes,
deeply stratified cultural deposits, and intact human burials. Nine areas of the site have
been defined, and excavation units have been opened in three these: area 1, area 2, and
area 5. Area 5, at the southern end of the site was occupied fi'om the Classic (AD. 400)
to the Late Postclassic period (ca. AD. 1350), area 2 during the Middle Postclassic and
Late Postclassic (A.D. 1000/1100 — 1350), and area 1 during the Early Classic and Late
Postclassic (ca. AD. 400 - 1525).

Excavations in area 1 (five test pits in 1990 and a block excavation of 10 adjacent
2 x 2 m pits in 1995) revealed a residential/administrative structure, containing at least
three sequential clay floors, all dating to the Late Postclassic (Pollard and Cahue, 1999).
Eighteen intact elite burials were excavated in the walls, floors and outside this structure
(Figure 8). The grave goods associated with these burials include bronze artifacts,
polychrome ceramic vessels, carved bone, spinning and weaving tools, knives and
projectile points (Pollard, 2000, Pollard and Cahue, 1999). An underlying structure

constructed of large adobe bricks has been dated to the Classic period (AD. 300- 500),

63

and is currently the earliest occupation at the site. No burials were associated with this
structure (Pollard and Cahue, 1999).

Three test pits were excavated in area 2 in 1990 and a block excavation was
partially excavated in 1995. There are no intact discrete burials from area 2. Artifactual
and soil analyses indicate that area 2 represents three layers of cultural deposits (Pollard,
2000). First, a residential deposit (probably commoners) dating to the Middle
Postclassic, which included the disarticulated human remains of two individuals (pollard
and Cahue, 1999). Second, a residential deposit with a series of clay floors dating to the
Late Postclassic; and third, a residential midden above a red clay floor dating to the
sixteenth century (Pollard, 2000).

Area 5 consists of artificially terraced slopes separated by large retaining walls.
In 1990, two test pits were excavated adjacent to two pyramid platforms located on a
terrace. A block (10 x 6 m) was excavated in the terrace in 1994, incorporating one of
the test pits in which part of a tomb was found. In the upper levels of all units there was a
heavy representation of artifacts associated with Tarascan ritual and state religion
(Pollard, 1993, Pollard and Cahue, 1999). Below a rock pavement, there is a secondary
fill dating to the Early Postclassic (ca. AD. 900 - 1000); and this fill oaverlays a primary
deposit consisting of a series of scaled floors dating fiom the Late Classic to the Epi-
Classic period (AD. 500 — 900). Below the floors of the elite residential structure is a
sealed stone-lined tomb from which eighty-seven grave goods and five burials (ten

individuals) were recovered Pollard and Cahue, 1999).

The Urichu skeletal remains provide a temporal control for this study because
they represent the only collection of intact, stratigraphically controlled and
radiometrically dated burials in the Lake Patzcuaro Basin (Pollard, 1995).

Thirty-three burials and several disarticulated and scattered remains were
recovered, representing a minimal number of individuals of forty. In area 1(Figure 8),
there were 19 burials (21 individuals), in area 2 there were only disarticulated scattered
remains (2 individuals), and in Area 5 there were 14 burials (17 individuals). Of the
burials in Area 5 (Figure 9), five were inside a tomb, and three were among the flat
stones that formed the roof of the tomb. It is possible that two of these burials were
deposited on the roof of the tomb. The other burials are inside walls, in flat-stone boxes,
or directly on the bedrock.

The early (area 5) burials were in extended position, in tombs, flat-stone boxes, or
in walls. The more recent (area 1) burials were in a flexed position, most on the right
side. There are no observable patterns with regard to the orientation of the bodies. Four
of the burials recovered fiom Area 1 during the season of 1995 were looted (B28, B29,
B30 and B32) during excavations, and it is not known which was their exact position.
Some of these burials had been partially excavated prior to disturbance, in these cases it
was possible to determine the individual's approximate position. Looted burials were
reconstructed in the laboratory, based on size of bones, side of the bones, and the color
and state of preservation of the surface of the bone.

Preservation is varied, ranging from very poor to very good. The degrees of

completeness of the burials are average to good, but the remains are mostly fragmented.

65

Complete inventories for each burial can be found in the final project report to INAH

(Pollard, 2000).

Tbcuaro (Rank 4 or 5)

The Tocuaro collection was excavated during the 1996 field season of the
"Proyecto Carretera Patzcuaro-Uruapan" of the Subdireccion de Salvamento Arqeologico
- INAH, under the direction of the archaeologist Salvador Pulido Mendez. A total of
seven individuals were recovered (Pulido Mendez, personal communication). An
accelerated mass spectrometry (AMS) date from bone places this site in the Late
Postclassic (ca. AD. 1425). The burials were incomplete and the preservation was

generally poor. All burials were recovered from the fill of several structures (Figure 10).

Atoyac - Rank 4 or 5

The site of Atoyac was excavated as part of the “Proyecto Arqeolégico de la
Cuenca de Sayula” (PACS) (Acosta, 1996; Acosta et al., 1998). The site is located on the
eastern lakeshore of Lake Sayula. Excavations were conducted as a salvage project in the
neighborhood of San Juan in Atoyac, J alisco. A total of one hundred and fourteen
burials (MNI= 140) were recovered fi'om four areas of the site: residential area, burial
area 1, burial area 2, and burial area 3.

These areas were determined to be chronologically distinct. Based on Kelly’s
(1948) ceramic chronology, the residential area and burial areas 1 and 2 date to the
Amacueca phase (AD. 1100 — 1523), which corresponds to the Late Postclassic (AD.

900 — 1523); burial area 3 dates to the Sayula phase (AD. 600 — 1100) which

66

corresponds to the Middle Classic — Early Postclassic (A.D. 500/600 - 1100) (Acosta et
al., 1998). Isolated remains representing two individuals were also recovered and are
dated to the Verdia phase (0 — AD. 600) corresponding to the Late Preclassic — Early
Classic (0 — A.D. 500/600) (Acosta et al., 1998).

The Verdia phase is not well represented in the Sayula Basin. The shaft tomb
tradition begins to disappear during toward the end of this period. The absence of Verdia
remains in the Basin may be due to settlement shifts secondary to economic and political
restructuring of the basin (Acosta et al., 1998). The Sayula phase is characterized by
marked population explosion as settlements grow in number and density. There is
evidence for craft specialization (for example, salt extraction, ceramic production,
architecture, shell workshops) (Acosta etal., 1998). The presence of monumental
architecture and craft specialization suggest the presence of a complex social system
(perhaps a chiefdom) capable of harnessing human labor and securing the exploitation
and distribution of resources (Acosta et al., 1998). During the Amacueca phase there is a
restructuring of settlement patterns coupled with population growth, new ceramic styles,
and new techniques for salt extraction. All of these changes suggest 3 shift in economic
and social systems. There is also a change in mortuary treatment, as evidenced by the
type of grave goods. There is evidence for differential access to resources, and for
ascribed elite status (Acosta et al., 1998).

Of the one hundred and fourteen burials, ninety-two are individual and twenty-
two were multiple. There are thirty-two burials in area 2 (Sayula phase), thirty-three in
area 3 (Amacueca phase), twenty-seven in area 1 (Amacueca phase) and 20 in the

residential area (Amacueca phase (Acosta, 1996). Combining the areas, there are 32

67

 

 

 

 

 

 

 

:'
§ Ossuary
Q

B
TZINTZUNTZAN
ARCHAEOLOGICAL ZONE

A g F

0 “100 meters '13 -
_ 2 Chapel Ruins

 

 

 

 

 

 

 

Figure 5. Main Platform in the archaeological zone of Tzintzuntzan and excavated
structures. Most intact burials were excavated from block Y3/PNW (Modified from
Pollard, 1993).

68

.533 20¢ 52 05 watsu 38608 flags SSE 53> £52th _ x83 cougaoxm .o Emmi

 

 

 

2 m

 

 

 

3m

 

 

 

 

 

a:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

323m? .. ENHZDNHZ—NH

 

 

69

.83 .256 23 EEE ace 8585
@2388 80? >95 £5 E 32:03 flags 883 05 80:3 308 wage—a snotD .«o cam 05 no 932 .h oEwE

 

 

 

      

docuouohuoum

n Gouda I
53:93.“.

n don< I

ONUHMD QZMUNA

OUmuOZ<Mm 24m

 

 

 

DEOHMD

 

70

 

.aa_ .230 wee 3:8 aee BEBE macs Me 82.2: 332... a 82 gen: he 32 .w 2&3

 

 

8 if... E 5w. 5.?

 

me
am ......
em
2

 

_. m: a.-. a
_ .. _. eaten
Nm 58225.
~ 3.3.

Bug

 

 

 

 

 

 

71

 

 

 

 

B23

 

 

 

 

 

 

 

 

 

 

URICHU
AREA5
TERRACE 1 ‘ 825
1994 SEASON 2‘ ./
03
09
I Subadults $0
03
I Adults 00
Tomb 1
H B15 B13
21 B13 B14 812
B19
.817 -
..°_.
1N
l,_______.__' I,” l l
0 1 m

 

Figure 9. Map of Urichu, Area 5 showing the location of burials (modified
fi'om Pollard and Cahue, 1999).

72

 

.ESI: .8_m28=_:<
3:08“an 62:5 cadmium 89a seaming 5m? gamma—on— 23 382 Bob 352:5
53m 05 E 39:05 $3.89 mo dogwoo— ofi 9:305 33 3.5m .36 8550B .2 033%

 

 

 

 

 

 

 

H. imam 95:05.5 3.51—
OEDUOH.

 

 

 

 

 

73

burials in the Sayula phase (MNI = 45), and 80 burials in the Amacueca phase (MNI =
94) (Acosta, 1996).

Only twelve individuals in the Sayula phase have associated grave goods which
include bead necklaces and bracelets, shell objects, figurines, and ceramic vessels typical
of the Sayula ceramic style (Acosta, 1996). There is variation in the grave goods
associated with the Amacueca phase burials. The residential area burials are associated
with ceramic vessels, copper tweezers and needles, obsidian lip plugs, copper bells, stone
beads, and shell bracelets (Acosta, 1996). Area 1 burials are associated with few grave
goods, which include miniature vessels, obsidian lip plugs, copper wire and rings, and in
one case, shell beads (Acosta, 1996). Area 3 burials are associated with Tarascan
polychrome ceramics, and Amacueca style ceramics. Also included are copper bells and
tweezers, obsidian lip plugs, and in fewer quantities, shell pendants (Acosta, 1996).

The collection was under study at the time of the pilot study, but 10 samples: five
fi'om the Sayula phase (area 2), and five from the Amacueca phase (area 3) were seemed

for inclusion in the study.

Sample Composition

Sixty-five individuals previously identified as elites (based on grave goods,
spatial context, mortuary and ethnohistoric evidence) were selected for the study; only
adults were included. Seventeen individuals were included in the Pre-Tarascan sample;
11 males and 5 females and 1 undetermined. The Tarascan sample is composed of 48

individuals; 27 are males and 17 females and 4 undetermined (Table A1).

74

Permission for extraction and export of samples from the collections housed in
Mexico was obtained from the Consejo de Arqueologz'a, and the Direccio'n de
Antropologia F isica, Instituto Nacional de Antropologia e Historia (INAH). Samples of
five grams of bone were collected from skeletal elements that were the least diagnostic
for anatomical or pathological study. Bones were sampled with a preference for already
fiagmented skeletal elements since the available evidence suggests that there is not
difference in isotopic signature between skeletal elements in adults (Schwarcz and
Schoeninger, 1991).

F aunal remains, representing dietary items, were sampled, and they include fish,

rabbit, turkey, deer, peccary and dog.

Age and Sex Determinations

The fragmentary nature of many of these remains made the determination of age
and sex difficult. The age and sex determinations of the human skeletal remains were
assessed by Cahue at the end of each field season; all determinations were revised by L.
Cahue and N. Sauer at the end of the 1996 season during Sauer’s field supervisory visit.
Age estimates were determined primarily by examination of the degree of wear on the
molar teeth. Cahue and Sauer independently seriated all the dentition available and
assigned the individuals to one of six age categories: subadult (0-12), adolescent (12-20),
adult (20 +), young adult (20—35), middle adult (35-50), and older adult (50 +) (Buikstra
and Ubelaker 1994). Of the 40 individuals examined, the observers agreed on age
categories for 39 cases, and disagreed in one case where most of the teeth were missing.

In cases where only long bones were available, the age was assigned as adult (20+).

75

The assignation of sex for this population proved to be a more complex problem.
The method of choice for most physical anthropologists for the assessment of sex in the
human skeleton is, without a doubt, Phenice’s (1969) method (St. Hoyme and Iscan
1989). The Phenice method, based on morphologic indicators of sexual dimorphism in
the hip bone, while reliable, is inappropriate for very fragmented remains where the
necessary morphological features are unobservable, as is the case in most of the skeletal
remains in the study (Pollard and Cahue, 1999). In addition to the difficulties inherent in
sexing fragmented remains, this population is characterized by gracile postcrania with
robust crania. This biological characteristic ofien led, in the preliminary stages of
analysis, to conflicting sex assignations for the same individual, depending on which part
of the skeleton was being scored (postcranium or cranium). This pattern appears to
reflect the changes in sexually dimorphic features occurring with age. The morphology
of both male and female skulls tends to become more “masculine” as individuals get
older (White and F olkens, 1991; Meindl et al., 1985). Post-menopausal females develop
more “masculine” cranial features while developing an increasingly gracile postcranium
due to a greater susceptibility to age-related bone loss (Raisz, 1982; Walker, 1995).
Males continue to develop their sexually dimorphic cranial traits until the age of thirty
(Walker, 1995). Unrecognized, these age-related changes in sexually dimorphic features
can be problematic, especially in poorly preserved, fiagmentary and incomplete skeletal
collections where there is a tendency for young males to be classified as females or in an
undetermined category. Conversely, older females tend to be classified as males

(Walker, 1995). These biases lead to erroneous demographic profiles (Weiss, 1972).

76

To overcome these biases all available skeletal collections were surveyed to
assess the range of variation in sexually dimorphic features. Age was estimated before
sex was assigned. The survey reveled that the mandible was the best represented across
all the samples. A relatively new method for the assignation of sex in skeletal remains
using the mandibular ramus flexure was applied. The sex assignations determined by this
method were compared to those previously determined, and there was a one hundred
percent agreement.

Scoring of the morphological indicator of sexual dimorphism in the mandibular
ramus flexure (Loth and Henneberg 1996) was chosen because in most of the cases
examined, the mandible was the best represented across the population, the feature was
easily observed and scored, and it did not require an intact bone, thus, we were able to
include fi'agmented mandibles.

In spite of the challenges posed by incomplete, fragmented and sometimes poorly
preserved remains, two independent observers (Cahue and Sauer) agreed on the sex of all
but two individuals in the Urichu sample.

For the Tzintzuntzan materials, samples were taken from crania wherever
possible. In these cases, sex assignation was determined by analyzing sexually dimorphic
features of the skull. The long bone samples were clearly labeled and were recovered
from very distinct areas of the site, eliminating the possibility of sampling the same
individuals represented by the crania. The Tzintzuntzan-ossuary sample was obtained by

selecting left-sided fragmented mandibles.

