.. c: Q. . .u. -...‘. ......-4..., "n.“-

 

 

‘
A ‘
' \ v

 

 

 

" "mu -....,-..,..- a

 

 

 

 

'H YT? FEES
".7
5/

{3
£3
C‘\
Ci \

This is to certify that the

thesis entitled

Neglected Artifacts: A Study of Re-Worked Ceramic Sherds
From The Lake Patzcuzro Basin, Mexico

presented by

Catherine Anderson Phillips

has been accepted towards fulfillment
of the requirements for

M.A. Jegree inAnthropology

[M040 ?” “Ala/(£4.40

Major professor

Dr. Helen P. Pollard

Date 5’ ZUZJD/

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

 

 

 

LIBRARY

University

Michigan State

 

 

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

 

rtaogmm

 

94219;

 

FMA" ‘ ‘ M“
1% E 6“?

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIRC/DateDue.p65.p.15

NEGLECTED ARTIFACTS: ,
A STUDY OF RE-WORKED CERAMIC SHERDS FROM THE LAKE PATZCUARO
BASIN, MEXICO
By

Catherine Anderson Phillips

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
MASTER OF ARTS
Department of Anthropology

2002

ABSTRACT
NEGLECTED ARTIFACTS: A STUDY OF RE-WORKED CERAMIC SHERDS
FROM THE LAKE PATZCUARO BASIN, MEXICO
By

Catherine Anderson Phillips

Recent archaeological work in the Lake Patzcuaro Basin in Michoacan, Mexico
has revealed a set of artifacts whose function has heretofore been unexplored. The
artifacts are comprised of roughly circular, re-worked ceramic sherds. Only isolated
references to similar such artifacts are made in the extant literature. This paper examines
the largest known grouping of such artifacts and attempts to ascertain their function and
their place in Prehispanic Tarascan society. Three hypotheses concerning the artifacts’
function are examined: that they are spindle whorls, that they are gaming pieces, and that
they are fishing net weights. Examination of the statistical and ethnographic data suggest

that they functioned as the latter.

Dedicated to my loving and ever-optimistic husband, Doug, and my beautiful
children, Anna and Sean

iii

ACKNOWLEDGMENTS

I would like to thank Dr. Helen Pollard for providing me with the original data set
and opportunity for research, as well as for her unflagging patience and assistance with
the completion of my graduate degree. Her unsurpassed knowledge of and enthusiasm

for the archaeology of Mesoamerica was the inspiration for my post-graduate studies.

iv

TABLE OF CONTENTS

LISTOF FIGURES ... .

INTRODUCTION
Description ofartifacts... .... ....

Objectives ......................................................................

CHAPTER 1

METHODOLOGYANDANALYSIS OF THE DATA...

Methodology...

Statistical Analysis of the Disks ..
Diameter, Thickness, and Weight
Notch and Perforation Morphology

Spatialand Temporal Patterning................m..
Density ofDisks per Site...............
Intra—Site Density andDistribution...
Density ofDisks by Phase......

CHAPTER 2
HYPOTHESES...

Malacate HypOthesis... .

Gaming Piece HypotheSis

Net SinkerHypothesis.........

CHAPTER 3
ARCHAEOLOGICAL CONTEXT...

Ceramic Disks from Prehispanic Memco... - . , ,

Similar Stone Artifacts from North American sites

CHAPTER 4

FISHANDFISHING INTI-IE TARASCAN EMPIRE.........

Fish in the Tarascan Diet and Economy... .. . . ..

FishinMesoamerican Ideology......................
FishinTarascanIdeology............. .. ..

Fishin Tarascan Mythology...... ..

Fish in Tarascan Burials... .. III.)

SUMMARYANDCONCLUSIONS......

APPENDIX (TarascanDiskRawData)... ..

...vii

. ...viii

..... 2

HOOOOQUIUIAA

....16
......16

...25
...29

......34
.....34
...34

...48
....48
...54
...59
.......60
.....63

...67

..73

ILLUSTRATIONS..................

vi

...94

138

LIST OF TABLES

Table l. Phases Representedby Ceramic Disks...

Table 2. Size Parameters forMesoamen'can Disks.........

Table 4. TarascanDisk RawData.........

vii

...11
.....20
Table 3. FishNomenclature.......................... ......48

...74

LIST OF FIGURES

Figure 1. Lake Patzcuaro and surrounding sites......... ..

Figure 2.Prehispanic Tarascan disks............... .. ..

Figure3.Histogramofdiskdiameters(alldisks)...... .. ..
Figure4.Diametersummarystatistics.... ..
Figure 5. Histogram ofdiskthickness (all dlSkS) ..
Figure 6.Thickness summary statistics.........................
Figure7.Histogramofdiskweights(alldisks)
Figure 8. Weightsummarystatistics.............................................................
Figure 9a. Regression plot of disk diameter vs. thickness and weight....................
Figure 9b. Pearson correlations for variables of diameter, thickness, and weight. .. . ..
Figure 10. Notchmorphology(all disks)
Figure11.Notchmorphology(whole dlSkS)
Figure 12. Notchmorphology(brokendisks)..................................................
Figure13.Diskspersurveyzone......... .. ..
Figure 14. Ratioofceramic sherdsperdrsk

Figure 15. Map of Lake Patzcuaro showing exposed lake bed due to lowered water levels

(from O’Hara 1993)..

Figure 16. Map ofXaracuaro units withthe highest disk counts... ..

viii

.95

...96

.97

.98

.99

100

101

102

..103

104

.105

106

107

108

109

1..10

111

Figure 17. Map ofPareo units with the highest disk counts.............. 112

Figure 18.MapofUrichuunits withthehighestdiskcounts................................113
Figure l9.Disksperphase(allsurveyzones)...................................................114
Figure20.Disksperphase(Urichu)............................................................115
Figure21.Disksperphase(Xaracuaro)........................................................116
Figure 22. Map of Xaracuaro units with highest Late Urichu phase disk counts ......... 117
Figure 23. Map ofXaracuaro units with highest Tariacuri phase disk counts... 1 18
Figure24.Disksperphase(Pareo)..............................................................119
Figure 25. Map ofPareo units with highest Tariacuri phase disk counts...'... 120
Figure 26. Map of Pareo units with highest Late Urichu phase disk counts ............... 121
Figure27.Prehispanic spinning(fromFlorentine Codex)...................................122
Figure 283. Prehispanic malacates from central Mexico

(fromParsonsandParsons1990)...............................................................123
Figure 28b. Prehispanic malacates from Cojumatlan (from Lister 1949) .................. 124
Figure 28c. Prehispanic malacates from Zacapu Lake Basin, Michoacan (from Arnauld et
Figure 29. T-test for diameter: Tarascan disks vs. Tzintzuntzan malacates...............126
Figure 30. T-test for thickness: Tarascan disks vs. Tzintzuntzan malacates..............127
Figure 31. T-test for diameter: Tarascan disks vs. Urichu malacates. .. . 128
Figure 32. T-test for thickness: Tarascan disks vs. Urichu malacates. .. 129
Figure 33. T-test for weight: Tarascan disks vs. Urichu malacates... 130
Figure 34. T-test for diameter: Urichu malacates vs. Tzintzuntzan malacates. .. 131
Figure 35. T-test for thickness: Urichu malacates vs. Tzintzuntzan malacates. .. .. 132

ix

Figure 36. Drilled stone disks from Lovelock Cave, Nevada (from Loud and Harrington
1929) .u.u.u.n.u.n.u.u.n.u.u.u. “ “ ..n.”.u.. . ..H.” .133