77

HYPOTHESES

In order to evaluate the ability of elites to use their political and economic power
to obtain maize from outside the Lake Patzcuaro Basin, two sets of hypotheses regarding
the isotopic composition of the bone collagen of elites in the Tarascan state were tested.
One set evaluated temporal shifts in maize consumption, and another, the differences in
maize consumption between administrative centers within the Tarascan state of difl‘erent
rank.
Temporal Shifts in Maize Consumption

This set of hypotheses is concerned with temporal shifts in the relative
proportions of maize in the diet. Ethnohistoric sources indicate that maize was one of the
principal items flowing into the Lake Patzcuaro Basin as a tribute, indicating that maize
was of economic value to the state. Evidence for the symbolic value of maize to the
Tarascans can be found in its pervasive role in the language describing the anatomy, life
cycle and harvesting seasons, as well as metaphoric references for birth, strength, and
renewal of the society (Pollard, 1993). Given its nutritional, economic and symbolic
value, it is very likely that the Tarascans would have used all means available to obtain
this item. Therefore, this study postulates that the Tarascans used their political and
economic power to obtain maize, maintaining its dietary level.

If the elite were indeed, able to obtain maize for their consumption needs, the
513C and S'SN values from the bone collagen of Pre-Tarascan and Tarascan elites are
expected to show no difference. A downward shift in 613C would indicate a decline in
the relative proportion of maize in the diet. This could mean that 1) although maize

flowed into the Patzcuaro Basin the elite did not consume it, 2) the flow of maize into the

78

Basin was insufficient to offset supply deficits, or 3) the ethnohistoric record is inaccurate
and the elite did not have enough power to obtain maize from outside the Basin.

An upward shift in 8‘3 C would indicate that not only did the elite have enough
power to obtain maize, but also obtained enough to consume more of it. This, however,
would be very unlikely since an increase in the amount of maize in the diet would result
in a percentage of calories derived from maize higher than any reported (75%).

If the proportion of maize in the diet decreased it is possible that the population
may have attempted to meet calorie-energy needs by consuming more meat. This
scenario, however, is unlikely since the Basin’s supply of terrestrial meat resources was
also insufficient. In the hypothetical diet, less than 10% of the calories are derived from
meat protein and fish combined. For meat to provide 25% of calories, an individual
would have to reach a daily consumption of approximately 550g of meat and fish
combined. Current estimates for meat availability would only be enough for 14g/day per
person. An alternative would have been to increase the consumption of fish since it was
produced at a surplus. Fish provides 67ca1s/100g; it would take more than 500g of fish to
provide the same amount of calories in 100g of maize (358cal/100g). As shown in
Chapter 3, even under the assumption of equal distribution of food resources, and the
consumption of 100% of available resources (as estimated by Pollard, 1982), the
available fish harvest would only be enough to provide 162g fish/day per person (80,000
population). Therefore, no changes in the relative proportion of meat consumed are
expected.

NULL HYPOTHESIS I: There is no significant difference in the mean 813C
values of Pre-Tarascan and Tarascan elites within the Lake Patzcuaro Basin.

79

ALTERNATIVE HYPOTHESIS I: There is a significant difference in the
mean 813C values of Pre-Tarascan and Tarascan elites within the Lake Patzcuaro
Basin.

NULL HYPOTHESIS Ia: There is no significant difference in the mean 6‘5N
values of Pre-Tarascan and Tarascan elites within the Lake Patzcuaro Basin.

ALTERNATIVE HYPOTHESIS Ia: There is a significant difference in the

mean SUN values of Pre-Tarascan and Tarascan elites within the Lake Patzcuaro

Basin.

Ethnohistoric and archaeological data indicate that the site of Atoyac in the
Sayula Basin was part of the Tarascan state’s borderlands. The integration of this Basin
into the state took place between AD. 1440 and 1500. The high ratio of maize
consumed/maize carried for one individual making the trip to Sayula indicate that
Tarascan interest in the Sayula Basin was not maize. Tarascan interest in gold and silver,
coupled with the lack of salt in the Lake Patzcuaro Basin suggest that interest in the
Sayula Basin was most likely related to their rich gold and silver mines and to their salt
resources (Valdez and Liot, 1996). How did integration into the Tarascan State affect the
diet of elites in the Sayula Basin?

Recall that the Tarascan elite in the Lake Patzcuaro Basin exported elite culture
and military service in return for tribute. Tarascan control of conquered settlements
involved the presence of Tarascan elites from the geopolitical core, mainly from
Tzintzuntzan as administrators. Atoyac was under the control of the Tarascan state, and
it was located on the frontier margins of the state’s territory, or ethic segregation zone.
These settlements were populated by different ethnic communities. Local elites were
selected and approved by the Tarascan king to administer their own communities.

Tarascan elites from Tzintzuntzan were only in charge of small Tarascan communities

80

(Pollard, 1993). The Tarascan ambassadors would also insure the loyalty of all
communities to the state. The local elites in the Sayula Basin would have been able to
coerce labor from local populations, to work in the gold and silver mines, produce salt for
tribute and to maintain the production of maize for their consumption.

Dietary signals of bone collagen are set for life by adulthood. With a very slow
bone turnover rate (20 years), they vary very little throughout the life of an individual
(Stenhouse and Baxter, 1979). Therefore, it is possible to identify individuals whose
early-life diets were different from those of the local population. If the Tarascan
Amacueca burials represent local individuals who acquired the Tarascan ethnic identity,
their diets should be the same as the Pre-Tarasean Sayula sample. If they were Tarascan
elites sent from the capital of Tzinztuntzan, their diets should be the same as that of Elites

from Tzintzuntzan.

NULL HYPOTHESIS II: There is no significant difference in the mean 613C
values of Pre—Tarascan and Tarascan elites at Atoyac in the Sayula Basin.

ALTERNATIVE HYPOTHESIS II: There is a significant difference in the
mean 8'3 C values of Pre-Tarascan and Tarascan elites at Atoyac in the Sayula
Basin.

NULL HYPOTHESIS Ila: There is no significant difference in the mean S'SN
values of Pre-Tarascan and Tarascan elites at Atoyac in the Sayula Basin.

ALTERNATIVE HYPOTHESIS Ila: There is a significant difference in the
mean 8‘5N values of Pre-Tarascan and Tarascan elites at Atoyac in the Sayula
Basin.

Dietary Differences between Administrative Centers of Different Rank

Within the Tarascan core, the distribution of resources to elites may have been

uneven, and determined by the rank of the settlements they administered. Given the

81

symbolic and nutritional importance of maize in the Lake Patzcuaro Basin (as it was in
the rest of Mesoamerica), it is unlikely that elites would have tolerated reduced levels of
maize consumption. Thus, no differences between settlements in the proportion of maize
consumed are expected.

A decrease in the 613 C of Rank 3 and Rank 4/5 centers compared to Tzintzuntzan
(Rank 1 and Rank 2) would indicate that administrative elites were not consuming the
same amount of maize as the Tarascan royal family and nobility. This could be due to
uneven distribution of resources from Tzintzuntzan. It could also reflect the amount of
lacustrine, fertile lands under the control of the administrative centers, and their
population densities (Pollard, 1993). Elites at centers with less fertile land and higher
population densities may have been successful enough to prevent temporal shifts in diet,
but not as successful as other centers with more fertile land and less people.

An upward shift in 5‘3 C could reflect a form of “silent resistance.” to the political
and economic control from Tzintzuntzan. Local elites could have rebelled to state control
by hiding or hoarding maize, and “unofficially” consuming more maize than the nobility
at Tzintzuntzan. The severe punishments for such activities recorded in the ethnohistoric
records suggest that it may have been a problem for the state.

NULL HYPOTHESIS III: There is no significant difference in the mean 6'3C

values of elites at Tzintzuntzan and lower administrative centers (Urichu and

Tecuaro).

ALTERNATIVE HYPOTHESIS 111: There is a significant difference in the

mean 513C values of Tzintzuntzan and lower administrative centers (Urichu and

Técuaro).

NULL HYPOTHESIS IIIa: There is no significant difference in the mean 61$N
values of Tzintzuntzan and lower administrative centers (U richu and T6cuaro).

82

ALTERNATIVE HYPOTHESIS Illa: There is a significant difference in the
mean 8‘5N values of Tzintzuntzan and lower administrative centers (Urichu and
Tocuaro).

The ossuary at the capital city of Tzintzuntzan may contain the remains of war
victims, or individuals who served the royal family and nobility. Whether those victims
were captured warriors later sacrificed, or Tarascan waniors killed during war, is not
clear. Males engaged in warfare would have consumed a greater proportion of meat than
the rest of the population (Pollard, 1993).

If the ossuary remains represent Tarascan warriors, the 513 C values in bone
collagen from individuals in the ossuary are expected to be the same as those from the
T arascan nobility at Tzintzuntzan. If these remains represent servants, the 6'3 C values in
bone collagen from individuals in the ossuary are expected to be lower than those from
the Tarascan nobility at Tzintzuntzan.

If the ossuary remains represent captured warriors, their isotopic values should
reflect the elite diets of their settlements, for example Aztec.

The 8‘5N values of the Tzintzuntzan-Ossuary individuals is not expected to differ
from the Tzintzuntzan nobility. An upward shift in 515N may indicate the army members
consumed a better quality diet, while a downward shift would suggest that since members
of the army were drawn from elite centers throughout the state, their diets would reflect

be similar to the diet of other elites in the state.

NULL HYPOTHESIS IV: There is no significant difference in the mean 613C
values of Tzintzuntzan elites and the individuals from the Tzintzuntzan ossuary.

ALTERNATIVE HYPOTHESIS IV: There is a significant difference in the
mean 6'3 C values of Tzintzuntzan elites and the individuals fi‘om the Tzintzuntzan

ossuary

83

NULL HY POTHESIS IV a: There is no significant difference in the mean 6'5N
values of Tzintzuntzan elites and the individuals from the Tzintzuntzan ossuary.

ALTERNATIVE HY POTHESIS IV a: There is a significant difference in the

mean 8'5N values of Tzintzuntzan elites and the individuals from the
Tzintzuntzan ossuary.

METHODS
Sample Preparation

All samples were mechanically cleaned with a brush and a Drcmel tool to remove
the outer layer of bone. Clean bone pieces were rinsed in a 0.01 M HCl solution for 10
seconds, and immediately rinsed in cold (4°C) deionized water to neutrality. The samples
were then air dried under a hood at room temperature, for 48 hours.

The organic matter (“collagen”) was extracted from bone fragments by slow
demineralization (Moore et al., 1989; Sealy, 1986). Approximately 0.5g of clean, dry
bone were soaked in 50 m1 of a 0.25 M HCl solution for two weeks, and checked every
three days for translucency as an indicator of complete demineralization. The acid
solution was changed every five days to maintain the concentration gradient. The
demineralized fiagments were rinsed to neutrality, and soaked in 0.125 M NaOH for 20
hours, rinsed to neutrality again, and dried under vacuum at 55°C for 24 hours. Dry

samples were weighed and stored in combusted glass vials.

Analysis of Extracted Organic Matter (“collagen ”)
Between 0.4g and 0.6g of dry sample were weighed into tin'sample cups, and

placed in an automated sampler tray. An internal standard (sulphanilamide) was weighed

every 10 samples. The samples were flash combusted in a Carlo Erba elemental analyzer
attached to a Finnigan MAT Delta Plus mass spectrometer.

Each sampled was prepared for combustion twice: once for a measure of 613C,
and a second time for a dual measure of 613C and SISN to obtain data for carbon to
nitrogen ratio (C/N) calculations. Replicate samples were prepared and measured for a
randomly selected sub-sample. Carbon to nitrogen ratios were evaluated and compared

to collagen extraction yields to evaluate the integrity of the organic material.

Statistical Analysis

Choosing the appropriate statistical test in any scientific study is of paramount
importance. The choice of which test to use depends on the type of data, the shape of the
population from which samples are drawn, the size of the samples, whether samples are
paired (matched) or independent, and the standard deviation of the populations form
which the samples were drawn (Edmondson and Druce, 1996). In addition, the direction
of the alternative hypothesis must also be considered.

Parametric statistical tests make assumptions about these parameters. The
validity of these assumptions determines the meaningfulness of parametric tests (Siegel,
195 6). A requirement of parametric tests is that scores be measured at least at an interval
ratio level, that the shape of the populations be that of a normal distribution (bell shape),
and that the samples drawn be large. For the purpose of this study, sample sizes are
defined as: large (above 30); small (20 to 30); and very small (less than 20) (Edmondson

and Druce, 1996).

85

Nonparametric statistical tests do not specify conditions about the population
from which the sample was drawn. The assumptions made by nonparametric tests are
that the observations are independent and the variable under study is continuous. The
strength of measurements required for nonparametric tests is not as strong as that
required by parametric tests. Nonparametric tests require only that data be in at least an
ordinal scale.

Although the observations in this study are measured at an interval scale, the data
fail to meet most of the assumptions of parametric statistical tests. The shape of the
population is unknown, the sample sizes are very small, and the standard deviation of the
population is unknown. For these reasons, the nonparametric Mann-Whitney U test will
be used to test differences between two independent samples. This test is one of the most
powerful nonparametric tests, and it is the best alternative to the t-test, when parametric
assumptions cannot be met (Siegel, 1956).

The computation of the test statistic, U is based on the ranks of the values rather
than the values themselves. For an excellent step-by-step explanation, sec Edmondson
and Druce (1996). When one of the sample sizes is larger than 20 values the sampling
distribution of U rapidly approaches the normal distribution. When this happens, a
normal approximation of the U test is computed (Siegel, 1956). This study uses both
calculations, and Tables B1 and B2 provide examples of their calculation and formulas.

All of the tests in this study are two-tailed and were conducted at a 5% level of
significance (or= 0.05); results are not significant for p > 0.05. The Mann-Whitney U

tests were used to statistically test the hypotheses presented

86

In summary, two pooled samples were formed from various skeletal collections in
the region. One samples is Pre-Tarascan (AD. 1 - 1300) and another is Tarascan (AD.
1300 - 1525). Two sets of hypotheses were formulated to test the ability of the Tarascan
elites to obtain maize form elsewhere. These hypotheses test for diachronic and
synchronic differences in relative proportions of maize in the diet. The samples available
for study were very small, not permitting parametric statistical testing, therefore, this
study used nonparametric statistical tests (Mann—Whitney U Test). The descriptions of

the data, summary statistics and results of the tests are presented in the next chapter.

87

Chapter 5

RESULTS

The extraction yields, C/N values, and stable isotope ratios obtained for each of
the sixty-five specimens are listed in Table C1. The overall precision (extraction and
analysis) of the isotope analysis was i 0.1. The extraction yields (yields of extracted
protein) and carbon to nitrogen ratios of extracted proteins were measured during sample
preparation to assess reliability of the stable isotope ratios. Nineteen randomly selected
samples were analyzed several times to examine the reproducibility of the stable isotope
ratios (see Table C2). The measured standard deviations ranged fiom 0.01 to 0.3,
indicating good reproducibility.

Three of the samples analyzed were rejected as unreliable: samples URI-T-9.1
and TOC-T-7 are juveniles, and TOC-T-3.2 had a high C/N value. These samples are not
used in subsequent discussions.

The extraction yields ranged from 1.57% (SAY-PT-35.1) indicating poor
preservation, to 23.62% (TZN-T-13), characteristic of very well preserved bone.
Recommended cutoff limits for extraction yields range from 1% (Schoeninger and
Moore, 1992) to 5% (Schoeninger et al., 1989; Stafford et a1, 1991). This study will
follow the recommendations of Schoeninger and Moore (1992). All values are above the
recommended cutoff point of organic residue > 1% of original dry bone weight. The C/N
values ranged from 3.15 to 3.83 (TOC-T-3.2). All except one value (3.83) were within
the acceptable range (2.7 — 3.6) recommended by Schoeninger and Moore (1992). Figure

11 shows that the C/N values are less variable (3.33 :1; 0.13, expressed as the mean i one

88

standard deviation) than yield values (mean is 11.18% i 4.77), indicating no relation

between C/N and extraction yields (r = 0.335).