Figure 37. Fish nets from Lovelock Cave, Nevada (from Loud and Harrington 1929).. 134
Figure 38. Net sinkers from Port Moller, Alaska (from Weyer 1930)... 135
Figures 39a-b. Notched stones from Cook Inlet, Alaska (from De Laguna 1934). .. .. 136

Figures 40a-b. Notched stones from central Texas (from Watt 1938). .. 137

INTRODUCTION

From time to time man discovers, along his pathway in life, some simple

object from out of the remote past, small in size and insignificant in aspect

that heretofore he has consistently overlooked; suddenly he becomes

conscious of it, takes note of and acquires a deep interest in, only to find

as he studies and ponders over the various phases involved that it becomes

more and more an object of elusive determination as to what may be its

purpose, and where it fits into the life of some people (Watt 1938:21).
Description of Artifacts

The subject of this thesis is a previously undefined class of artifacts, ceramic
disks, from the Lake Patzcuaro Basin in Michoacan, Mexico. This artifact set was
collected from numerous sites located along the southwest comer of Lake Patzcuaro
under the direction of Professor Helen Pollard as part of the Development of the Tarascan
State: the Urichu, Xardcuaro, and Pareo Polities Project. Formal excavations were
conducted largely at the site of Urichu but also on the island of J aracuaro (known in
Prehispanic times as Xaracuaro) during the summers of 1990, 1991, 1994, and 1995.
Additionally, numerous site surveys were conducted from February through June of 1996
in the areas surrounding Urichu, Xaracuaro, and Pareo. The project area lies at the heart
of the Prehispanic Tarascan Empire, and the encountered occupation levels span the
Classic through Post-Contact periods (c. AD. 350-1525). (See Figure 1)

The artifacts themselves are re-worked ceramic sherds which have been fashioned
into roughly circular shapes (Figure 2). Some are rather squarish but most are circular.
In addition, most of the ceramic disks have either one or two notches. Only one is

perforated, or drilled, and one has grooves carved into it. Both broken and whole ceramic

disks, 522 in all, were recovered.

The function of this artifact group has been heretofore unclear. There are several
brief mentions in the literature of similar artifacts from other Prehispanic Mexican sites,
but they are always listed as “unknowns” or “miscellaneous.” Some researchers have
made informal suggestions as to the disks’ functional purpose, including their use as
spindle whorls and fishing net weights. Macias Goytia (1990283), for example,
speculates that, although the function of the pieces is not known, their discovery in great
numbers and in riverine contexts suggests that they could have been used as weights for
the nets used to catch charal, a term which refers collectively to several species of

Chirostoma, including Ch. grandocule, Ch. attenuatum, and Ch. patzcuaro.

Objectives

To date, however, no one has undertaken a formal analysis of the ceramic disks.
This thesis is the first formal analysis of these artifacts ever undertaken. The large
sample size (the largest documented to date) and the detail available regarding their
provenience and size measurements makes this collection of Tarascan disks ideal for
performing a formal analysis.

This thesis has several objectives. The first objective is to determine which of the
proposed functions, if any, can be correctly assigned to the artifacts. As stated above, to
date, these functions have only been put forth informally. This analysis will formally test
three hypotheses concerning the disks’ function(s). This will be achieved by spelling out
specific criteria which should be met to validate each particular hypothesis, and then
using a variety of methods and lines of evidence to support or reject that hypothesis. The

analysis will make use of elementary statistical techniques as well as ethnographic data.

The three hypotheses to be considered concerning the disks’ true function are 1) that the
disks are spindle whorls, or malacates, as they are known in Spanish; 2) that the disks
were gaming pieces of some sort; and 3) that the disks were fishing net weights, or net
sinkers. A thorough examination of the available data points to the third hypothesis as
the most likely explanation of the artifacts’ true function.

A further objective of this thesis is to clarify the artifacts’ place in Tarascan
society. Since the disks (net sinkers) represent an activity (fishing) which was central to
the Tarascan economy, the set of disks examined in this paper can be viewed as a proxy
for the Tarascan economy itself. They can be analyzed for spatial and temporal changes
which might reveal developments or fluctuations in the economy of the Tarascans. The
cultural and ideological associations of fish and fishing in Prehispanic Tarascan society
will also be explored.

In addition, the Tarascan disks will be compared with similar artifacts reported in
archaeological research from other areas in native North America, from both North
American and Mesoamerican cultures.

Finally, suggestions and criteria for identifying these artifacts in the existing

archaeological record and for future excavations will be set forth.

CHAPTER 1
METHODOLOGY AND ANALYSIS OF THE DATA
Methodology

Two basic methods of analysis will be employed in exploring the function of the
ceramic disks: statistical analysis and ethnographic analysis. Unfortunately, certain lines
of testing, such as use/wear analysis or examination of the ceramics’ paste or firing
hardness, are eliminated a priori because the artifacts are currently housed in their home
country of Mexico. This author has access only to measurements which were made at the
time of collection and initial examination. Fortunately, however, much can be done with
the available measurements, and it is believed this is sufficient to illuminate the artifacts’
function

Ethnographic analysis will comprise the other main method of examining the
data. Ethnographic data collected from the late nineteenth century through the mid-
twentieth century are available for all topics relevant to this paper: malacates (Parsons
and Parsons 1990; M. Parsons 1972; M. Parsons 1975), games and gaming pieces (Foster
1948; Bennett and Zingg 1935), and net sinkers (Foster 1948; Pollard, personal
communication). Good ethnographic accounts of fishing practices in native North
America (Meehan 1893; Rostlund 1952) and Mexico (Beals 1932; Bennett and Zingg
1935), including the Tarascans (Foster 1948) also exist. In this way ethnographic data
can be compared to the statistical findings. Where the two methods concur, they will
lend greater credence to a particular hypothesis.

It should be re-iterated that the hypotheses presented and tested in this paper are

not original to the author; they have been informally proposed elsewhere. Rather, it is the

first formal testing of these hypotheses which is the original contribution of the work

presented herein.

Statistical Analysis of the Disks

Statistical analysis of the ceramic disk assemblage will form a major portion of
this investigation. As Whallon posits in the opening of his “Simple Statistics” (19872135),
. . simple, descriptive statistics and display techniques are indispensable preliminaries to
the application of even the most basic inferential statistics or tests.” He further asserts
that . . in almost every instance, one can learn more, more quickly, more clearly, and in
more detail about one’s data with these (simple) techniques than through the use of
inferential statistics or tests”. Very basic, summary statistics will therefore be calculated.
For example, the means and range distributions of the disks’ weight, diameter, and
thickness will be determined and graphed. Notch and perforation morphology will also

be analyzed.

Diameter, thickness, and weight

A total of 522 ceramic disks was recovered. Histograms for the characteristics of
diameter, weight, and thickness all show positively skewed, unimodal distributions with
the preponderance of values concentrated (Figures 3, 5, and 7).

A histogram of the disks’ diameters exhibits a roughly unimodal distribution
(Figure 3), with a mean diameter of 30.7 mm, a median score of 28.2, a mode at 25.2
mm, and minimum and maximum scores of 14.9 mm and 59.6 mm respectively (Figure

4).

A histogram of the disks’ thickness shows a similarly shaped distribution (Figure
5), with a mean value of 5.7 mm, a median value of 5.1 mm, a mode at 5.2 mm, and
minimum and maximum scores of 2.7 mm and 19.5 mm respectively (Figure 6).

A histogram of the disks’ weight scores also demonstrates this same distribution
shape, with the bulk of the scores positively skewed about the mean and a gradual tailing
off of higher weight disks, with a few, much heavier, outliers (Figure 7). The mean
weight of the disks is 6.2 grams, with a median score of 4.2 grams and a mode at 2.4
grams. There is a wide range of weights, with the lightest a mere 0.4 grams and the
heaviest weighing in at 35 grams (Figure 8).

Although the mean, median, and mode of each of the three measurements are
close in value, the tightest distribution is that of thickness. This could indicate a high
degree of standardization in Tarascan pottery, out of which the disks were manufactured.
Although thickness can plausibly be viewed as an unselected-for variable resulting from
the re-use of ceramic pieces, it seems likely that weight and/or diameter were being
consciously selected for, with a range of variation centered around an ideal value.

For example, more than a third of all diameter values for the disks fall within the
23-29 mm range (see Figure 3). It is quite possible that the disk makers were selecting for
weight indirectly and used diameter to control for this, depending on the thickness of the
sherd being recycled. In fact, weight distribution is even more strongly peaked, or
leptokurtic, than that of diameter, with 52% of all disk scores falling in the 5-8 gram
range (see Figure 7). Of the variables thickness, weight, and diameter, it is the latter two

which are most strongly correlated, with a Pearson correlation coefiicient (r) of 0.78

(Figures 9a-b). This is not a random pattern and indicates a clear selection on the part of

the disk makers for weight and size.

Notch and perforation morphology

Comparing the frequencies of notch classes is also informative. Out of the 517
whole and broken disks with unambiguous notch morphologies, 66% have two notches
(Figure 10). A pie chart of only the whole disks displays this patterning even more
clearly, with 85% of disks having two notches (Figure 11). In contrast, this regularity is
much less clear in a chart showing only the broken disks (Figure 12). Furthermore, the
extremely low percentage of single notching among whole disks (just 1%) argues against
its being considered a legitimate class. It seems likely that two notches was the intended
norm. This morphology would certainly facilitate tying the disks to a net. Since 80% of
all disks and 86% of whole disks have some notching, perhaps the unnotched ones were
blanks to be finished at a later time.

Only one disk was perforated, or drilled. The rarity of this form argues against it,
too, being considered a legitimate class or the intended final form. It may or may not
have been used in the same capacity as the other Tarascan disks. Its diameter, thickness,
and weight measurements (37.1 mm, 2.8 mm, and 3.5 g respectively) do fall within the
parameters of at least some known malacates, or spindle whorls (see below).

Three disks had grooving on them. Again, they are the exception and not the rule.
Interestingly, these three disks are much larger than most of the other disks. Their mean
diameter, thickness, and weight measurements are much greater than the mean

measurements for the rest of the disks: 48.9 mm (diameter), 15.6 mm (thickness), and 26

grams (weight), vs. means of 30.7 mm (diameter), 5.7 mm (thickness), and 6.2 grams
(weight) for the rest of the disks. Although the sample size of the grooved disks is
extremely small it is at least possible that they are in fact a different kind of artifact and
were used in a different capacity than the other ceramic disks. It is equally plausible that
they were originally manufactured to function in a different capacity but were‘later re-

used as net sinkers.

Spatial and Temporal Patterning

In addition to analyzing the disks’ inherent characteristics, statistical techniques
can also be applied to their frequency distribution in time and space. All of the disks
have proveniences, and many of them can be dated to a particular cultural period. Disks
which were recovered during surveys rather than excavations could not always be
definitively assigned to a particular period, e.g. if they were found without other
diagnostic ceramic types. It should also be noted that if the disks were used as net sinkers,
disks which were recovered from units which were under water dming the Tariacuri
phase may have either dropped off Tariacuri phase fishing nets and settled onto the lake
bottom, or may in fact date from the earlier Late Urichu phase. In this case it is not
possible to assign the disks to either phase with complete certainty. Other factors which
can be analyzed include the density of ceramic disks at each site, their intra-site density
and distribution, their density as a ratio of other ceramic artifacts, and their density by

phase/date.

Density of disks per site

The areas around Urichu, Pareo, and Xaracuaro all yielded ceramic disks.
Seventy-three (14%) were recovered from the environs of Urichu, 324 (62%) from
Xaracuaro, and 124 (24%) from Pareo (Figure 13). Their relative density can be
quantified as a proportion ofall other (non-disk) ceramic sherds. These data are
summarized in Figure 14. The zone with the lowest proportion of ceramic sherds to
ceramic disks therefore has the highest density of ceramic disks relative to all other
ceramics recovered Pareo has the highest density of disks to sherds, with a ratio of 1:42.
Xaracuaro comes in second, with a ratio of 1:61. Urichu comes in at a distant third, with
a ratio of 1:1,389.

The density figures for the three zones is revealing. Urichu, with the highest
number of sherds per disk, is furthest from the lakeshore. The ratios are much lower for
Xaracuaro, which was in fact an island at the time of Late Postclassic occupation, and for
Pareo, which was even closer to the lakeshore then than it is at present, due to lowered
lake levels in recent times (Gorenstein and Pollard 1983; O’Hara 1993). According to
O’Hara, the waters of Lake Patzcuaro at the time of Conquest stood at between
approximately 2041.5 and 2045.5 meters above modern sea level. This is between five
and eight meters higher than at present (O’Hara 1993:54-55) (Figure 15). Pollard puts
the current lake level at 2033 meters above sea level, 12.5 meters lower than at Conquest
(personal communication). This line of evidence is entirely consistent with the hypothesis
that the ceramic disks were used as net sinkers, or at least for lake-related activities, for
one would expect their numbers to increase directly with proximity to the lakeshore. In

fact, it is likely that many of the disks recovered from (now dry) Xaracuaro and Pareo

units were underwater during the immediate pre-Conquest period. This would account
for the low numbers of other (non-disk) ceramics found in these areas compared with
Urichu. It also supports the net sinker hypothesis, for one would expect to find net

sinkers, but not other types of ceramic artifacts, in areas that were previously underwater.

Intra-site density and distribution

We can examine the density and distribution of the ceramic disks within each
zone both in absolute numbers and as a ratio to other ceramic artifacts recovered in hopes
of discovering some sort of pattern.