 

 

25 _
o o . r=0.335
b 20 _ 0.
g .
5 .2 15 _ 9°
8" >‘ e'
5 5 10 'v .
o '5 r o
3 Q: '
E 5 4 ~., .
III . .
0 l r I 1
2.00 2.50 3.00 3.50 4.00

C/N

Figure 11. Collagen extraction yields and C/N values for all treated samples in

the study.

The 8‘3 C and S'SN values for the sixty-two specimens that produced reliable

stable isotope ratios are listed in Table C1. The 6‘3C values range from -14.6%o to

-10.8%o, and the 6'5N values range from 7.1%o to 12.0%0. The lowest 8'3C value is

from a Tzintzuntzan female, and the highest is from a male in the Sayula sample. The

lowest 615N value is from the Area 2 female (URI-T-34) in the Urichu-T sample, and the

highest 615N value is from a female in the Tzintzuntzan sample. Table 3 shows summary

89

TABLE 4

 

Mean 513cm, and 5‘5Nco.I for Males and Females in
Pre-Tarascan and Tarascan Sites

 

TARASCAN
Urichu
Males (n = 10)
Females (n = 6)
Total (n = 16)
Tocuaro

Males (n = 3)
Females (n = 2)
Undetermined (n = 1)
Total (n = 6)
Tzint-Ossuary
Males (n = 2)
Females (n = 3)
Total (n = 5)
Atoyac-Amacueca
Males (n = 5)
Females (n = 0)
Total (n = 5)
Tzintzuntzan
Males (n = 7)
Females (n = 6)
Total (n = 13)

PRE-TARASCAN
Urichu
Males (n = 6)
Females (n = 5)
Undetermined (n = 1)
Total (n = 12)
Atoyac-Sayula
Males (n = 5)
Females (n = 0)
Total (n = 5)

 

 

 

6lSN (“0)

 

90

 

513C (x)

Mean S.D. Mean S.D.
—12- .3 0.8 9.3 0.8—
-11.7 0.6 8.8 0.5
-12.2 0.4 9.0 0.7
-11.8 0.5 8.9 0.6
-13.2 0.8 9.0 0.8
-l3.3 0.2 8.5 0.3
-13.3 0.0 8.7 0.0
-13.4 0.2 8.8 0.6
-12.6 0.2 9.2 0.2
-12.2 0.5 8.7 0.9
-12.4 0.4 8.8 0.7
-12.1 0.4 9.8 0.6
-12.1 0.4 9.8 0.6
-12.3 0.6 9.7 0.4
-12.4 1.2 10.1 1.2
-12.4 0.9 9.9 0.8
-12.2 0.7 9.4 0.8
~12.4 0.5 9.3 0.6
-12.2 0.7 8.8 1.0
-12.4 0.0 9.3 0.0
-12.3 0.6 9.1 0.8
-11.9 0.9 10.0 0.3
-11.9 0.9 10.0 0.3

 

statistics for the reliable stable isotope ratios from each site.

613C in Pre-Tarascan Samples

The mean 613C value for the Sayula sample (n=5) is more positive (-11.9 i 0.9)
compared with the Urichu-PT sample (n=12) (~12.3 : 0.6). This difference could be a
reflection of sex composition of the sample, since the Sayula sample does not include
females. A comparison between males shows a slight difference in means. The pooled

Pre-Tarascan sample (n=l7) has a mean 813C value of -12.2 i 0.7.

513 C in T arascan Samples

Among the Tarascan sites, the Tocuaro sample (n=6) produced the lowest mean
613 C value with the lowest standard deviation (-l3.4 _-l_- 0.2) compared with the other sites.
Of the remaining sites, Urichu-T (n=16) produced the highest (less negative) mean 813C
value (-1 1.8 i 0.5).

The difference in 613 C values between the Amacueca (n=5) (-l2.1 : 0.4),
Tzintzuntzan (n=13) (-12.4 i 0.9), and Tzintzuntzan-Ossuary (n=5) (-12.4 i 0.4) samples
is small. Urichu-T (n=16) has a 5% value (-11.8 i 0.5) that is 0.30/00 and 0.6%.) higher
than Amacueca and Tzintzuntzan and Tzintzuntzan-Ossuary respectively. Within the
Tzintzuntzan sample females show greater variation in 613C values (s.d. = 1.2) than males
(s.d. = 0.6).

The pooled Tarascan sample (n=45) has a mean 5'3C value of -12.3 i 0.8. The
mean 5'3 C value for pooled Tarascan males (n=27) is (-12.1 i 0.7) and for Tarascan

females (n=16) it is (-12.4 i 0.8).

91

6" 5 N in Pre-Tarascan Samples

The mean 8‘5N value for the Sayula sample from Atoyac is higher (10.0 i 0.3)
than that for the Urichu-PT sample (n=12) (9.1 i 0.8). A comparison among males alone
also shows that the Sayula males (n=5) have a higher mean 6‘5N value (10.0 i 0.3) than

the Urichu-PT males (9.3 i 0.6). The pooled Pre-Tarascan mean 8'5N value is 9.4 i 0.8.

SISN in T arascan Samples

The mean 8‘5N values for the Tarascan samples range from 8.8960 to 10.0%o.
The Tzintzuntzan sample (n=13) (9.9 i 0.8) and the Amacueca sample (n=5) from
Atoyac (9.8 i 0.6) have the highest 815N values. Within the Lake Patzcuaro Basin, the
Urichu-T (n=16) (8.9 i 0.6), Tocuaro (n=6) (8.8 i 0.6) and the Tzintzuntzan-Ossuary
(n=5) (8.8 i 0.7) samples have mean SUN values similar to one another, but, on average,
1.0%o lower than Tzintzuntzan (n=13) (9.9 _-I_~_ 0.8) and Amacueca (n=5) (9.8 i 0.6). The
pooled Tarascan (n=45) mean S‘SN value is 9.3 i 0.8. For males (n=27), the mean value

is 9.3 i 0.7, and for females (n=17) it is 9.2 i 1.1.

Statistical Tests
Differences in the mean 8‘3 C and 5‘5N values were tested with the Mann-

Whitney U statistical test. Tables B3 and B4 summarize the results of the tests.

Following is an examination of each null hypothesis with the Mann-Whitney U statistic.

NULL HYPOTHESIS I: There is no significant difference in the mean 613C
values of Pre-Tarascan and Tarascan elites within the Lake Pétzcuaro Basin.

92

The Mann-Whitney U test for differences in mean 513 C values between the pooled
Pre-Tarascan (-12.2 :1; 0.7) and Tarascan (-12.3 1 0.8) samples shows that there is not
enough evidence to reject the null hypothesis (Zen-p Zu; p>0.5). Thus, no significant

difference between the two groups was detected.

NULL HYPOTHESIS Ia: There is no significant difference in the mean 6‘5N
values of Pre-Tarascan and Tarascan elites within the Lake Patzcuaro Basin.

The Mann-Whitney U test for differences in mean 815N values between the
pooled Pre-Tarascan (9.4 i 0.8) and T arascan (9.3 _-|; 0.8) samples shows that there is not
enough evidence to reject the null hypothesis (Zen-p Zu; p>0.5). Thus, no significant

difference between the two groups was detected.

NULL HYPOTHESIS II: There is no significant difference in the mean 6‘3C
values of Pre-Tarascan and Tarascan elites within the Sayula Basin.

The Mann-Whitney U test for differences in mean 613 C values between the Pre-
Tarascan (Sayula) (-11.9 i 0.9) and Tarascan (Amacueca) (-l2.l : 0.4) samples from the
Sayula Basin (Atoyac) shows that there is not enough evidence to reject the null
hypothesis (U ’>U>Uc,i. ; p>0.5). Thus, no significant difference between the two groups

was detected.

NULL HYPOTHESIS Ila: There is no significant difference in the mean 6‘5N
values of Pre-Tarascan and Tarascan elites within the Sayula Basin.

The Mann-Whitney U test for differences in mean 6'5N values between the Pre-
Tarasean (Sayula) (10.0 1; 0.3) and Tarascan (Amacueca) (9.8 i 0.6) samples fi'om the

Sayula Basin (Atoyac) show that there is not enough evidence to reject the null

93

hypothesis (U>U’>Ucm; p>0.5). Thus, no significant difference between the two groups
was detected.

NULL HYPOTHESIS III: There is no significant difference in the mean 513 C

values of elites at Tzintzuntzan and lower administrative centers (Urichu and

Tocuaro).

The Mann-Whitney U test for differences in the mean 8” C values between the
Tzintzuntzan (-12.4 i 0.9) and Urichu-T (-11.8 i 0.5) samples shows that there is not
enough evidence to reject the null hypothesis (U>U’>Uen't; p>0.5). Thus, there is no
significant difference between the two groups.

The Mann-Whitney U test for differences in the mean 8” C values between the
Tzintzuntzan (-12.4 i 0.9) and Tocuaro (-13.4 i 0.2) samples shows that there is enough
evidence to reject the null hypothesis (U>Um-,>U’; p>0.5). Thus, there is a significant
difference between the two groups.

The Mann-Whitney U test for differences in the mean 8'3 C values between the
Tocuaro (-13.4 jg 0.2) and Urichu-T (-11.8 1; 0.5) samples shows that there is enough
evidence to reject the null hypothesis (U>Ucm>U’; p>0.5). Thus, there is a significant
difference between the two groups.

NULL HYPOTHESIS Illa: There is no significant difference in the mean 8'5N

values of elites at Tzintzuntzan and lower administrative centers (Urichu and

Técuaro).

The Mann-Whitney U test for differences in the mean 8'5N values between the

Tzintzuntzan (9.9 i 0.8) and Urichu-T (8.9 i 0.6) samples shows that there is enough

evidence to reject the null hypothesis (U ’>Uc,,-,>U; p>0.5). Thus, there is a significant
difference between the two groups.

The Mann-Whitney U test for differences in the mean 6'5N values between the
Tzintzuntzan (9.9 3; 0.8) and T6cuaro (8.8 i 0.6) samples shows that there is enough
evidence to reject the null hypothesis (U>Ucm>U’; p>0.5). Thus, there is a significant
difference between the two groups.

The Mann-Whitney U test for differences in the mean 61’N values between the
Tocuaro (8.8 i 0.6) and Urichu-T (8.9 i 0.6) samples shows that there is not enough
evidence to reject the null hypothesis (U>U’>Ucm; p>0.5). Thus, there is no significant

difference between the two groups.

NULL HYPOTHESIS IV: There is no significant difference in the mean 613C
values of Tzintzuntzan elites and the individuals fi'om the Tzintzuntzan ossuary.

The Mann-Whitney U test for differences in the mean 5'3 C values between the
Tzintzuntzan (-12.4 i 0.9) and Tzintzuntzan-Ossuary (-12.4 i 0.4) samples shows that
there is enough evidence to reject the null hypothesis (U>Ucm>U’; p>0.5). Thus, there is

a significant difference between the two groups.

NULL HYPOTHESIS IVa: There is no significant difference in the mean 5‘5N
values of Tzintzuntzan elites and the individuals from the Tzintzuntzan ossuary.

The Mann-Whitney U test for differences in the mean 6'5N values between the
Tzintzuntzan (9.9 i 0.8) and Tzintzuntzan-Ossuary (8.8 i 0.7) samples shows that there
is enough evidence to reject the null hypothesis (U>Ucm>U’; p>0.5). Thus, there is a

significant difference between the two groups.

95

In summary, the data from the pooled Pre-Tarascan and Tarascan samples support
the hypothesis that the proportion of maize in the diet of elites did not change over time.
The Mann Whitney U test shows that the differences between the two groups is not
significant. When examining the Lake Patzcuaro and Lake Sayula basins independently,
the data also indicate that there was no significant temporal change in the diet of elites.

The data examining differences between Tarascan administrative centers of
different rank shows that the 8‘3 C of elites at Tocuaro was significantly lower than that
all other sites in the study. The 815N data shows that mean values were significantly
different between Tocuaro and Tzinztuntzan, but not significantly different from that at
Urichu. The Amacueca individuals had a mean S'SN similar to that of Tzintzuntzan, and
the mean Sl’N value from the Tzintzuntzan-ossuary was similar to the Técuaro and

Urichu values.

96

Chapter 6

DISCUSSION, CONCLUSIONS AND FUTURE RESEARCH

The ability of elites to utilize economic power to obtain maize during the
development of the Tarascan state was tested by examining two sets of hypotheses. The
hypotheses were concerned with temporal shifts in diet and with dietary differences

between elites fiom differently ranked administrative centers.

Temporal Shifts

The absence of a temporal shift in mean 513 C and 6'5N values for the pooled Pre-
Tarascan and Tarascan samples support the hypothesis that Tarascan elites were able to
obtain enough maize to meet their dietary needs when the production of maize in Lake
Patzcuaro Basin was insufficient to support the population (Figure 12).

Disaggregating the data and examining differences between sites and between
differently-ranked administrative centers provides more insight on the way resources
were utilized within the state’s structure. The mean 613 C and SN values for each site
are presented in a bivariate plot (Figure 13). The plot shows that unlike elites at
administrative centers within the Lake Patzcuaro Basin, the elites at Atoyac (Sayula and
Amacueca phases, alike) had a diet very similar to that consumed by the elites at
Tzintzuntzan

The Sayula Basin incorporated into the state after AD. 1440. The process of state

expansion resulted in the incorporation of increasingly diverse communities. This

97

 

 

 

 

10.5 — o Pie-Tarascans
o 2) o. ; .Tarascans
10-0 - . ° ’ . o DPT-Mean
A O .00 o .T-Mean
9‘8 9.5 - .9” .
-'- ...,
z 9.0 - o ‘
:3 Q) o 0‘
4° . 8 O
8.5 - ° .0 O o
0 o
8.0 'r , -

 

 

-l4.0 -l3.0 -12.0 -11.0 -10.0 -9.0

8 13Ccoll (%0)

Figure 12. 813C and 8'5N values for Pre-Tarascan and Tarascan elites.

diversity challenged the state’s economic and political power. To ensure economic
exploitation of populations and resources, and to secure the integrity of its frontiers, the
state had to integrate these communities (Pollard, 1993).

The Sayula Basin was in the ethnic segregation zone, out in the fiontiers of the
state. The Tarascan burials fi'om the Sayula Basin are all males from Area 3 at Atoyac

(Amacueca phase), and they represent individuals buried with Tarascan burial goods.

98

 

 

 

 

Tarascan
SAY ‘

10.0 _ TZIN‘ o . Pre-Taras can
A ‘AMA
g 9.5 -
_-§ URI-PT

o URI-T

22 9.0 _ ‘
‘° r60 izmos

8.5 —1

800 l l T I

 

 

-l4.0 -13.0 -12.0 -ll.0 -10.0 -9.0

8 13C:eoll (0160)

Figure 13. 6'3C and 8N values for elites in Pre-Tarascan and Tarascan sites.