In terms of absolute numbers, mapping the formal survey units with the top five
scores for numbers of disks reveals a telling pattern In Xaracuaro, all but one of the
units with the most disks are clustered in the eastern part of the zone (see Figure 16).
Since Xaracuaro was an island at time of contact and is a peninsula today, all areas of the
zone can be considered to be in close proximity to the lakeshore. In this respect one can
draw no special conclusions regarding the spatial patterning of the top—ranking units.
However, the survey units with the most disks are clustered together in the same general
area, and it is possible that this area, which was under the waters of Lake Patzcuaro
during the time of the Tarascan empire, was a prime fishing area.

In Pareo, all of the top—ranking units are clustered along the northern part of the
zone, adjacent to the lakeshore (Figure 17). This pattern is even more striking in Pareo
because the zone covers such a large area. None of the southem-most units yielded any

disks at all. Only three areas in Urichu were excavated; all three yielded ceramic disks.

10

Again, these three areas are clustered together, but in this case, they are not adjacent to

the lakeshore. Instead, they are in the westem-most part of the zone (Figure 18).

Density of disks by phase

It was possible for Pollard’s research team to date most of the excavated disks
based on the ceramic typologies of other ceramic pieces found in the same levels. It
should be noted that the (style) type of the disks themselves, even if discernable, is not
adequate for dating because the disks are a secondary use artifact; that is, they were made
from sherds of pottery pieces which were originally manufactured and used for entirely
different (primary) purposes.

If the ceramic disks are net sinkers, we should expect that the ones recovered
from now-dry areas around Pareo and Xaracuaro to date to phases when lake levels were

significantly higher than they are today.

(phase table adapted from Pollard and Cahue 1999:262)

Table 1. Phases Represented by Ceramic Disks

 

 

Period , Local Phase Date

Late Postclassic Tariacuri A.D. 1350-1525
Middle Postclassic Late Urichu A.D. 1100-1350
Early Postclassic Early Urichu AD. 900/ 1000-1 100
Epiclassic Lupe-La Joya A.D. 600-900/ 1000
Classic Loma Alta 3/ Jaracuaro A.D. 350-600

 

Two hundred and forty-eight, or almost half (47%), of the disks date to the Late
Urichu phase (AD. 1100-1350) (Figure 19). The Tariacuri, and then Early Urichu,

phases are the next most numerous. An additional 63 disks could be dated to the Urichu

11

 

phase, but not specifically enough to be dated Early or Late. This means that 350, or
75%, of the disks which could be definitively assigned to a particular phase can be
assigned to either the Early or Late Urichu phases.

Consistent with the spatial data, the site of Urichu seems to break with this
pattern. Interestingly, none of the disks recovered from Urichu dates to the Late Urichu
phase (Figure 20). Pollard suggests that this is due to limited excavations in Area 2, the
only part of Urichu with Late Urichu deposits (personal communication 2002). Most of
the disks from Urichu date from the Tariacuri phase (n=31), Early Urichu (n=18),

J aracuaro (n=10), and Lupe-La Joya (n=9). The large number of Tariacuri phase disks
reflects the increased population density at Urichu during this time (Pollard, personal
communication 2002).

Xaracuaro yielded disks from all of the phases (Figure 21). By far, most disks
from this zone (n=178) date to the Late Urichu phase, representing 55% of the disks from
Xaracuaro. This is consistent with the fact that Xaracuaro experienced its highest
population density during the Late Urichu phase. By this time, the island was home to
full-time fishermen. The Urichu (n=61), Tariacuri (n=25), and Lupe-La Joya (n=l6) are
the next most numerously represented phases. Maps of the units which yielded the most
disks at Xaracuaro for each of the Late Urichu and Tariacuri phases (see Figures 22 and
23) show a definite shift in disk distribution within the site. Disks dating to the Late
Urichu phase are concentrated in the eastern half of the site. At this time, this area was
above water. In contrast, by the Tariacuri phase, when the eastern part of Xaracuaro was
submerged under water, the highest concentration of disks has shifted westward. The

disks dating to the Tariacuri phase which are located in the western portion of Xaracuaro

12

thus most likely reflect household areas where fishing nets were manufactured and
stored The disks which were found in Late Urichu phase deposits, on the other hand,
may represent both a household pattern of Late Urichu fishing net manufacture and
storage and later Tariacuri phase plainware sherds which were used as net sinkers and
dropped off Tariacuri phase fishing nets onto the earlier, Late Urichu phase deposits
below (Pollard, personal communication 2002).

Pareo yielded no disks at all from the J aracuaro or Lupe-La Joya phases (Figure
24). As at Xaracuaro, over half of the disks (n=70, 56%) from Pareo date to the Late
Urichu phase. Most of the remaining disks from this zone date to the Tariacuri (n=26)
and Early Urichu (n=12) phases. Again, mapping the highest disk concentrations by
phase is informative. Disks dated to Tariacuri phase deposits occur in high
concentrations in the western, central, and eastern part of the zone (see Figure 25).
Pollard (personal communication 2002) associates those in the western part of the site
with fields outside the Tariacuri phase village of Ardcutin. They were commoner
residential areas bordering the lakeshore, and likely were associated with fishermen
households. Likewise, the area to the east bordered both the Tariacuri phase town of
Pareo and the lakeshore. These areas too were likely either fishermen’s households or
ideal places to repair nets (as they are now). Pareo had a well-known market and it is
likely that fish caught here were exchanged via the market system. The small units in the
center of the Pareo map represent the only areas which were above water during the high
lake levels of the Tariacuri phase. They thus represent areas where fishing households or
storage/repair areas were located. High concentrations of disks associated with Late

Urichu phase deposits, on the other hand, are confined to the central portion of the Pareo

13

area map (see Figure 25). As at Xaracuaro, disks found here may be associated either
with fishing households or fishing net manufacturing/repair areas of the Late Urichu
phase, or with later, Tariacuri phase nets which lost sinkers to the earlier deposits below.
In light of the fact that there was a substantial population increase in this area during the
Tariacuri phase, it is interesting to note that although high concentrations of disks are
found in more areas of Pareo during the Tariacuri phase, larger numbers of them were
actually associated with Late Urichu deposits, when the population was lower. However,
if many of the disks which were found in Late Urichu deposits actually fell off of the nets
of Tariacuri phase fishermen, then the numbers of net sinkers associated with each of
these periods may more accurately reflect area population.

The temporal data suggests a sharp increase in the relative frequency of ceramic
disks during the Late Urichu phase. Even if all the disks which were generally assigned
to the Urichu phase in fact dated from the Early Urichu phase, which is unlikely, there
would still be a greater than 100% increase in disks in the Late Urichu phase. If, on the
other hand, all 63 Urichu phase disks are in fact fiom the Late Urichu phase, the Late
Urichu phase increase could be as high as eight-fold. In all likelihood the actual increase
lies somewhere between these two extremes.

Since the level of Lake Patzcuaro was lower during the Late Urichu phase and
higher during the subsequent Tariacuri phase (O’Hara 1993), two possibilities exist for
the disks dated to the Late Urichu phase which were found in areas which were under
water during the Tariacuri phase. Either they do in fact date from the Late Urichu and are

artifacts of shoreline settlements, or they may have dropped off the nets of later, Tariacuri

14

phase fishermen onto the lake bottom in areas that were previously inhabited during the

Late Urichu phase of lower lake levels.

15

CHAPTER 2
HYPOTHESES

Three basic hypotheses concerning the ceramic disks’ function(s) will be
presented and tested. Each hypothesis entails certain specific criteria to which the data
should conform in order to support it. The basic data set, which consists of a set of the
disks’ measurements and notch morphologies, as detailed above, will be analyzed in
order to glean any inherent patterning. The artifacts’ distribution in time and space will
also be examined. Thus, the ceramic disks’ formal characteristics as well as their spatial

distributions can be compared to those of the artifacts represented in each hypothesis.

Malacate Hypothesis

The first hypothesis to be considered is that the disks functioned as spindle
whorls, or malacates, as they are known in Mexico. This hypothesis has been previously
presented in other site reports but not tested formally. Malacates are essentially a type of
weight, attached to a spindle, used to help spin fiber of various sorts into yarn, thread, or
string (Figure 27). Malacates are ubiquitous in Mesoamerican societies and can therefore
easily be compared to the ceramic disks. Their morphology is fairly uniform, and can be
compared to that of the ceramic disks. If the two groups of artifacts differ significantly in
shape, size, or distribution, then it is unlikely that they represent a single class or type, or
that they performed the same function.

The spatial distribution of both artifact groups can also be compared. Since it is
well documented that spinning and weaving fibers was an activity performed by women

in Mesoamerica, one can reasonably expect that malacates will be concentrated in or near

16

household areas. If the ceramic disks have a different spatial patterning, the hypothesis
that they performed a different function as well is further supported. In sum, if the
ceramic disks were used as malacates, the following criteria should be met:

1. The disks’ shape/form should closely resemble in form known malacates fi'om
the Tarascan cultural area as well as other relevant areas.

2. The disks should be concentrated in household areas, as malacates are.

3. The disks’ average weight and size should fit within the ’size parameters of
known malacates.

As far as morphology is concerned, known malacates from the Tarascan cultural
area as well as those from the Valley of Mexico and the Teotihuacan Valley exhibit a
much broader range of forms than do the Tarascan ceramic disks. These forms include,
but are not limited to,: circular, biconical, truncated biconical, truncated conical,
discoidal, straight and convex cylindrical, jar-shaped, and effigy-shaped. These and other
forms are reported at length in Lister (1949), M. Parsons (1972, 1975), Parsons and
Parsons (1990), Kelly (1947), and Pollard (1993). (Figures 28a-c). In contrast, almost all
of the Tarascan disks are roughly circular and none exhibits any of the more exotic
shapes.

Malacates differ from the Tarascan disks in several other morphological
characteristics as well. The former are commonly decorated with punctate stamping or
incised designs, and some are even painted. None of the Tarascan disks exhibits any
such decoration. Indeed, this artistic embellishment would be completely unnecessary
for something as utilitarian and “invisible” as a net sinker, if that is indeed what the
Tarascan disks were. In addition, malacates were fashioned out of wet clay, often in

molds. In contrast, the Tarascan disks were reworked sherds of pottery that had already

been fired and initially used for another purpose. Perhaps most importantly, malacates

l7

were always perforated in order to accommodate a spindle around which the spun fiber
was wound. In contrast, only one of the Tarascan disks is perforated. If perforation was
important for the Tarascan disks, they probably would also, like the malacates, have been
fashioned out of wet clay, for it would have been much easier to form a hole in wet clay
than to perforate an already fired piece. All of these morphological differences make it
most unlikely that malacates and the T arascan disks were used for similar functions.

Carrasco (1976:224-26) (in Parsons and Parsons 1990:314) discusses Aztec
women’s spinning of cloth. He reports that commoner women spun fiber and wove cloth
in their own households for tribute. In addition, specialized high-quality cloth was also
produced by women who worked at the rulers’ palaces. In both these instances, cloth
was produced at the household level. Although none of the authors cited above discusses
intra-site provenience of the malacates, in light of the ethnographic documentation it is
entirely reasonable to assume that they are found mainly, if not exclusively, in household
contexts. Little is known about the production of cloth in the Tarascan empire but it is
reasonable to presume that it, too, may have been produced at the household level. More
research needs ,to be done in this area.

While intra—site provenience of this detail is also lacking for the Tarascan disks, it
is readily apparent that the latter tend to be concentrated along what was shoreline or
under water at the time of the Tarascan empire. Using a count of ceramic disks as a
proportion of other ceramic sherds one can gauge the relative density of disks in a given
area. At the Pareo sites, the survey units with the highest diskzsherd ratios were found in
areas P33, P57, P17, P29, P42, and P51. All but P57 and P51 are adjacent to the

lakeshore; and these two are close to it.

18

When compared by zone, diskzsherd ratios are quite revealing. The overall ratio
for Urichu, which is the farthest from the lake, is by far the lowest, with a ratio of
121,641. The ratio is much higher for Xaracuaro, which was in fact an island at the time
of protohistoric occupation: 1:61. Pareo, which was even closer to the lakeshore then
than at present, due to lowered lake levels in recent times (O’Hara 1993), has the highest
ratio of all, 1:41 (see Figure 14).

One possibility for this patterning might be that the disks are household artifacts
and that Tarascan households were concentrated near the lakeshore. However, the fact
that many of the Xaracuaro disks were found in areas that would have been underwater at
the time of Prehispanic Tarascan occupation argues against this. This spatial patterning
suggests instead that the significant factor for the Tarascan disks is proximity to
lakeshore, rather than presence at the household level. More data regarding household
distribution within the sites would be necessary to definitively reject the possibility that
the disks are malacates or other household artifacts, but the fact that ceramic disks are
concentrated near lakeshore areas, even places that would have been underwater at the
time of use, strongly supports the net sinker hypothesis rather than the malacate
alternative. It should also be noted that if the disks were malacates, then Urichu, as the
largest community, should have the highest ratio of disks to other ceramics, not the
lowest one.

Finally, the size and weight of known malacates can be compared to the Tarascan
disks. Again, a significant size discrepancy would support their separation into two

distinct artifact classes, especially when combined with the evidence cited above

19

regarding their differences in morphology, method of manufacture, and spatial
distribution.

Macias Goytia (1990) and Parsons and Parsons (1990) do not provide
measurements for the malacates they discuss. Arnauld et al. (19932161) report three
examples from the Zacapu Lake Basin in Michoacan, ranging from 21-48 mm in
diameter and 19-38 in thickness. While the malacates’ diameters are well within the
range exhibited by the Tarascan disks (14.9-59.6 mm), their thickness is not. The
thickest Tarascan disk is only 19.5 mm, well exceeded by the 38 mm Zacapu piece. N o
weights are provided by Arnauld et al. but it is likely that at least the thicker examples
outweigh the Tarascan disks. In comparison, Arnauld et al. report 41 examples of
tepalcates recortados, or reworked sherds, with diameters and thicknessses ranging from
12-17 mm and 2-10 mm, respectively. While the diameters of these artifacts are a bit
smaller than those of the Tarascan disks, their range does overlap. The thickness
measurements, on the other hand, are closer to those of the Tarascan disks, which have a

mean of 5.7 mm and a range from 2.7-19.5 mm.