Current interpretations are that these individuals represent Tarascan
administrators sent out from the geopolitical core to secure the extraction of resources
(primarily silver and gold) and ensure the loyalty of local populations to the state, and
Tzintzuntzan elites in particular (Valdez and Liot, 1996). Dietary signals are fixed by
young adulthood, and stable carbon and nitrogen isotopes do not detect dietary shifts that
occur later in life (Stenhouse and Baxter, 1979). All the males at Atoyac (Sayula and
Amacueca) are between 25 and 50 years of age. Thus, the stable isotope data is
measuring the diets consumed early in life. If the “Tarascan” administrators were sent
out as adults, it is not surprising that their diet would be close to that of the Tzintzuntzan

elite. If they were local elites who acquired a Tarascan ethnic identity, the data may

99

indicate that diet served as an ethnic marker, depending on when the Tarascan identity
was acquired. While the loyalty of these administrators may have been assumed by
Tzintzuntzan, the danger of insurrection may have been reduced by providing the
administrators with resources (including food items) that identified them with the state.
The consumption of foods like those consumed by the Tzintzuntzan elite would have
been an effective ethnic and status marker, and a source of power for local elites. Stable
oxygen isotope analysis is in progress to determine if these individuals spent their
developing years in the Lake Patzcuaro Basin, or in the Sayula Basin.

A focus on the Sayula Basin reveals that male elites seem to have experienced a
slight temporal decrease in 8'5N, possibly reflecting a downward trophic level shift.
Valdez and Liot (1996) argue that subsistence strategies in the Sayula Basin underwent
seasonal shifts as the lake bed dried, exposing rich salt beds. They propose that the
population in the Sayula Basin practiced agriculture during the rainy season and stored
surpluses for consumption during the dry season, when they specialized in the production
of salt. Salt was used in exchange networks to supplement their needs. Earlier
populations are believed to have relied more on hunting and gathering than maize
agriculture (Valdez and Liot, 1996).

Estimates of population density in the Sayula Basin show an increasing trend, as
settlements move further away from the lakeshore into the uplands. An increase in
population density may have reduced the game available for consumption, reducing
dietary trophic levels and relative S‘SN values (10.0 to 98%, although not statistically
significant). Furthermore, the Basin’s incorporation into the Tarascan territory would

have increased demands for tribute items (metals, salt and minerals). To meet tribute

100

demands, people would have diverted their labor energy toward the production of salt and
mining for metals, and away fiom hunting activities. Assuming that the Amacueca
sample does represent Tarascan administrators sent from Tzintzuntzan, they would have
used power derived from their ethnic and elite status to exact maize and animal protein
fiom local tribute.

Temporal trends in the Lake Patzcuaro Basin can only be evaluated at the site of
Urichu. The 813C and SI’N values for the Pre-Tarascan sample (Area 5) and the Tarascan
sample (Area 1) show an increase in maize and a decrease in animal protein. The Mann-
Whitney U tests, however, show that the differences are statistically insignificant.

Within the Lake Patzcuaro Basin, Tarascan sites below Rank 2 (Urichu and
T6cuaro) cluster together in terms of 6'5N values, suggesting that their levels of meat
protein were similar. Urichu has a slightly higher (although not statistically significant)
mean 6'5N value, but significantly higher (less negative) 613C values. The hypothesis
that elites in administrative centers in the Lake Patzcuaro Basin consumed the same
relative proportions of maize is not supported by this data. Figure 13 clearly shows that
the diet of elites at Tzintzuntzan was different from the diets of elites in Técuaro (Rank
4/5), and Urichu (Rank 3).

The 513C data show that the elite at Tocuaro consumed less maize than at Urichu
and Tzintzuntzan. The 5'5N values between Urichu (Rank 3) and Tocuaro (Rank 4/5) do
not differ significantly, but the difference between Urichu and Tzintzuntzan is significant.
Comparing the S'SN mean value of Tocuaro with that of Tzintzuntzan, the greatest
difference is evident. The 1.1%o difference in 6'5N mean values clearly indicates that

the Tzintzuntzan elite consumed more meat protein, possibly deer, since this resource

101

was coming into the Basin as a tributary item and did not flow through the market
network. If the Tocuaro elite were obtaining their meat from non-elites though local
tribute, deer and rabbit would not have been a significant part of the diet. Another factor
affecting the T6cuaro diet is population density. Estimates for population density at
Tocuaro and surrounding settlements indicate that this portion of the Lake Patzcuaro
Basin was the most densely populated (McCosh, et al., 1999).

The Tarascan elites in Tzintzuntzan are consuming more animal protein than the
elites in the administrative centers, except for those elites out in the state’s frontier
borders. This difference in the allocation of such an important food resource parallels the
stratified status structure of the state, a trend that is less apparent in the distribution of
maize. The elites from administrative centers may have been less willing to tolerate a
shortage of maize compared to animal protein, in their diets. The position of their
settlement (rank) within the state’s sociopolitical structure may have influenced the
power of administrative elites to acquire resources.

Within Tzintzuntzan, the Tzintzuntzan-Ossuary sample has a 813 C value that is
the same as that of the Tzintzuntzan elite, and a S‘SN mean that is lower. These data may
indicate that the Tzintzuntzan elite’s diet was at a slightly higher trophic level. The
Tzintzuntzan elite may have consumed relatively more meat (Schoeninger, 1985), or they
could have been exploiting more aquatic resources, higher in 15N than terrestrial
(Katzenberg, 1989).

The 613C and 8N mean values of the Tzintzuntzan-Ossuary sample cluster more
closely with Urichu (Figure 13), indicating that their diet was similar to that of Rank 3

administrative elites. This finding does not support the hypothesis that the Tzintzuntzan-

102

Ossuary individuals had a diet similar to the diet of the Tzintzuntzan elite. The warriors
captured by the Tarascan army would have been members of elite society. If these men
resided in their respective communities, they would have been consuming diets similar to
those of other elites. It appears that the consumption of any Ca-based ceremonial foods
or beverages may have been too infrequent, or occurred too late in life to leave a
discernible isotopic signature. Therefore, any differences in diet between the
Tzintzuntzan-ossuary individuals and those from the main platform at Tzintzuntzan
morelikely, reflect the elite status of the victims used in sacrificial ceremonies that were
captured during wars.

The samples in the study did not allow adequate tests of differences between
males and females. In spite of this limitation, bivariate plots of the 613C and 8‘5N mean
values for males and females at each site do show some trends. Figure 14 shows that
among females alone, the Tzintzuntzan females have the highest 615N mean value (10.1 i
1.2), and are clearly different fi'om females at all the other sites in the Lake Patzcuaro
Basin (including the Pre-Tarascan group).

The Tzintzuntzan females differ in mean 6‘5N values from their Urichu,
Tzintzuntzan-Ossuary, and Tocuaro counterparts by 1.1%0, 1.4%o and 1.6%o,
respectively. Assuming the sample sizes were adequate, the magnitude of the variation
within the Tzintzuntzan female sample is still a problem for detecting a statistically
significant difference (see Chapter 4 for discussion).

The Tarascans forged kinship ties to conquered settlements by marrying females
of local elite families. The degree of variation within the Tzintzuntzan female group may

be a reflection of exogamy. After marriage to Tarascan nobility, these women would

103

move to the Lake Patzcuaro Basin, and in particular to Tzintzuntzan (Pollard, 1993).

 

10.5 _

 

 

a Tarascan
A 10.0 3 ‘ Tin o Pre-Tarascan‘
8 9.5 _
3
z 9-0 - A Uri .r
3 Uri -P'I‘ : .
Q 8.5 _ ‘ Tan-Os
Toc
8.0 I

 

 

-l4.0 -l3.0 -12.0 -11.0 -10.0 -9.0

8 13Ccoll (0160)

Figure 14. Mean 8‘3 C and SN values for Pre-Tarascan and Tarascan females.

Amacueca male samples from Atoyac in the Lake Sayula Basin cluster close to
Tzintzuntzan males and have the highest 6‘5N values (Figure 15). The Tzintzuntzan
males also show an elevated 815N mean value compared to males elsewhere in the Lake
Patzcuaro Basin, but the magnitude of this difference is not as large as that shown by
females. The only statistically significant difference is found betweenmales at Urichu
(8.8 + 0.5) and males at Amacueca (9.8 + 0.6) (Mann-Whitney U: U’>Ucm>U; p>0.5).
Of the Tarascan group, males in the Tzintzuntzan-Ossuary have higher SUN mean values

than T6cuaro and Urichu males; the Urichu males have the lowest value.

104

Is the elevated 8'5N mean value of females in Tzintzuntzan a record of diet, or
does it signal metabolic and physiological stress? If the former were to be assumed, the
idea that females consumed more animal protein than males is worthy of thorough
investigation, especially since the Tarascan state was firmly set on a system of gender
stratification. Yet, this elevated 615N may reflect an increased exploitation of aquatic
resources (fish), of which there were plenty. Gender divisions of labor were flexible
among commoners but less so among elites. The ireri, the king’s wife, ran the place, and
supervised elite women in various positions (Pollard, 1993). There were those who
ground corn, those who prepared the fish, and those who brought him drinks, among

many others (RM, 1980).

 

 

 

 

10.5 ‘ . Tarascan
10.0 3 SAY . . Pre-Tarascan
’3 TZIN ‘ AMA
53 9.5 4
___ . URI-PT
8 a
ml 9.0 2 ‘ TZIN-OS
“to TOC A URI-T
8.5 _
8.0 -

 

 

-l4.0 -13.0 -12.0 -11.0 -10.0 -9.0

5 13C:rtoll (0160)

Figure 15. Mean 813 C and SN values for Pre-Tarascan and Tarascan males.

105

Among contemporary Tarascans, women are in charge of preparing food. While
they will serve men first before serving themselves, women will “sample” the food as it is
being prepared, and in fact, many will have their “meals” by the cooking fire. In many
instances, the quality of the foods they consume may not be the most culturally desirable,
but in terms of nutrition they are as valuable. Women may save the breast, legs and
thighs of a duck or turkey for men, and eat the liver, heart, feet and neck themselves. It is
difficult to imagine hungry women not taking small portions of meat as they prepared the

meals for the king.

CONCLUSIONS

The inability of the Lake Patzcuaro Basin to support its population as the
Tarascan state developed created pressure on local populations to gain access to
resources, in particular food. The non-elite specialized in the exploitation of lacustrine
resources, which they would exchange for needed items through the state’s market
networks. The elite obtained their goods directly from Tzintzuntzan, where all tribute
items were stored for later distribution. In addition, local tribute and outright ownership
of local production were important sources of goods for elites at administrative centers.
These strategies served to protect elites fi'om the potentially harmful consequences of
dietary change. This study used dietary change as a measure of the power of elites to
maintain their diet. As a regional group, the elite had enough power to obtain maize and
other resources that functioned as elite markers.

Within the region, however, the power of elites seems to have varied. The

isotopic data reveals that there was variation in diet between sites. This variation clearly

106

reflects the rank of elite centers and suggests that elites in lower rank centers obtained
less maize and animal protein foods than elites at higher rank centers.

Another unexpected finding was the similarity in diet between the Atoyac-
Amacueca males, and the Tzintzuntzan elite. Although distant from Tzintzuntzan, the
need to ensure political loyalty to the state prompted Tzintzuntzan to place highly ranked,
loyal elites at this settlement to administer the exploitation of labor and resources and to
avert any insurrections. Because dietary signals are fixed by adulthood, these individuals
may have acquired the Tzintzuntzan dietary signal before leaving Tzintzuntzan, or at
Atoyac in the Sayula Basin. If the signal was acquired at Atoyac, dietary signals may be
serving as ethnic makers and a signal for power. To evaluate the place of origin, oxygen
and strontium isotopes are currently being analyzed. The dietary evidence in this study
shows that the people represented in the Tzintzuntzan-Ossuary may represent sacrificed

captured warriors from enemy elite centers.

CONTRIBUTIONS, LIMITATIONS AND FUTURE RESEARCH

This dissertation makes a valuable contribution to Tarascan prehistory in that it is
the first study to address the effect of political economic processes on the biology of
human populations. The greatest strength of the study is that it challenges uni-
dimensional explanations for the dietary and health patterns among human populations.
While incomplete, the study provides an essential foundation for the building of a
comprehensive and systematic research program in the region.

The focus on elites in this study brings into sharp view the debate over the role of

elites in the development of socioeconomic inequality. Are elites system-serving

107

individuals using their power for the good of the community (providing access to
resources in times of need), or, are they self-serving individuals who exploit and
manipulate the population for their own benefit (Hayden, 1990)? If elites were system-
serving, the non-elite populations at the administrative centers would have consumed
nutritionally adequate diets, even if different from elite diets. To address this question it
is necessary to first evaluate the effect of state formation and environmental change on
the diet and health of both elites and non-elites. A comparison of elites to non-elites,
while ideal, was not possible at the time this study was conducted. To date, non-elite
skeletal materials have not been recovered. One way in which to correct this bias in the
skeletal record is to conduct field excavations at sites that are not elite centers, e.g.,
market settlements.

Does the ethnohistoric record truly reflect the nutritional status and health of the
Tarascan elites? Are the apparent trends in dietary differences between males and
females an artifact of inadequate sampling? How did populations at other basins cope
with their incorporation into the state? Current interpretations are that these individuals
represent Tarascan.

To answer these questions, it will be necessary to address the most serious
limitation of this study: sample size. Plans are to collect samples from Tingambato (Pre-
Tarascan), and to expand the Atoyac (Pre-Tarascan and Tarascan) samples.

A project to examine the role of maize in the socio-political and economic shifts
detected in the Zacapu Basin is scheduled to begin January, 2001. As part of his project,

temporal shifts in diet will be examined, the skeletal materials will be systematically

108

studied to assess their nutritional status and health, and a comprehensive study of the
isotopic ecology of the Patzcuaro, Zacapu, Sayula and Cuitzeo basins will be completed.

The independent and systematic study of health and nutrition is urgent in West
Mexico. As cultural-historical models of west Mexican prehistory are replaced, new
hypotheses emerge about the role regional and interregional interactions have on the
emergence or suppression of local elites. Local, regional and imperial elites affected the
nature of interaction within Mesoamerica. How were these social, political and economic
shifts experienced by people, the real actors forging these behavioral patterns? What was
the effect of social complexity on the health or the populations of west Mexico?

Today, the relationship to the lake of the diverse, but mostly Tarascan
(Pureh'pecha) population in the Patzcuaro basin parallels prehistoric patterns in many
ways. Since 1990, the lake level has been receding and exposing fertile land. This
exposure of fertile land has come at a time when national and international political and
economic policies have resulted in shifts in land ownership and access to resources,
especially food. The effects of these shifts and the competition for control of resources,
have resulted in a reconfiguration of sociopolitical and economic power within, and
between, the basin's communities. Because the traditional Tarascan (Pureh'pecha) diet
continues to be mainly composed of maize, beans, chile and fish, the current changes in
the environment, and sociopolitical and economic power are sure to have an effect on the
health of people living in the Lake Patzcuaro basin, particularly that of the Tarascan

(Pureh'pecha) populations.