Table 2. Size Parameters for T arascan Disks

 

 

mean minimum maximum
diameter (mm) 30.7 14.9 59.6
thickness (mm) 5 .7 2.7 19.5
weight (g) 6.2 0.4 35

 

Lister (1949) and M. Parsons (1972, 1975) report much larger sample sizes of
malacates which may therefore provide more reliable comparisons to the Tarascan disks.
Lister. reports 97 malacates from his excavations at Coj umatlan, Michoacan. They range

from 10-55 mm in diameter. While the upper range overlaps those of the Tarascan disks,

20

 

the lower range is smaller than the smallest Tarascan disk (14.9 mm). Lister provides
thickness measurements for only two out seven types of malacates, but both of these are
larger than the Tarascan disks’ mean thickness of 5.7 mm. Three jar-shaped malacates
have a mean thickness of 20 mm, and fifteen flat malacates average 7 mm.

M. Parsons (1972) reports over 200 spindle whorls from a surface survey in the
Teotihuacan Valley, which she divided into stylistic types. Eighty-two Type I whorls
were relatively large, ranging from 35-61 mm in diameter. Most fell between 40-52 mm
(1972147). The range of the Type 1 whorls is larger than that of the Tarascan disks’ 14.9-
59.6, and the range of the majority seem to be much larger than the Tarascan disks’ mean
of 30.7 mm. Thirty-four examples of Type II malacates were larger still, with a diameter
range of 45-73 mm (1972:50). This is significantly larger than the range and the mean of
the Tarascan disks. Seventy-eight Type III whorls ranged from 15-31 mm in diameter,
with 88 % falling between 18-28 mm (1972:52). These overlap only the lower end range
of the Tarascan disks. Two Type IV malacates averaged roughly 53 mm in diameter and
weighed approximately 51 grams (1972253). While this diameter is within the range of
the Tarascan disks, it is much larger than their mean of 30.7 mm. No Tarascan disk
comes close to weighing 51 grams; the heaviest is 35 grams and the mean is only 6.2
grams. Nine Type V malacates range from 39-56 mm in diameter and weigh between 26-
62 grams. While the range in diameters overlaps the upper end of that of the Tarascan
disks, the malacates’ weight is, again, much larger overall than that of the Tarascan disks,
which range from a mere 0.4-35 grams. Two examples each of Type VI and VII whorls
follow the same sort of pattern in comparison with the Tarascan disks, with diameters of

42, 47, 51, and 52 mm, and weights of 44-558 grams. Two examples each of Type VIII

21

and D( malacates fall within the Tarascan disks’ range in both diameter and weight, but
again, towards the higher end of the scale: 29-44 mm in diameter and 7 .4-31 grams in
weight (1972:54-5). In sum, while the Teotihuacan Valley malacates do overlap some of
the measurement ranges of the Tarascan disks, there is enough size and weight
discrepancy to prevent viewing them as members of the same artifact class.

The more than 600 Aztec malacates recovered from the Valley of Mexico also
exhibit size discrepancies compared to the Tarascan disks (M. Parsons 1975). When
analyzing the malacates’ weight, hole diameter, and total diameter, M. Parsons found “a
clear bimodal distribution” indicating that the whorls were “used to spin two different
fibers:” cotton and maguey fiber (ixtli). The larger whorls were used to spin ixtli, while
the smaller ones were used for spinning cotton (M. Parsons 1975:208). Parsons reports
that the typical cotton whorl measured about 20 mm in diameter, while maguey whorls
averaged about 50 mm in diameter (1975:210) (see Figure 28a). While both these figures
are within the range of the Tarascan disks, as averages they are significantly lower, in the
case of cotton whorls, and higher, in the case of the maguey whorls, than the Tarascan
mean of 30.7 mm. In addition, a histogram of the Tarascan disks’ diameters shows a
clearly unimodal distribution (see Figure 3).

Pollard reports 11 cotton malacates recovered during her 1970 archaeological
survey of Tzintzuntzan (Pollard 1993 ). Although the sample is small, it may be the best
comparison to the ceramic disks because Tzintzuntzan also lies in the Lake Patzcuaro
basin and is thus culturally Tarascan. Fortunately, Pollard provides diameter and
thickness measurements of the malacates so a precise comparison is possible. The nine

complete examples have diameters which range from a minimum of 13 mm to a

22

maximum of 26 mm and have a mean of 20 mm. Their height at center, which
corresponds to the thickness measurement of the ceramic disks, ranges from 9 mm at
minimum to 12.5 mm at maximum, with a mean of 10.4 mm. By contrast, the mean of
the ceramic disks’ diameter is 30.7 mm; of their thickness, 5 .7 mm. A t-test calculated
for diameters results in the rejection of the null hypothesis; i.e. demonstrates that the two
sets of artifacts differ significantly in their measurements of diameter and therefore do
not, in all likelihood, represent the same population (Figure 29). A t-test for thickness, on
the other hand, failed to reject the null hypothesis (Figure 30). In this instance, the
thickness of the Tarascan disks does not differ significantly from that of the Tzintzuntzan
malacates. Since no weight measurements were made for the Tzintzuntzan malacates at
their time of collection, no tests were possible for this characteristic.

A statistical comparison of the Tarascan disks to a set of 19 cotton malacates
excavated by Pollard at Urichu as part of the same research project yielded mixed results.
Once again, a t-test on the diameter measurements of the two artifact sets firmly rejects
the null hypothesis that the two groups are part of the same population (Figure 31).
Again, though, as with the Tzintzuntzan malacates, a t-test for thickness measurements
failed to reject the null hypothesis. The thickness measurements of the Tarascan disks do
not therefore differ significantly from those of the Urichu malacates (Figure 32). A t-test
for weight likewise fails to reject the null hypothesis (Figure 33).

By way of comparison, t-tests were performed comparing the Urichu and
Tzintzuntzan malacates for the measurements of diameter and thickness (again, it was not
possible to compare weights). As one might expect, the test for diameter failed to reject

the null hypothesis (Figure 34) . The test for thickness similarly failed to reject the null

23

hypothesis (Figure 35). Thus, the tests seem to validate that the two malacate sets are
from the same population, a finding which is consistent with the archaeological data.

On the other hand, although t-tests for thickness and weight turned up no
significant statistical differences between the populations of Tarascan disks and known
Tarascan malacates, I do not believe that this means that the two artifact groups should
be viewed as belonging to the same class. That the t-tests performed for the variable of
diameter firmly reject the null hypothesis argues that they come from different
populations and thus should not be classified together. The similarities between the two
artifact classes in terms of thickness and weight, however, do explain why several
researchers have postulated that the Tarascan disks might be malacates. Seen within the
larger body of evidence discussed within this paper, however, including the statistical
evidence, it is clear that the Tarascan disks represent a class of artifacts distinct from
Tarascan or other Mesoamerican malacates.

Kelly (1947:98) provides additional examples of discoidal, worked sherds, which
she distinguishes from the malacates she analyzes. She reports 32 artifacts of this sort, 27
of which are unperforated and five of which are perforated. Although a higher
percentage of this small sample is perforated compared to the Tarascan disks, the two
samples are similar in that the vast majority are not perforated. Kelly does not provide
weight or thickness measurements but reports a range in diameters from 15-65 mm,
almost identical to the Tarascan disks’ range of 14.9-59.6 mm. This suggests that the two
groups may well belong to the same artifact class.

In conclusion, measurements from reported samples of Mesoamerican malacates,

while exhibiting some overlap with the Tarascan disks, differ enough to prevent grouping

24

them as a single artifact class. When combined with differences in spatial patterning and
the overwhelming morphological differences, it is clear that the Tarascan disks are a
distinct artifact class and that the hypothesis that they functioned as malacates can be

definitively rejected.

Gaming Piece Hypothesis

The second hypothesis to be explored is that the ceramic disks were some kind of
gaming piece. Limited ethnographic information exists concerning common games from
modern Mexican peoples, including present-day Tarascans. Again, this information can
be applied to the artifacts in question. For this hypothesis to be considered viable, the
following predictions should be satisfied:

1. The ceramic disks should closely resemble known gaming pieces from the
Prehispanic and/or modern Tarascan cultural area in size and shape.

2. The ceramic disks should be found in relatively small clusters, and not in great
concentrations.

3. The ceramic disks should be concentrated in either household areas or common
(non-ceremonial?) public areas.

4. The disks should all be of uniform size or fall into several discrete size ranges,
as might be expected of gaming pieces.

Interestingly, the Tarahumara of northern Mexico produced artifacts which are
similar to the ceramic disks which are the subject of this thesis. In this case, the disks are
known ethnographically to be used in the popular game of cuatro, or dihibama, as the
Tarahumara call it. This game is played with two teams, usually comprised of two men
each, who try to throw the disks into a small goal in the opposing team’s field. As
reported in Bennett and Zingg (1976:341), the disks may be chipped out of stone, or

“made of old potsherds, rounded into shape by grinding" (italics mine). It is in the latter

25

case that the disks would, presumably, most resemble those artifacts under study here.
How then to distinguish the two classes of artifacts? Are they in fact one and the same?

We do have available several points of comparison. Most obviously, perhaps, the
dihibama disks would not be found in the immediate riverine environment, where fishing
took place, but instead would most likely be found in the vicinity of living quarters or out
in the open where a game court could be situated. The Tarascan disks, on the other hand,
seem to be concentrated near the lakeshore where fishing nets would have been used.

As far as size is concerned, the two types of disks seem to have some similarities.
Bennett and Zingg (19762341) report that the Tarahumara disks can be as much as 3-4
inches (76.2-101.6 mm) in diameter and up to one inch thick (25.4 mm), although they
can be much smaller. Their size is dependent upon the size of the court being used to
play the game. In comparison, the Tarascan disks have a mean diameter of just 30.7 mm,
with a minimum of 14.9 mm and a maximum of 59.6 mm. They have a mean thickness
of 5.7 mm., with the smallest at just 2.7 mm thick and the largest at 19.5 mm. Although
there seems to be some overlap in size ranges between the Tarascan disks and the
Tarahumara disks, on the whole the Tarascan ones seem to be smaller, in some cases
substantially so. Since no precise averages are provided by Bennett and Zingg, exact size
differences are difficult to assess; however, the suggestion of an overall size discrepancy
between the two groups of artifacts could reflect differing uses within their respective
societies.

Perhaps most significantly, there is no evidence that the disks manufactured for

playing dihibama are, or ever were, notched. In contrast, notching is clearly the norm for

26

Tarascan disks. Based upon these points of comparison, then, the Tarascan disks and the
Tarahumara disks seem to comprise two separate and distinct artifact classes.

Brand (1951) makes no mention of a dihibama-like game being played in his
ethnography of Quiroga, a modern Tarascan settlement in the Lake Patzcuaro Basin; nor
does Smith (1965) mention it in his dissertation on the inhabitants of the island of La
Pacanda.

Foster describes several games played by present-day Tarascans in his 1948
ethnography, Empire ’s Children: The People of ’llzintuntzan (1948:238-40). Among
those utilizing gaming pieces is a game called peleche, similar to hopscotch. Played by
two or more persons, each player has a pebble or potsherd which he tosses into various
boxes marked out on the ground with chalk or in dirt according to the rules of the game.
Although this game utilizes potsherds, it is doubtful that enough players could be
involved to produce the quantity of disks found in the Prehispanic levels of the areas
explored by Pollard. In addition, there is no mention by Foster of any purposeful
modification of the potsherds, such as side-notching, used in the game.

The only other game Foster describes which uses game pieces is e! coyote or
coyote y gallinas (“the coyote” or “coyote and chickens”). Similar to checkers, the game
is played on a 30 x 60 cm. square which is drawn on paper, on a board, or in the dirt.
Beans, com, or beer bottle tops serve as markers representing the coyote and 12 chickens.
Foster does not report potsherds, modified or otherwise, being used as markers for this
game, even though they are obviously used in peleche. It is therefore unlikely that
markers used for this game account for the Prehispanic disks, assuming it was even

played in that period.

27

Beals and Carrasco (1944) also discuss games played by the modern Tarascans,
albeit those of the mountain areas and not necessarily those around Lake Patzcuaro.
They describe a “hockey-like” game played with sticks or paddles and balls. None of the
equipment used in this game bears any resemblance at all to the Tarascan disks. Three
other games described by the authors, kolicatékua, also known as quince or patolli;
méskukua; and kuaio, or Coyote and Hens, do involve gaming pieces; unfortunately,
however, these are not described in any detail so that they cannot be compared to the
Prehispanic disks. Kolicatakua requires two players with four game pieces each, and
mOSkukua and kuaio are both played by two players, each with 12 pieces. Again, it
would be difficult to imagine game-playing on a scale sufficient to produce the numbers
of Tarascan disks found, especially in areas that would have been underwater. At the
time of Beals and Carrasco’s ethnography several of these games were on the wane; it is
entirely possible that some of them are no longer played at all. If they are, however, it
would be extremely informative to collect some of the gaming pieces in order to compare
their form with that of the disks. In the case of quince, the authors mention that it is
played either in a household or in the plaza, depending upon the time of year. It is '
reasonable to assume that the other “board” games, which were played on the ground,
were played in similar spaces. In this case, then, the distribution of these artifacts within
a site would be different from that of the Prehispanic disk artifacts, thereby suggesting a
possibly different function.

Although I originally hypothesized that a distribution of several discrete size
ranges might be a marker for gaming pieces rather than net sinkers, it appears that none

of the ethnographically documented games played by modern Tarascans uses pieces of

28

more than one size in any given game. Therefore, this might not be a useful criterion for
classification. In fact, the disks falling into discrete size categories could even be used to
support the net sinker hypothesis, if different size nets were used It may be, however, a
moot point, since a histogram of the disks’ diameters shows a distinctly unimodal
distribution (see Figure 3). Aside from this, the lack of similarities in form (e. g.
notching) between the Prehispanic Tarascan disks and any gaming pieces described in
modern ethnographic sources strongly argues against classifiying them together. In
addition, the sheer quantity of Prehispanic disks, as well as their spatial patterning (e.g.
proximity to water) also seems to be unmatched by modern gaming pieces. Therefore, it
seems reasonable to reject the hypothesis that the Prehispanic Tarascan disks were made

as, or used for, gaming pieces.