109

APPENDICES

110

APPENDIX A

LIST OF SAMPLES INCLUDED IN THE STUDY

111

 

+2.. 2 5088.5.-on 0685 £83. m Nd—
+o~ 2 508.3965 3830 :88. m Em.
8&— 2 5883-05 0mm - ashram Each m «.m—
Vmém m 5083505 0mm - agoram ESE m fix—
emrom m 5083505 «ma—c.5835 58m $22 m 2
+8 2 508.3505 820-55 ..th m 3
3-5 5 588558-05 mun—93m 988. m m—
emiou m 53505 0mm - mun—0.3m :83. m 5
+o~ 2 5088.5. 28255 83 m E5 _ 2
91mm 2 5088p. Emma—88m 83 m E5 _ 3
2.2 D 5883. 3828855 83 m E5 _ Em
+om m 508.8% Emma—88¢ 23 m E5 _ w
31mm m 5088... Emma—Swen 83 v E5 _ N.
3.1mm m 508.39 Emma—Snow 8m..— v E5 _ c
8.5 E 5088a. 38388.5 83 N E5 _ m
91mm 2 5.08.38 Emma—88m 85 v E5 _ v
3-5 2 .8885. Emma—28¢ 83 m E5 _ Nu
+8 2 5885. Emma—88¢ 23 m E5 _ _.~
+3 2 588mb Emma—38m 23 m E5 _ _
Diva:
ou< new 3.5 E5 8.3. 3.5:

 

scam 2.. a. 8.5.2: 8.9.5.

.—< Hanan—r

112

21mm
vmém
+cN
+om
+o~
+o~
«umém
Vmém
emém

+o~
mafimm

+Om
VméN
91mm
«WON
avfim
emém
vmém
Vmém

255222225

2252252555

 

0u<

 

>1
0
U)

505B
58555.
505.5.
58.55;
58.59
5855.5.
505.5.
58555.
5885.

5058-05

5855.5.
505.5.
5855B
58.55%
585%
5855.5.

50555605
585 H05
585550-05

 

2.0.5285 23 5. .55 32.59 a >65
2.0.8.285 23 3259 w >65
256.85 23 3259 5 >65
83285 23 2% 6m 5 .8 5 >65
8328.5 23 5mm .5 m .55 8 >65
2.0.3285 23 2% :5 m 5., 6 >65
25.285 23 m .65 m >65
25.285 23 3259 N >65
25.285 23 m .am 325.» _ >65
725.2555...
26.5285 5.5. N «m
235225 23 cm az _ mm
225225 23 um 22 _ mm
256285 23 5 dz _ 5
256285 23 B 22 _ om
25.285 23 um 22 _ mm
25.285 23 mm :2 _ em
2550 82 m a
256 558‘ m N:
2350 555 n ZN
5.5 .2555
2.5 55. 32 .555

 

A9283 ~< ”ma—ah.

113

 

 

55.2 $5 58555.
mg. 5 58555.
+8 5 58555.
35.3 ...5 58555.
+ON D gogh.
+ON D 503.3%
551mm 55 58555.
912 $5 58555.
5.8 ..5 5255
emém $5 58555.
55mm ..5 58555.
5.0% 2 503.35
«Won 55 58555.
+8 55 58555.
+o~ 5 5855 5.
+o~ ..5 5855 .5.
ONIN g hm 555508.35
0u< now

05855085 055
05855085 055
05855085 055
05855085 055
05855085 055
05855855 055
05855855 055
05855055 055

ommww_oum0nm Quad
Damn—030nm Quad
gnaw—030nm Quad
Gama—030nm 053
ommmflgwom OHNA

058550555 055
05855055 055
05855085 055
05855055 055

55 5.550
5 5550
5 5550
5 55550
5 5550
5 5550
5 5550
5 55550

QM .55

mm
<0
55-53
mm
mm
<5.-<©
<5.

5:80
5:80
5:80
5:80
5:80

V0500

5

0555005.

5 >65
5. >65
5. >65
5 >65
5 >65

bu:80..Z<.—.NZDN5.Z5NH

 

0555

58.555 5 55555. 2 >65
555.8555 5 55555. N5 >550:5
5:50.555 5 55555. 5 5 >555555
555. .555 325m> o5 >550555
.5550 .Z<N5.ZDN5.Z5N5.

55.5 502 55555.55

 

55.55.85 2 5.5.5555

114

 

 

min 2 58855. 83285 23 m 5.5
55-8 52 50855. 8328.5 33 m :2.
8-3 2 5085 88.285 33 m 5.2.
8-5 2 5085. 83380 23 m 2.5
8-8 2 53855 2.0.3285 23 m 5.8
van :5— «coauaa<unv<>o.—.<
8-55 2 5035.25 830-555 N :2.
8-3. 2 5355-25 2.30-555 N 5.8
8.3 2 58555-25 886-555 N 2.5
55-8 2 50355.25 886.5 N 5.5
8-8 2 5885.85 836-35 5 Z:
W 85.: 5.558-355:
03.. now I 0555 55:55 50.35 555.555

 

55.585 2 5.5:

115

APPENDIX B

MANN-WHITNEY U TESTS

116

TABLE Bl

6'3Cmu (%o) For Amacueca and Tocuaro Elites

 

 

 

 

 

 

 

Raw Measures Ranked Measures
AMA TOC AMA TOC
n1=5 n2=6 n1=5 n2=6
72.7 -13.6 '7 1.5
-12.2 -13.6 8.5 1.5
-12.2 -13.5 8.5 3
-l 1.9 -13.3 10 4.5
-11.6 -l3.3 11 4.5
-13.2 6
n = 5 6
SUM of RANKS 45 21
AVG. of RANKS 9.0 3.5

 

Mann-Whitney U Test
Two-Tailed test, p = 5%

 

n1= n of smaller sample

n1= n of larger sample

R1 = sum of ranks for smaller sample
U = [n.nz + n,(n, + l)/2] - R,

U' = ulnz - U

CALCULATIONS

ninz = 30
n1(n1+l)/2= 15

U=O
U'=30
Una-=3

EVALUATE HYPOTHESES
Reject Ho and accept Ha if:

smaller of U and U‘ is less than or
equal to Ucrit

 

Ho: There is no significant difference in mean 6 I3Ca,“ between TOC and AMA elites.
Ha: There is a significant difference in mean 6 13Ca," between TOC and AMA elites.
U'>U¢,.,>U: reject the Ho. P> 0.05

117

TABLE BZ
Example of Mann—Whitney U Test Worksheet - Z Approximation

 

 

 

 

 

 

 

 

 

Raw Measures Ranked Measures Mann-Whitney U Test
Pre-Tar arasca Pre-Tar Tarascani Two-Tailed test, p = 5% (0.05)
nl=ll n2=27 n1=11 n2=27

-12.9 -l3.6 5 1.5 n1= n of smaller sample

-12.9 -13.6 5 1.5 n; = n of larger sample

-12.7 -13.3 10 3 R1 = sum of ranks for smaller sample

-12.6 -12.9 13.5 5 U = [n,n2 + n,(n, + 1)/2] - R,

-12.6 -12.8 13.5 7.5 U' = mu; - U,

-12.4 -12.8 18 7.5

-12.3 -12.7 19 10

-12.2 -12.7 21 10 n,n2 = 297

-l 1.5 -12.6 31.5 13.5 n,(n,+l)/2= 66

-11.2 -12.6 34 13.5 U = 154.5

-10.8 -12.5 38 16.5 U' = 142.5

-12.5 16.5
-12.2 21 When n; increases in size, the sampling
-12.2 21 distribution approximates normality.
-11.9 24.5
-1 1.9 24.5 Mean = p.” = n,n2 I 2
-11.9 24.5
-1 1.9 24.5 _ _ 111112 (“#112“)
-11.8 27.5 S'D-°”_ \/ 12
-11.8 27.5
-11.6 29.5 Z= U-uU
-ll.6 29.5 T
-11.5 31.5
-11.2 34 Reject Ho and accept Ha if the positive
-11.2 34 value of Zn is greater than Zc,“
-1 1.1 36
-10.6 37
n = 11 27 Zn = 0.19
SUM of RANKS 208.5 557 Zed, = 1.96
AVG. of RANKS 19.0 19.9

 

 

Ho: There is no significant difference in mean 6 UCw“ between PT and T males
Ha: There is a significant difference in mean 8 13Ca," between PT and T males
Zerlt > ZU: the test fails to reject the H0.

118

cm “8.? 2 sum anwaAb M: _- 2 29::

 

EoDADAD EqubAD an 36 m . 9: Ya _ - m _ 555555.
cm 86.8.— 8 sum _.m T m «858:2.
atoDADAb EoDWbAD 2 an ow v.2- 2 ggficfih
mmanZm—U E—Héhw—Z—Eg
50min.
o: “8.6.— 8 mam EoDWDAb fimT n Amnaomfiab «858:3.
ESSA: anwas N we n: 3 7 m Amfioaassé 223m

525 «imam - 3.32

on so? 9 GE E2333 3 _- 2 8835
“EBAY: E0353 mm wm 42 3 _- 2 “5035-8;
525 c.3355— .. ...—3.5

 

o: 3%: 9 ea ....NA:N as- 2 ”some...
DNAEN SA a: .... .8 03 86 Nd- : 3835-5
....N :N 8:55 8.8m
.5325 a: .8? 3 5.220 .3: b a :82 .. mEEm 45.225.
3.. ....o: m

 

Ame... u .— .uozahékb mos—Er =80 nae—2 ..8 38,—. D 3533-:52 ..e ran—8.6
an

m»— Ham—<8

119

 

 

 

 

 

8: 6%: anwsAb on _- m 222;
568:5 beeswax: _ o a we- a $555.15
on so? 2 ea .EDwDAb 2 _- 2 32.5
E533 anwaa 3 2 mm Md- 5 gafisfl
8.32 5823.
cm .8? 2 fin £5.39 am _ - 2 museum
E595 “Samar: a: § 4mm «.2. R as: ,
589:5.
o: 28.? 2 :5" “Samoa: 2 _- m 838$
£5.59 £03.53 : 3m on ad. : 832
59338.9...—
magnum—ES
......EEE 9,2 542
a: 3%: “Emma: 4.2. o 8.32:.
56:53 .8353 a o 8 M: _- 2 22.5
a: .8?— anwaAb 4.2. 8 855$
.2533 E0353 2 8 fl «.2. m Engages
.538: a: .8? 3 .225 .5: .2 a :82 a
3% =39: w

 

A9233 mm Ham—4H.

120

 

2. .88. 8. ...... ..85w5A5 .2.- m 8.2.8...
285585 ..85w5A5 m 2.. n. . E . - o 8...:
:aomaaahé..m-=._u_.5
.... 8...... 0. ...... ..85w5A5 3.- m 5880-25.
.8555 ..85w.5A5 . 3 m... 8.5- . 8858......
o5 .88. 8. ...... .8555 2.- 8 2.25
285A 5 v.5 ..85w.5A5 m ... ... .25- . 58.5.82...
8...th engaging.
a: «camou— atoDWDAb —.N ~ 1 m «00...an
..85 u .55 ..85.v..5A5 o o 2 “.2- m 888.
.... .88. 8. ...... ..85w5A5 ..N . - m 88:88...
.8555 ..85w-5A5 w .. an ... .- o. 2.2.5
.... .88.. ..85w5A5 n..- m 288.
52855 ..85w5A5 m o on ... .- o. 2.25
.... .88. 8. ...... ..85w5A5 ..N .- m .8882
.8555 .mlw5wa5 m m. N. ......- 2 885.828...
.33qu c: 39.3.. 3 3.8.50 .5: .D D :32 .—
..... =25: w

 

2......8. mm 55.2...

121

85 .88. 8. ......
DNmEN

cm 80.8. 8 an
:NAEN

2. .88. 0. ......
28555

cm .8 .6. 8 :92
EoDADAb

o: .8 mo. 8 :am
...oDAbAD

8.285

 

«ENADN

3A «a an ..8

...aNAaN
3A :— .... ..8

.toDWDAb
...oowAD

.toDWDAb
EoDWbAD

EoDWDAb
.toowAD

 

a: 3&9. 3 «€220

 

 

Nd _ - w m 332
0mg :6 Wm _ . NN mug—mach
.EN 5N 233.3“. 6.3 3333.70...—
N. N _ - hm mgommbwh.
ea; m—d ”NT : gab-ohm
...:N 5N 8...: 1225
1N # u N. ~ macaw-EH
2 9w? mbm N.NT m mgomauwhéum
83:5..— 60.8.—
vd _ .. c mug—«Eon
c m.NN md~ WNT h. 3:32
5.55.555.
~23- o mofifiom
: 92 Wow 5; T o— 832
anguish-=52:
at»: .D D :60: E
A3». ..802 w

 

2......8. 8 555;

122

 

 

 

 

8: .83. ..85w5A5 ..o o. 2.2.5
5285A... ....o5w.5A5 an 3: m. . N .3 m. 885.822..
0.. .88. o. ...... ..85w5A5 ...o m 88:85.
.5555 ..85.v..5A5 N. EN EN a... N. 58.5825
9.59sz 525252.23.
avg-N
8.. .88. o. ...... ..85w5A5 .3 m 5588...: 888.5.
..85A5A.5 285M5A5 N 3.. w. . ...... m 55.8.5-8... 233
£35 «3.3m .. 2:33.
on .88. o. ...... ..85w5A5 3 ... .5885
..55A5A5 ..55w.5A5 mm 3.. n? ..m N. 28885-8...
525 cuazufiam - ...—3.5
cm .8? o. ...... ....NAaN 8.. N... N... a. 5.88....
:NkEN ..NAN: .... .... ....N ..N ...N N. 8885.8...
8.58am 60.25
5.285 a: 88...: 3 282.5 .....5 .5 5 58:. : m-Eim Son—2.:-
.3... .32.. m

 

Ame... u a .tozahékb mos—3r glam :32 ..8 38h. D 50523-552 «a 5.3885

3— Eda—<8

123

 

 

 

 

... .8... o. ...... ....5w5A5 .... m .8......
....5A.5A5 ....5w.5A5 . m .. N. N 585.82...
... .8.... ....5w5A5 w... o. 2.8.5
5A....5Ab ....owA: ... «.3 n.“ N. N 2.52.8222.-
$_a2 5.89:...-
o.. .8... .. ...... ....5w5A5 N. e. 8.2.8...
....5A5A5 ....5w.5A5 mm. nNN ...... ... NN 8...).
5.8.2:...
... .8... .. ...... ....5w5A5 .... m 8.2....
....5A5A5 ....5w.5A5 .. ..N. 5N... ... .. 8......
5933.70...—
Seaman-....
2.222... 572 ...-2:.
... .8... .. ...... ....5w5A5 .... o .88..
....5A.5A5 ....5w.5A5 .N ... om ... .. 2.8.5
... .8.... ....5w5A5 .... o .8......
52.55 .....5w.5A5 .. N. 8 ... m. 585.82....
8.28.. ... .8.... ... 282.0 ....5 .5 5 =8... ._ .
A30 =82... w

 

APE... .... 25......-

124

 

2. .8... .. .... ....5w5A5 .... m 8.2....
....5A5A5 .....5w5A5 ... m... m... m... . 8......
flaomahah.&hhu=:omha
... .8 ... .. .... ....5w5A5 N... m 2.8.0-2....
....5A.5A5 ....5w.5A5 . ..N .... ..o. . 2.2.222...
2. .8 ... .. ...... ....5w5A5 .... . 2.2.5
2.5275 .....5w5A5 m nN SN ...: . 2.22.22...-
8.2.8... 582......
... .8... .. ...... ....5w5A5 ... m .8282...
2.5255 ....5w5A5 o N m. .... m .28..
... .8.... ....5.v.5A5 .... m .8282...
52.552. .....5w.5A5 .. n... m... .... o. 2.2.5
... .8 ... .. .... .....5w5A5 .... m .8......
....5A5A.5 ....5w.5A5 m .. ... .... o. 2.2.5
2.. .8... .. .... ....5w5A5 .... m 8.2.2....~
...oDADAb ...oDWbAD m mm m. Num 5. 55.55%...
5.23: c..— .8?. 8 3.23.0 ...... .D D :32 .—
..... ..sz: ..

 

A9253 3— Mada—(h.

125

 

... .8... 8. ...... ....NA..N 8.. 3.. .... ... 8...).
DNAEN ..NA N2 .... .... ....N ..N ... NN 8.2....
...—32...; ...... 3333...“...—
om .8... .. .... .ENAaa .... .... ... NN 2.88...
DNmEN ..NA ... .... ... ....N ..N ... .. 2282......
«222 ...—cc.—
.5 .8... .. ...... ....5w5A5 N. N. 2.8....
2852.5 .....5w5A5 m. N. .... .... m .8.88......
8.2.3.... 8.2...
... .8... o. ...... ....5w5A5 ...: . 8.2....
.8555 ....5.v-.5A5 . .....N m... N. N 8.2).
5.5.5552...
... .8... .. .... ....5w5A5 .... o 82......
.5555 ....5w.5A5 .. m... 3N M... o. 8....
233.28-5523
5.289 a: 30?. 2 ...-.820 ....D .D D :32 .—
..... =37... ...

 

......28. .... ...-.39

126

APPENDIX C

STABLE ISOTOPE DATA

127

 

8.5-.55. 822...... 8... 8.:- E... a»... 2.5 . ...-.55
N8... 8...- NN. NN. ...-8 n. ...--25
N..N ....N.- NN. 8... +8 .. 89.55
8... 8...- ..N8 N8. 8-8 2 .385
8.. 8...- 8.... 8.. 98m .. 8-2-5.5
N... ..N...- 8.. 8... 8-8 .2 89.55
8... 8...- 8.. N... 218 2 .N-..-5.5
8... 8.5- 8.. 88. +8 .2 ...-5.5
8... 8....- 88 8... 8.8 2 ...-2-25
8.. 85. ..N.m 8N. +8 .. .-.-5.5
8.. .N.N.- 8.. 8... 8:8 .. 3.55
8... 8.5. NN.m 8.8. .78 . ...-5.5
NS. .....- NN.m 8.8 +8 .2 .-.-.25
8.. N....- ...... Ne... .88 s. 3-5.5
o... N...- 8.. ...... 8.5 2 23.55
NN... 8.... 8... 2.... +8 2 ..N-.-5.5
.8... N... .- 8.. N... +8 .2 .-.-5.5
2.2.5
5.85:9
$sz50 7...... o... 25 22> ... ...... .6... x8

 

82.3 2.83— ..oua.._z ...... ..c...a0 038m ...... 2.0 .20; 28.3.2... noun—.50

~0 ”Hum—<3.

128

 

8.2 .22- 2.. 8... +8 _2 ._-.-2..
...8 8:- 8.. 3.. +8 "2 2.2..
8.2 ...:- 8.. 2.8 +8 . :-.-2..
8.. ....2. 8.. ...: 8-2 . 2.2..
.... 2.:- 8.. ...: ...... m .-.-2..
... .....- .... .... ...-8 _2 .-.-2..
8.. 8d- ... , 8.. +8 2 .-.-2..
.... 8.2- 8.. 8.2 +8 2 .-.-2..
t... :82- .... .... +8 .2 .-.-2..
.... 8.2- .... 2... +8 .2 3-2..
.... 3.3- 8.. ...S ...-8 ..2 .-.-2..
8.. ....2- .... 2.2 ...-8 m .-.-2..
8.2 8.:- .... 8.. ...-8 2 7.2..
flaNuE—NHENH.
...umsmz...=§§ 8.. .....- a... ... . ... . ...-.8.
...-um... 38 ...... ... .....- .... ...» +8 . ...-.68
.... ....2- 8.. 8.2 ...-.. 2 .-.-co.
.... 8.._- 6.. 8.. +8 ”2 .-.-Do.
2.. 2.2. 3... .... ...-8 . ...-Do.
.... 8.2- 8.. .... +8 . 2..-.-00.
... 8.2- .... ...: .... 2 .-.-co.
.... 8.2. 8. 8.. .1. 2 _-.-uo.
6.338.!
82.550 2... Us. 25 so.» ... :5 .3. 2..

 

29...... 6 man...

129

 

a... .2:- om. mm..- mm-om E _..v-Hm-><m
9...: SN? mu. .... on-.. 2 _.mm-._-m-><m
md S...- v... o... 2.... 2 ~.v-Hm-><m
.2: 2.2- .... mm... ...-mm 2 _.m-._-m-><m
2.2 mm; 7 w... 3.2 mmém E 2. -Hm-><m
...—2am
72.038-82.—
89 $4 7 .... .5... ...-mm 2 flaw-8%
3.2 .2 T e... mm... mm-.. 2 2.89-22)?
3.. 3.2. m... 2.. on... 2 _.E--._--.<Z<
3.. 5.7 vu. 3... mm-.. 2 five-5&2...
wad— _~.m_- NN. we.» mm-.. 2 _.oc-H-<E<
60052.5(
a... ....— T .m. cm; _ ...-o... m ...-20.25.
N... ...~ 7 o... N: _ ...-cm 2 8.20.25.
9. 9.23- NN. o... 9.... m .-.-mO-ZNH
3.. 3.3- of. 2.2 +3 2 whammy-ZN..-
wv. we. _ - a... 8.2 ...-cm 2 TH-mO-ZN.
Paammcéfih

...-2.2.5 2:. Us. 25 2... .x. .25 .9. 2mm

 

.5233 U .55..

130

 

2.. ...2- o... 2.. +8 n. ...-2.25
8.. 8.:- .... 8... ...-8 ... 8-2-55
.... ...2- ..... 8... ...-8 2 ...-2-55
.... 8.2- 8.. .... ...-8 2 ...-2-25
.... 8.2- .... .... +8 2 ...-2-55
.... 8.:- 2. 8.. +8 2 ...-2-55
8.. ...2- o... 8.2 8-2 2 ...-2-25
.... .....- .... 8.. ...-8 ,2 ...-2-55
.... 8.:- .... ...: ...-8 2 2.2.2:
8... ...2- 8. ...... +8 2 ...-2-55
.... ...2- .... 8... +8 . 2-2-55
.... 8.. .- 8.. 8.. ...-8 2 2.2.5
......5
Dam-2.2.5 z... 5... 25 2.... .... a... .5“. x...

 

2......8. .5 2.5...

131

 

 

 

 

2.. .... 8.. 8.. 2.. .

8.. .- .... .- 5.. .
.... .... 8.. 8.. 25 ...-55

.... .... 2...

2.. .....- 2...:- .o...- 5.. m
8.. ... 25 ...-55

.... .... 2...

o... 2.:- 2.:- ....2- .....- 5...
.... .... 25 .-.-55

2.. 8.. .... .... 8.. 2.. .

8.. 3.:- 2..:- ...:- .....- 5...
8.. .... .... .... .... 25 ...-.55
35.5.35 5.. .... 2.2.2 . 2.2 . .92 ...... . .22 5 ...-224m

225.. 252

 

83...» 2.3:..— ..»uch E... 52.20 038m E... 7:0 33.33—

NU Human—-

132

8:25:50

Ed

cod

vmd

$6

56

:6

mod

86

 

 

 

 

 

.... ... .... 2...

...2- v..... 5... .

.... .... .... 25 ...-2-55

.... 8.. .... 2...

9... .- 2... .- 5... .

.... .... .... 25 ...-2-55

.... ... .... 2.. m

8.:- E .- 5.. .

.... .... 8.. 25 2:255

.... . .... .... 2.. m

...2- :..2- 5.. .

8.. 8.. .... .... 25 ...-.5
5: .... 24.52 . ...... . .5. ...... ...... I 5 2.22...

.25.. 24.2

 

.538. .5 .55..

133

 

 

 

 

. .3. ...... .8...... 8.. 8.. .... 2.. .

.....aomoeo. ...... 8.2- 8.2- 5.. .

8.3.6.. 2...... 3.. ...... .... 25 .-.-5o.
. ..3. ...... .8...... .... .... .... 2.. .

8.28.82. .... .... 8...- ...2- ......- E...- 5.. w

8.8.... 2...... .... .... .... 25 .-.-5o.
. .3. 2.... .8...... .... .... .... 2.. .

8288...... ...... .....- ...- ...- 5.. .

8.3.6.. 2...... .... 8.. .... 25 ...-.50.
. .3. 2.... .8...... .... .... 8.. 22 .

50.5225: 5... 8.... GMT 0.. w

3.3.... 2...... 8.. 8.. . . .. 25 .-.-50.
3585.5 >2. .... 252 . ..3. . ..3. . .3. . .3. ... ...-224m

22.... .22

 

.....ES. .5 .55...

. .3. 2.... .8....
vowiowofio: 2...

3.0530. 2.8.6
855890

 

mud

mad

mod

cad

86

Ed

86

NNd
3.2-
Rd

«ad—
054 7
end

and
3.2-
5}”
wad
awn...
Gd

 

 

>09 3w EH2

 

 

 

 

.... .... 2...
.....- 5...

.... .... 25 3.5.6.
...... .... 2...
.... .- N... .- .....- 5.. .

.... ...... 25 2.9.25‘
.... .... 2...
.....- 5...

a... .... 25 5-25
.... ...... 2...
.....- 5...

.... .... 25 3-55..

. .3. a .3. ..3. . .3. .. a. 21...)...

22.... >52

 

......8. .5 ......fi

135

 

 

 

 

 

 

.... .... ... .... 2.. .
.....- .....- 5.. .
o... ...... ... .... 25 .....-fig
8... .2: 2...
3.. .....- .... .- .... .- 5.. .
.... .... 25 ....-fig
.2: .2: 2...
.... .....- .....- .....- .....- 5.. .
.... .... 25 ....-ES...
....835 >3. .... 24.2 n .3. N .3. ..3. ..3. ... 2......
.25.. .22

 

......8. N5 2......

136

BIBLIOGRAPHY

137

BIBLIOGRAPHY

Acosta, Ma. del Rosario (1994). Las Entierros deI F raccionamiento San Juan Atoyac,
Jalisco. In Williams, E. (ed.): Contribuciones a la Arqueologia y Etnohistoria del
Occidente de Mexico. Zamora, Michoacan: El Colegio de Michoacan, pp 93-114.

Allen, Lindsey (1984). Functional Indicators of Nutritional Status of the Whole
Individual or the Community. Clinical Nutrition 35: 169-75.

Ambrose, Stanley H. (1990). Preparation and Characterization of Bone and Tooth
Collagen for Stable Carbon and Nitrogen Isotope Analysis. Journal of
Archaeological Science 17: 430-51.

Ambrose, Stanley H. (1991). Effects of Diet, Climate and Physiology on Nitrogen
Isotope Abundances in Terrestrial Foodwebs. Journal of Arcaheological Science
1 8: 293-3 17.

Ambrose, Stanley H. (1993). Isotopic Analysis of Paleodiets: Methodological and
Interpretive Considerations. In Sandford, M. K. (ed.): Elemental and Isotopc
Analyses: Understanding diet and Disease in Past Populations. New York:
Gordin and Beach Science Publishers, pp 59-130.

Ambrose, Stanley H and M. J. DeNiro (1986). The Isotope Ecology of East Afi‘ican
Mammals. Oecologia 69:395-406.

Ambrose, Stanley H. and L. Norr (1993). Experimental Evidence for the Relationship of
the Carbon Isotope Ratios of the Whole Diet and Dietary Protein to those of
Bone Collagen and Carbonate. In Lambert, J .B. and Grupe, G. (eds): Prehistoric
Human Bone: Archaeology at the Molecular Level. Berlin: Springer-Verlag, pp
1-37.

Araya, H., M. Flores, G. Arroyave (1981). Nutritive Value of Basic foods and Common
Dishes ofgthe Guatemalan Rural Populations. A Theoretical Approach. Ecology
of Food and Nutition 11: 171-76.

138

Amauld, Charlotte and B. F augere-Kalfon (1998). Evolucion de la Ocupacién Humana
en el Centro-Norte de Michoacan (Proyecto Michoacén, CEMCA) y la
Emergencia del Estado Tarasco. In Darras, V. (ed.): Genesis, Culutrals y
Espacios en Michoacan. Mexico City: CEMCA, pp 13-34.

Baker, Paul (1996). Adventures in Human Population Biology. Annual Review of
Anthropology 25: 1-18.

Blitz, J. H. (1993). Big Pots for Big Shots: feasting and Storage in a Mississippian
Community. American antiquity 58: 80-96.

Boserup, E. (1966). The Conditions of Agricultural Growth. Chicago: Aldine.

Bourque, B.J. and Krueger, H. W. (1994). Dietary Reconstruction from Human Bone
Isotopes for Five coastal New England Populations. In Sobolick, K.D. (ed.):
Palaeonutrition: The Diet and Health of Prehistoric Americans. Carbondale:
Center for Archaeological Investigations, Southern Illinois University, pp 192-
209.

Boutton, T.W. (1991). Stable Isotope ratios of Natural Materials. II. Atmospheric,
Terrestrial, Marine and Freshwater Environments. In Coleman, CD. and Fry, B.
(eds.): Carbon Isotope Techniques. New York: Academic Press, pp 173-85.

Boutton, T.W., M.J. Lynott, and MP. Bumsted (1991). Stable Carbon Isotopes and the
Study of Prehistoric Human Diet. Critical Reviews in Food Science and Nutrition
30: 373-85.

Buikstra, J .E. and Milner, GR. (1991). Isotopic and Archaeological Interpretations of
Diet in the Central Mississippi Valley. Journal of Archaeological Science 18:
3 19-29.

Buikstra, J .E., W. Autry, E, Breitburg, L. Eisenberg, and NJ. van der Merwe (1988).
Diet and Health in the Nashville Basin: Human Adaptation and Maize Agriculture
in Middle Tennessee. In Kennedy, B.E. and LeMoine, G.M. (eds.): Diet and
Subsistence: Current Archaeological Perspectives. Calgary: Archaeology
Association, University of Calgary, pp 243-59.

Buikstra, J.B. and Ubelaker, D. H. (eds.) (1994). Standards fro Data Collection from
Human Skeletal Remains. Proceedings of a Seminar at the Field Museum of
Natural History. Arkansas Archaeological Survey Research Series No. 44.

139

Bumsted, P. (1984). Human Variations: 613C in Adult Bone Collagen and the Relation to
diet in an Isochronous C4 (Maize) Archaeological Population. LA-10259-T Ph.D.
Thesis. Los Alamos National Laboratory.

Bye, R. (1981). Quelites - Ethnoecology of Edible Greens - Past, Present, and Future.
Journal of Ethnobiology 1: 109-23.

Cabrera Castro, R. (1987). Tzintzuntzan: Décima Temporada de Excavaciones. In
Homenagie a Roman Pifia Chan, UNAM,Instituto de Investigaciones
Antropologicas, Serie Antropolégica 79, Mexico, pp 531-65.

Cahue, L. (1999). Environmental Change, Sociopolitical Power and the Diet and Health
of Elites During the Emergence of the Tarascan State (abstract). American
J oumal of Physical Anthropology, Supplement 28, Abstracts to the Sixty-eighth
annual meeting of the American Association of Physical Anthmpologists.
Columbus, Ohio, pp 101.

Cahue, L., N]. Sauer and HP. Pollard (1998) Lip Plug (bezote) Abrasion Facets in
Tarascan Burial form Urichu, Michoacan. American Journal of Physical
Anthropology, Supplement 26, Abstracts to the Sixty-seventh annual meeting of
the American Association of Physical Anthropologists. Salt Lake City, Utah, pp
1 l7.

Cahue, L. and Pollard, H. P. (1998). Cambios en las Costumbres Funerarias de Urichu:
La Importancia de la Antropologia Fisica. In Avila, et al. (eds.): El Occidente de
Mexico: Arqueologia, Historia y Medio Ambiente. Guadalajara: Universidad de
Guadalajara, ORSTOM, pp 375-86.

Cahue, L. and Pollard, HP. (1999). Mortuary Patterns of Regional Elites in the Lake
Patzcuaro Basin of Western Mexico. Latin American Antiquity, 10: 259-80.

Cannon, A., H.P Schwarcz, and M. Knyf (1999). Amrine-Based Subsistence Trends and
the Stable Isotope Analysis of Dog Bones from Manu, British Columbia. Journal
of Archaeological Science 26:399-407.

Caplan, P. (1997). Approaches to Food, Health and Identity. In Caplan, P. (ed.): Food,
Health and Identity. London: Routledge, pp 2-31.

Castro Leal, M. (1986). Tzintzuntzan, Capital de los Tarascos. Gobiemo del Estado ed
Michoacan, Morelia, Mexico.

140

Chacon Torras, A. (1993). Lake Patzcuaro,Mexico: Watershed and Water Quality
Deterioration in a Tropical High-altitude Latin American Lake. Lake and
Rmois Management, 8: 37-47.

Chisholm, BS. (1989). Variation in Diet Reconstructions Based on Stable Carbon
Isotopic Evidence. In Price, T.D. (ed.): The Chemisz of Prehistoric Human
Bone. Cambridge: Cambridge University Press, pp 10-3 7.

Cohen, MN. and Armelagos, G. J. (1984). Paleopathology at the Originsof Agriculture.
New York: Academic Press.

de Garine, I and GA. Harison (1988) Coping with Uncertanity in Food Supply.
Clarendon Press. Oxford.

DeNiro, M.J. and Epstein S. (1978). Influence of Diet on the Distribution of Carbon
Isotopes in Animals. Geochimica et Cosmochimica Acta 42: 495-506.

DeNiro, M.J. and Epstein S. (1981). Influence of Diet on the Distribution of Nitrogen
Isotopes in Animals. Geochimica et Cosmochimica Acta 45: 341-51.

DeNiro, M.J. and Weiner, S. (1988). Chemical Enzymatic and Spectroscopic
characterization of “Collagen” and Other Organic Fractions from Prehistoric
Bones. Geochimica et Cosmochimica Acta 52: 2197-2206.

Dietler, M. (1996). Feasts and Commensal Politics in the Political Economy. Food,
Power and Status in Prehistoric Europe. In Wiessner,P. and Schiefenhovel, W.
(eds.): Food and the Status Quest. An Interdisciplinary Perspective. Providence:
Berghahn Books, pp 88-125.

Douglas, M. (1975). Deciphering a Meal. In Douglas, M (ed.): Implicit Meanings.
London: Routledge and Kegan Paul.

Draper, H.H. (1977). The aboriginal Eskimo Diet in Modem Perspective. American
Anthropologist 79: 309-16.

Edmondson, A. and Druce, D. (1996). Advanced Biology Statistics. Oxford: Oxford
University Press.

141

Energy and Protein Requirements: Report of a joint FAO/WHO/UNU Expert
Consultation (1985). WHO Technical Report Series 724, Geneva.

F aure, Gunter (1998). Principles and Applications of Geochemistry. New Jersey:
Prentice Hall.

Farquhar, G.D. M.H. O’Leary and J .A. Ben'y. (1982). On the Relationship between
Carbon Isotope Discrimination and the Inter-cellular Carbon Dioxide
Concentration in Leaves. Australian Journal of Plant Physiology 9: 121-37.

Fisher, C.T., H.P. Pollard and C. Frederick (1999). Landscape Change and Socio-
political Development within eh Lake Patzcuaro Basin. Antiquity 73: 642-49.

Gervais, Veronique (1989) Deformations Artificielles de cranes Préhispaniques au
Guatemala et au Mexique. These de Doctorat. UFR des Sciences de la Vie,
Université de Caen, Caen- CEMCA, Mexico.

Giddens, A. (1987). Sociology: A Brief bu Critical Introduction. San Diego: Harcourt
Brace J ovanovich.

Gilberti, R.P. Fr. Maturino (1975). Diccionario de la Lengua T arasca 0 de Michoacdn
(1559). Edicion facsirnilar, Pirnentel 1902, Balsal-Editores,S.A., Morelia,
Mexico. -

Goodman, AH. and Leatherman, T. L. (eds.): Building a New Biocultural Synthesis.
Political economic Perspectives on Human Biology. Ann Arbor: University of
Michigan Press.

Goodman, AH. and Leathennan, T. L. (1998). Traversing the Chasm between Biology
and Culture: Introsduction. In Goodman, AH. and Leathennan, T. L. (eds.):
Building a New Biocultural Synthesis. Political economic Perspectives on
Human Biology. Ann Arbor: University of Michigan Press, pp 245-67.

Goody, J. (1982). Cooking, Cuisine and Class. Cambridge: Cambridge University Press.

Gorenstein, S. and Pollard, HP. (1983). The Tarascan Civilization: A Late Prehispam'c
Cultural System. Nashville: Vanderbilt University Publiactions in Anthropology
28.

142

Gould, SJ. (1991). The wonderful Life: The Burgess Shale and the Nature of History.
New York: W.W. Norton.

Gould, SJ. and Lewontin, R. (1978). The Spandrels of San Marcos and the Panglossian
Paradigm: A Critique of the Adaptationist Approach. Proceedings of the Royal
Society of London 205: 581-98

Hare, RE. (1980). Organic Geochemistry of Bone and its Relation to the survival of
Bone in the Natural Environment. In Behrensmeyer, K. and Hill, A.P. (eds.):
Fossils in the Making. Chicago: University of Chicago Press, pp 208-19.

Harris, M. (1987). Foodways: Historical Overview and Theoretical Prolegomenon. In
Harris, M. and Ross, E.B. (eds.): Food and Evolution. Toward a Theory of Human
Food Habits. Philadelphis: Temple University Press, pp 57-90.

Hayden, B. (1996). Feasting in Prehistoric and Taditional Societies. In Wiessner,P. and
Schiefenhovel, W. (eds.): Food and the Status Quest. An Interdisciplinary
Perspective. Providence: Berghahn Books, pp 127-47.

Hayden, B. and Gargett, R. (1990) Big Man, Big Heart? A Mesoamerican View of the
Emergence of Complex Societies. Ancient Mesoamerica 1: 3-20.

Harris, M. and Ross, EB. (1987). Introduction. In Harris, M. and Ross, E.B. (eds.): Food
and Evolution. Toward a Theory of Human Food Habits. Philadelphis: Temple
University Press, pp 1-3.

Hastorf, CA. and Popper, V.S. (eds.) (1988). Current Paleoethnobotany: Analytical and
Cultural Interpretations of Archaeological Plant Remains. Chicago: University of
Chicago Press.

Heaton, T.H.E.; Vogel, J .C.; von la Chevallerie, G.; and Gollett, G. (1986). Climatic
Influence on the Isotopic Composition of Bone Nitrogen. Nature 32: 822-23.

Hedges, REM. and Law, LA. (1989). The Radiocarbon Dating of Bone. Applied
Geochemistry 4: 249-253.

Hoefs, J. (1987). Stable Isotope Geochemistry. New York: springer-Verlag.

143

Huelsbeck, D. (1988). F aunal Remains and the Identification of Social Groups in the
Archaeological Record. In Webb, B. (ed.): Recent Developments in
Environmental Analysis in Old and New World Archaeology. Oxford: BAR
International Series, pp 131-49.

Ivanhoe, F. (1978). Diet and Demography in Texcoco on the Eve of the Spanish
Conquest: A Senriquantitative Reconstruction from Selected Ethnohistoric Texts.
Revist Mexiacna de Estudios Antropolégicos 34: 137-45.

Jackson, HE. and Scott, S.L. (1995). The Faunal Record of the Southeastern Elite: The
Implications of Economy, Social Relations, and Ideology. Southeastern
Archaeology 14: 103-19.

Joharmssen, S. (1988). Plant Remains and Culture Change: Are Paleoethnobotanical
Data Better than we Think? In Hastorf, CA. and Popper, V.S. (eds.): Current
Paleoethnobotany: Analytical and Cultural Interpretations of Archaeological Plant
Remains. Chicago: University of Chicago Press, pp 145-66.

Jones, TL. (1996). Mortars, Pestles, and Division of Labor in Prehistoric California: A
View from Big Sur. American Antiquity 61: 243-64.

Katz, S.H. M.L. Hediger and LA. Valleroy (1974). Traditional Maize Processing
Techniques in the New World. Science 184: 765-73.

Katzenberg, M.A. (1989). Stable Isotope Analysis of Archaeological F aunal Remains
from Southern Ontario. Journal of Archaeological Science 16: 319-29.

Katzenberg, M.A. (1992). Advances in Stable Isotope Analysis of Prehistoric Bones. In
Saunders, SR. and Katzenberg, M.A. (eds.): Skeletal Biology of Past Peoples:
Research Methods. New York: Wiley-Liss, pp 105-19.

Katzenberg, M.A. and Harrison, R.G. (1997). What’s in a Bone? Recent Advances in
Archaeological Bone Chemistry. Journal of Archaeological Research 5: 265-
293.

Koch, P.L. AK. Behrensmeyer, N. Tuross and ML. Fogel (1990). Isotopic Fidelity
during Bone Weathering and Burial. Annual Report of the Director of the
Geophysical Laboratory, Carnegie Institute of Washington 1989-1990: 105-10.

144

Krueger, H.W. and Sullivan, CH. (1984). Models for Carbon Isotope Fractionation
between Diet and Bone. In Tumlund, J .R. and Johnson, P.E. (eds.) Stable
Isotopes n Nutrition, American Chemical Society Symposium Series,
Washington, DC, pp 205-20.

Lagunas, Fray Juan Baptista de (1574) (1983). Arte y Diccionario con Otras Obras en
Lengua Michuacana. Edicién dacsimilar, (edited by J. Warren). FIMAX
Publicistas Editores, Morelai, Mexico.

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

Larsen, C.S., R. Shavit and M.C. Griffin (1991). Dental Caries Evidence for Dietary
Change: An Archaeological Context. In Kelly, M.A. and Larsen, C.S. (eds.):
Advances in Dental Anthropology. New York: Wiley-Liss, pp 179-202.

Leathennan, TL. (1998). Illness, Social Relations. and Household Production and
Reproduction in the Andes of Southern Peru. In Goodman, AH. and Leatherman,
T. L. (eds.): Building a New Biocultural Synthesis. Political economic
Perspectives on Human Biology. Ann Arbor: University of Michigan Press, pp
245-67.

Lee-Torpe, J. and van der Merwe, NJ. (1987). Carbon Isotope Analysis of Fossil Bone
Apatite. South African Journal of Science 83: 712-15.

Lee-Thorpe, J. and M.J. van der Merwe (1991). Aspects of the Chemistry of Modern and
Fossil Biological Apatites. Journal of Archaeological Science 18: 343-54.

Linkswiler, H.M., C.L. Joyce and CR. Anand (1974). Calcium Retention of Young
Adult Males as Afected by Level of Protein and Calcium Intake. Transactions
New York Academy of Science, Series 2 36: 333-40.

Little, EA. and Schoeninger, M.J. (1995). The Late woodland Diet on Nantucket Island
and the Problems of Maize in Coastal New England. American Antiquity 60:
351-68.

Little, M. (1995). Adaptation, Adaptability, and Multidisciplinary Research. In Boaz,
NT. and Wolfe, L.D. (eds.): Biological Anthropology. The State of the Science.
Corvallis: Oregon State Unversity Press, pp 121-47.

145

Little, M., N. Dyson-Hudson, R. Dyson-Hudson, J .E. Ellis, K.A. Galvin, P.W. Leslie and
D.M. Swifi (1990). Ecosystem Approaches in Human Biology: Their History and
a Case Study of the South Turkana Ecosystem Project. In Moran, E.F. (ed.): The
Ecosystem Approach in Anthropology: From Concept to Practice. Ann Arbor:
University of Michigan Press, pp 389-434.

Loth, SR and Henneberg, M. (1996). Mandinular Ramus F lexure: A New
Methodological Indicator of Sexual Dirnorphism in the Human Skeleton.
American Journal of Physical Anthropology, 99: 473-85.

Lumholtz, Carl and Hrdilcka, A. (1898). Marked Human Bones from a Prehistoric
Tarasco Burial Place in the State of Michoacan, Mexico. Bulletin of the
American Museum of Natural History 10: 61-79.

Macias Goytia and Vackimes Serret A. (1989). Las Turquesas de un Lago. In Garcia
Moll, R. and Garcia Cook, A (eds.): Homenaje a Roman Pifia Chan. Coleccidn
Cientifica, Serie Arqeologia, No. 187. Mexico: INAH, pp 41-71.

Macias Goytia, A. (1990). Huandacareo: Lugar de Juicios y Tribunal. INAH,
Coleccién Cientifica, No. 222, Mexico.

Marshall, F. (1986). Implications of Bone Modification in a Neolithic Faunal.
Assemblage for the Study of Early Hominids Butchery and Subsistence Practices.
Journal of Hmnan Evolution 15: 661-72.

Martin, BL. (1998). Owning the Sins of the past: Historical Trends, Missed
Opportunities, and New Directions in the Study of Human Remains. In
Goodman, AH. and Leathennan, T. L. (eds.): Building a New Biocultural
Synthesis. Political economic Perspectives on Human Biology. Ann Arbor:
University of Michigan Press, pp 171-90.

Martin, D.L., A.H. Goodman and 6.1. Armelagos (1985). Skeletal Pathologies as
Indicators of Quality and Quantity of Diet. In Gilbert, Jr. RI. and Mielke, H.
(eds.): The Analysis of Prehistoric Diets. Orlando: Academic Press, pp 227-79.

McCosh, R., H.P. Pollard and A. Hirshman (1999). Paper presented in the symposium
Beyond ethnohisotry: The Archaeology of Prehispanic Michoacan,Mexico.
Annual Meetingof the Society for American Archaeology, Chicago, March.

146

Meindl, R.S., Lovejoy, C.O., Mensforth,, RP. and Carlos, L.D. (1985). Accuracy and
Direction on Error in Sexing of the Skeleton: Implications for Paleodemography.
American Journal of Physical Anthropology 68: 70-85.

Mintz, SW. (1996). Tasting Food, Tasting Freedom. Excursions into Eating, Culture,
and the Past. Boston: Beacon Press.

Moore, K., M. Murray and M.J. Schoeninger (1989). Effects of Preservatives on Trace
Element and Stable Isotope Analysis of Bone. Journal of Archaeological Science
16:437-46.

Noli, D. and Avery, G. (1988). Protein Poisoning and Coastal Subsistence. Journal of
Archaeological Science 15: 395-401.

Norr, L. (1991). Nutritional Consequences of Prehistoric Subsistence Strategies in Lower
Central America, Ph. D. Dissertation, University of Illinois, Urbana.

O’Hara, S. (1991). Late Holocene Environmental Change in the Basin of Patzcuaro,
Michoacan, Mexico. Unpublished D.Phil. Thesis, University of Oxford.

O’Hara, S. (1993). Historical Evidence of Fluctuations in the Level of Lake Patzcuaro,
Michoacan, Mexico Over the Last 600 Years. The Geographical J oumal 159: 51-
62.

O’Hara, S. S. Metcalfe and A. Street-Perrot (1994). On the Arid Margin: The
Relationship between Climate, Humans and the Environment. A Review of
Evidence from the Highlands of Central Mexico. Chemosphere 29: 965-81.

Ortiz de Montellano, BR. (1990). Aztec Medicine, Health and Nutrition. New
Brunswick: Rutgers University Press.

Ortner, DJ. and Putschar, W.G.J (1981). Identification of Pathological Conditions in
Human Skeletal Remains. Smithsonian Contributions to Anthropology 28.

Ortner, DJ. (1992). Skeletal Paleopathology. Probabilities, Possibilities and
Irnpossibilities. In Veran, J .W. and Ubelaker D.H. (eds) Disease and
Demography in the Americas. Washington, DC: Smithsonian Institution Press,
pp 5-13.

147

Parkington, J. (1991). Approaches to Dietary Recosntruction in the Western Cape: Are
You What You Have Eaten? Journal of Archaeological Science 18: 331-42.

Pereira, G. (1992). Problemes Relatifs au F onctionnement d’une Sépulture Collective: le
Cas de la Structure F unéraire 1 de Guadalupe (Michoacan, Mexique). Tesis de
DE. A., Laboratoire d”Anthropologie, Université de Bordeax 1. ‘

Phenice, T. (1969). A Newly Developed Method of Sexing in the Os Pubis. American
Journal of Physical Anthropology 30: 297-301.

Pickering, RB. (1996). El Significado de las Cubiertas de Crisalidas de Insectos que
Aparecen en las Figurillas del Occidente de Mexico. In Avila, et al. (eds.): El
Occidente de Mexico: Arqueologia, Historia y Medio Ambiente. Guadalajara:
Universidad de Guadalajara, ORSTOM, pp 333-43.

Pifia Chan, Roman. and 01, K. (1982). Exploraciones Arquelogicas en Tingambato,
Michoacan. INAH, Mexico.

Pollard, HP. (1982). Ecological Variation and Economic Exchange in the Tarascan
State. American Ethnologist 9: 250-68.

Pollard, HP. (1993). Tariacuri’s Legacy: The Prehispanic Tarascan State. Norman:
Oklahoma Press.

Pollard, HP. (1995). The Transformation of Regional Elites in Central Michoacan.
Paper presented to the Symposium: The Classic to Postclassic Transition in
Western Mexico. Annual Meeting of the Society for American Archaeology,
Minneapolis, MN, May.

Pollard, HP. (1997). Recent Research in West Mexican Archaeology. Journal of
Archaeological Research 5: 345-84.

Pollard, HP. (1999). The Tarascan Empire in the Mesoamerican World-System. Paper
prepared for the conference: The Postclassic Mesoamerican World-System, held
at the Kellogg Center, Michigan State University, April 17-23, 1999. Organized
by M. E. Smith and F. Berdan.

148

Pollard, HP. (2000) Inforrne Final del Proyecto Los Senorios Urichu, Xarécuaro y Pareo:
Desarrollo del Estado Tarasco. Archivo del Consejo de Arqeologia, INAH,
Mexico City, Mexico.

Puaux, Olivier (1989) Les Practiques firnéraires Tarasques (Approache Archéologique et
Etnohistorique). Theses de Doctorat. Université de Paris 1, Paris.

Relacion de Michoacan (1541) (RM) La Relacion de las Ceremonias y Ritos y Poblacién
y Gobiemo de Michoacén (1980). La Relacidn de Michoacén . Version
Paleografica, Separacion de Textos, Ordenacién Coloquil, Estudio Preliminar y
Notas de F. Miranda, Estudios Michaocanos V. Fimax, Morelia, Michaocan,
Mexico.

Relaciones Geograficas (1578-1581) (RG). Relaciones Geograficas del Siglo XVI:
Michoacan. Serie Antropolégica 74. Edicion de René Acufla, UNAM, Insituto
de Investigaciones Antropolégicas, Mexico.

Romero Molina, Javier (1986) Catalogo de la Coleccion de Dientes Mutilados
Prehispanicos. Coleccion Fuentes, V014. INAH, Mexico.

Rose, J .C. and Harmon, A. (1986). Enamel Microwear and Prehistoric North American
Diets. American Journal of Physical Anthropology 69: 257.

Roseberry, W. (1988). Political Economy. Annual Review of Anthropology 17: 161-85.

Roseberry, W. (1998). Political Economy and Social Fields. In Goodman, AH. and
Leatherman, T. L. (eds.): Building a New Biocultural Synthesis. Political
economic Perspectives on Human Biology. Ann Arbor: University of Michigan
Press, pp 75-91.

Rubin de la Borbolla (1941) Exploraciones Arqueolégicas en Michoacan: Tzintzuntzan
Temporada HI. Revista Mexicana de Estudios Amtropolégicos 5: 5-20.

Saitta, DJ. (1998). Linking Political Economy and Human Biology: Lessons from North
American Archaeology. In Goodman, AH. and Leathennan, T. L. (eds.):
Building a New Biocultural Synthesis. Political economic Perspectives on
Human Biology. Ann Arbor: University of Michigan Press, pp 127-46.

149

Schoeninger, M.J. (1985). Trophic Level Effects on l5N/‘"N and l3C/‘ZC Ratios in Bone
Collagen and Strontium Levels in Bone Mineral. Journal of Human Evolution 14:
5 15-25.

Schoeninger, M.J. and Moore, K. (1992). Bone Stable Isotope Studies in Archaeology.
Journal of World Prehistory 6: 247-96.

Schurr, M. and Schoeninger, M.J. (1995). Associations between Agricultural
Intensification and Social Complexity: An Example from the Prehistoric Ohio
Valley. Journal of Anthropological Archaeology 14: 315-39.

Schwarcz, H. P. (1991). Some Theoretical Aspects of Isotope Paleodiet Studies. Journal
of Archaeological Science 18: 261-275.

Schwarcz, H. P. (in press). Some Biochemical Aspects of Carbon Isotopic Paleodiet
Studies. In Ambrose, SH. and Katzenberg, M.A. (eds.): Biogeochemical
Approaches to Dietary Analysis. New York: Plenum Press.

Schwarcz, H.P. J. Melbye, M.A. Katzenberg and M. Knyf (1985). Stable Isotopes in
human Skeletons of Southern Ontario: Reconstructing Paleodiet. Journal of
Archaeological Science 12: 187-206.

Schwarcz, HP. and Schoeninger, M.J. (1991). Stable Isotope Analyses in Human
Nutritional Ecology. Yearbook of Physical Anthropology 34: 283-322.

Sealy, J. (1986). Stable Carbon Isotopes and Prehistoric Diets in the South-westem Cape
Province, South Africa. BAR International Series 293.

Sealy, J .; van der Merwe, N.J.; Lee-Thorpe, J.A.; and Lanham, J .L. (1987). Nitrogen
Isotopic Ecology in Southern Afi'ica: Implications for Environmental and Dietary
Tracing. Geochimica et Cosmochimica Acta 51:2707-17.

Sepr’rlveda y H., Ma.T. (1988). La Medicina entre los Purépecha Prehispanicos. Serie
Antropolégica 94. UNAM, Mexico.

Siegel, S. (1956). Nonparametric Statistics for the Behavioral Sciences. New York:
McGraw-Hill Book Co.

150

Sillen, A. (1986). Biogenic and Diagenetic Sr/Ca in Plio-Pleistocene Fossils of the Omo
Shungura Formation. Paleobiology 12: 311-23.

Smith, B.N. and Epstein, S. (1971). Two Categories of 13C/ 12C Ratios for Higher Plants.
Plant Physiology 47: 380-84.

Speth, J .D. (1983). Bison Kills and Bone Counts. Chicago: University of Chicago Press.

Spielrnann, K.A. (1990). Plains-Pueblo Interdependence and Human Diet at Pecos
Pueblo, New Mexico. American Antiquity 55: 745-65.

Spurr, GB. (1990). The Impact of Chronic Undemutrition on Physical Work Capacity
and Daily Energy Expenditure. In Harrison, GA. and Waterlow, J .C. (eds.): Diet
and Disease in Traditional and Developing Societies. Cambridge: Cambridge
University Press, pp 5-23.

St. Hoyme, LE. and Iscan, M.Y. (1989). Determination of Sex and Race: Accuracy and
Assumptions. In Iscan, M.Y. and Kennedy, K.A.R. (eds.): Reconstruction of Life
from the Skeleton. New York: Alan R. Liss, pp 53-93.

Stafford T.W., Hare, P.E., Currie, L, Jull, A.J.T., and Donahue, DJ. (1991). Accelerator
Radiocarbon Dating at the Molecular Level. Journal of Archaeological

Stenhouse, M.J. and MS. Baxter (1979) The Uptake of Bomb MC in Humans. In
Berger, R and H. Suess (eds.): Radiocarbon Dating. Berkley: University of
California Press, pp 324—341.

Street-Perrot, RA. and Harrison, SP. (1985). Lake Levels and Climatic Reconstruction.
In Hecht, A.D. (ed.): Palaeoclimate Analysis, and Modeling. New York: John
Wiley and Son, pp 291-340.

Szathmary, E.J.E., C. Ritenbaugh and CS. Goodby (1987). Dietary Change and Plasma
Glucose Levels in an American Population Undergoing Cultural Transition.
Social Science and Medicine 24: 791-804.

Thomas, RB. (1998). The Evolution of Human Adaptability Paradigms: Toward a
Biology of Poverty. In Goodman, AH. and Leathennan, T. L. (eds.): Building a
New Biocultural Synthesis. Political economic Perspectives on Human Biology.
Ann Arbor: University of Michigan Press, pp 43-73.

151

Tieszen, LL. (1991). Natural Variation in the Carbon Isotope Values of Plants:
Implications for Archaeology, Ecology, and Paleoecology. Journal of
Archaeological Science 18: 227-48.

Tieszen, LL. and Fagre, T. (1993). Effect of Diet Quality and Composition on the
Isotopic Composition of Respiratory C02, Bone Collagen, Bioapatite and Sofi
Tissues. In Lambert, J .B. and Grupe, G. (eds.): Prehistoric Human Bone:
Archaeology at the Molecular Level. Berlin: Springer-Verlag, pp 121-55.

Ulijaszek, SJ. (1995). Human Energetics in Biological Anthropology. Cambridge:
Cambridge University Press.

Urufluela y Ladron de Guevara, G. (1998). Organizacién Social, Medio Ambiente y
Enfermedad: Identificacién de Treponematosis en un Individuo de la F ase Sayula
en Atoyac. In Avila, et al. (eds.): El Occidente de Mexico: Arqueologia, Historia
y Medio Ambiente. Guadalajara: Universidad de Guadalajara, ORSTOM, pp
363-74.

Urufiuela y Ladron de Guevara, G. (1996). E1 Cuerpo Humano y su Adaptacién a1
Medio: 1a Poblacion Prehispanica de Atoyac, Jalisoco. In Williams, E. and
Weigand, P. (eds.): Las Cuencas del Occidente de Mexico (Epoca Prehispénica).
Zamora: El Colegio dc Michoacan, ORSTOM, CEMCA, pp 427-54.

Valdez, F. and Liot, C. (1994). La Cuenca de Sayula: Yacimientos de Sal en la Frontera
Oeste del Estado Tarasco. In Larneiras, B. (ed.): El Michoacan Antiguo.
Morelia: El Colegio de Michoacan And E1 Gobiremo del Estado de Michoacan,
pp 285-305.

Valdez, F. and Liot, C. and Schondube, O. (1996). Los Recursos Naturales y su Uso en
las Cuencas Lacustres del Sur de Jalisco: E1 Caso de Sayula. In Williams, E. and
Weigand, P. (eds.): Las Cuencas del Occidente de México (Epoca Prehispanica).
Zamora: E1 Colegio de Michoacan, ORSTOM, CEMCA, pp 325-66.

van der Merwe, NJ. (1982). Carbon Isotopes, Photosynthesis, and Archaeology.
American Scientist 70: 596-606.

Vogel, J .C. and van der Merwe, NJ. (1977). Isotopic Evidence for Early Maize
Cultivation in New York State. American Antiquity 42: 238-42.

152

Walker, Phillip L. (1995) Problems or Preservation and Sexism in Sexing: Some Lessons
from Historical Collections fro Palaeodemographers. In Sanders, SR. and
Herring, A. (eds.): Grave Reflections: Portraying the Past through Cemetery
Studies. Tronto: Canadian Scholars Press, pp 31-47.

Wallerstein, I. (1977). Rural Economy in the Modern World-Society. Syudies in
Comparative International Development 12: 29-40.

Warren, B. (1971). F ray J cronimo de Alcala: Author of the Relacidn de Michoacan? The
Americas 27: 307-26.

Warren, B. (1977). La Conquista de Michoacan 1521-1530. Morelia: FIMAX
Publicistas.

Warren, B. (1983). Writing the Language of Michoacan: Sixteenth Century Franciscan
Linguistics. In Morales, F. (ed.): Franciscan Presence in the Americas.
Maryland: Potomac.

Waterlow, J .C. (1990). Mechanisms of Adaptation to Low Energy Intakes. In Harrison,
GA and Waterlow, J .C. (eds.): Diet and Disease in Traditional and Developing
Societies. Cambridge: Cambridge University Press, pp 5-23.

Weiss, KM. (1972). On the Systematic Bias in Skeletal Sexing. American Journal of
Physical Anthropology 37: 239-50.

West, R. (1948). Cultural Geography of the Modern Tarascan Area. Institute of Social
Anthropology 7. Washington, DC: Smithsonian Institution.

White, CD. and Armelagos, G.J. (1997). Osteoporosis and Isotopes in X-Group
Nubians. American Journal of Physical Anthropology 103: 185-99.

White, C.D., Healy, RF. and Schwarcz, HP. (1993). Intensive Agriculture , Social
Status, and Maya Diet at Pacbitun, Belize. Journal of Anthropological Research
49: 347-75.

White, CD. and Schwarcz, HP. (1992). Isotopic Evidence for Regional and Temporal
Distinction in Ancient Maya Diet. Journal of Vertebrate Paleontology 12: 59A
(abstract).

153

White, CD. and Schwarcz, HP. (1989). Ancient Maya Diet: As Inferred from Isotopic
and Elemental Analysis of Human Bone. Journal of Archaeological Science 16:
45 1-74.

White, C.D., M.W. Spence, H.L.Q. Stuart-Williams and HP. Schwarcz (1998). Oxygen
Isotopes and the Identification of Geographical Origins: The Valley of Oaxaca
versus the Valley of Mexico. Journal of Archaeological Science 25: 643-55.

White, T and Folkens (1991) Human Osteology. San Francisco: Academic Press.

Whitmore, T.M. (1992). Disease and Death in Early Colonial Mexico. Simulating
Amerindian Depopulation. Dellplain Latin American Studies, No. 28. Boulder:
Westview Press.

Wright, LE. and Schwarcz, HP. (1996). Infrared and Isotopic Evidence for Diagenesis
of Bone Apatite at Dos Pilas, Guatemala: Palaeodietary Implications. Journal of
Archaeological Science 23: 933-44.

Wolf, E. (1982) Europe and the People without History. Berkeley: University Hof
California Press.

154

IIIIIIIIIIIIIIIIIIIIIIIIIIIIII

111111111musjljjjjjjjljwl

3 1293 0