Net Sinker Hypothesis

The final hypothesis posited for the ceramic disks is that they functioned as
fishing net weights, or “net sinkers.” Again, certain patterns in the data set would make
this hypothesis more or less likely. For instance, a strong distribution of ceramic disks
along present or past lakeshore would provide one indication that they functioned in this
capacity. Other predictions concerning notch prevalence and morphology can be tested
as well. Specifically, the following criteria should be demonstrable:

1. Ceramic disks should be distributed primarily in lakeshore areas.

2. Notching, or more specifically, double notching, should be prevalent, or the
norm, as it would facilitate attachment of disks to fishing nets.

3. The disks should be found in rather large concentrations, at least in some areas
where they were likely to have been used.

29

4. The disks should be of a suitable weight and size in order to be functional as
fishing net weights.

5. The disks should have a limited range of variance for diameter, weight, and
thickness measurements. The distributions may be either unimodal, bimodal, or
multirnodal, depending on how many types or sizes of nets were used.

As we have seen in Chapter 1 above, the Tarascan disks do indeed seem to be
distributed in or near lacustrine areas. This is consistent with descriptions contained in
The Chronicles of Michoacdn (1970: 127). This patterning is especially strong in the
survey zones of Xaracuaro and Pareo. The exception is in the Urichu zone, where there
were only three units which yielded any disks at all, from excavations conducted at the
site of Urichu. These are clustered together in the western part of the site. Perhaps this
represents an area where the artifacts were manufactured or stored. Ifthe artifacts are
fishing net weights this could be expected to take place at the household level, perhaps in
separate storage sheds such as those used by the modern fishermen of La Pacanda (Smith
1965:23). Further research would be useful in clarifying this. At any rate, the spatial
distribution patterning of the Tarascan disks is consistent with descriptions of Prehispanic
fishing practices contained in The Chronicles of Michoaccin (1970: 127,13 1), including
the one already noted above. In addition, The Chronicles of Michoaca’n describes how
Lake Patzcuaro fishermen “put their nets out to dry on some poles near the shore and had
placed their fish there to dry also” (19702131). The occurrence of Tarascan disks in close
proximity to shoreline is consistent with the net sinker hypothesis since cheremekua nets
were placed in shallow waters near the shore and also set near the shore to dry.

As mentioned above, double opposing side notching appears to have been the

norm for the Tarascan disks. This morphology would certainly facilitate tying the disks

30

to a net. It bears repeating that the notching morphology is absent in all known malacates
and gaming pieces from the Mesoamerican cultural area.

Although large numbers of ceramic disks were recovered from all three sites
(Urichu, Pareo, and Xaracuaro), their concentration in any single excavation unit is not
such that their use as gaming pieces can be ruled out. For example, even the top-ranking
units typically yielded anywhere from 14-21 disks. As discussed above, this is consistent
with the number of pieces that could have been used for a game of mo’skukua or
kolicatdkua, providing that they were played in the Prehispanic era. The criterion for the
artifacts being found in large concentrations, then, appears not to be entirely met,
although the fact that they are found in largest concentration in lake areas (at least at two
sites) seems to support the artifacts’ use as net sinkers, not as malacates or gaming pieces.

Whether or not the ceramic disks are of a suitable weight and size to function as
net weights may be more difficult to ascertain, but ethnographic information provided by
Foster (1948:102) is useful in this regard. Clearly the Tarascan disks are much too small
to function as weights for the chinchorro nets which were being used then in
Tzintzuntzan. According to Foster, these were weighted only when fishing for bass, with
unworked stones of 4-8 kilos. The Tarascan disks, on the other hand, weigh only from
0.4-35 grams. Perhaps they are analogous to the “tiny pebble sinkers” used to weight the
chere’mekua nets. These nets are used to catch khuerepu, or the young of a group of
closely related fishes collectively referred to as pescado blanco, or white fish. These are
members of the Chirostoma genus, including Ch. estor, jordani, bartoni, and

michoacanae (194821024). The young ch. bartoni is specifically referred to as choral.

31

Foster reports that small cherémekua, or gill, nets are about 60-80 cm wide; large
ones three to four times that. Lengths for individual nets are up to about 25-30 meters,
although they may be joined together to form huge nets as long as 250 meters. These
large nets are used to fish bass (an exotic species introduced in modern times) from

canoes. Foster provides the following description of the smaller cherémekua nets:

The cherémekua gill net, though used to a considerable extent in
other parts of Lake Patzcuaro, is little used in the vicinity of Tzintzuntzan.
The small-sized one is suspended between two poles, usually near the
shore, with tiny pebble sinkers carrying it nearly to the bottom. Then the
fishermen, either walking in the water and splashing or making noise from
a canoe, drives the khuerepu and choral toward the net, and in trying to
pass the fish are caught by the gills (1948:104).

Foster provides no additional description or illustration of these sinkers. Neither
does a similar description in The Chronicles of Michoacdn ( 1970: 127) of fishermen who
“put (their) nets along the shore to catch fish.” Were they literally pebbles or is it
possible that they could have been re-worked potsherds such as the Prehispanic disks?
Perhaps potsherds were used in Prehispanic times for the same purpose? Perhaps they
are used in other parts of the Lake Patzcuaro basin where cherémekua nets are more
popular? Pollard’s observation that modern Tarascan fishermen have been known to (re-
)utilize the Prehispanic disks as net weights makes this entirely possible (Pollard,
personal communication).

The final criterion for this hypothesis, that the disks should exhibit a limited
distribution in terms of size measurements, seems to be supported by the data. As
discussed in Chapter 1 above, histograms and other graphing techniques all show rather
tight, unimodal distributions for each of the measurements of diameter, thickness, and

weight. This makes sense in light of the ethnographic data reported by Foster regarding

32

small cherémekua nets, the only type of net which uses small sinkers. These nets are all
similar in size; larger nets are formed only by combining a number of smaller ones, and
these use only large stone sinkers.

In summary, the statistical and ethnographic data seem to support the net sinker
hypothesis most strongly. Although the numbers of ceramic disks found are not
inconsistent with the gaming piece hypothesis, they are also not inconsistent with the net
sinker hypothesis The data are consistent with all of the criteria put forth for weighing
the net sinker hypothesis (no pun intended). In contrast, there are serious problems with
both of the other hypotheses, that the disks were malacates or that they were gaming
pieces. In light of available information, then, it seems pragmatic to accept the net sinker
hypothesis as the most likely explanation of the Tarascan disks, at least for now. Future

anthropological research may support or weaken this conclusion.

33

CHAPTER 3
ARCHAEOLOGICAL CONTEXT

Considerable confusion regarding the function of these North American artifacts
exists in the literature. Similar artifacts are variously classed as different items, and,
conversely, different types of artifacts are sometimes assumed to have the same function.
These will be examined on a case-by-case basis to determine which, if any, functioned

similarly to those from the Tarascan cultural area.

Ceramic Disks from Prehispanic Mexico

Ceramic disks similar to those recovered in Pollard’s researches have been
reported from other areas of the Lake Patzcuaro Basin (Toledo et al. 1993) as well as
from several other lake basins in Mexico. Arnauld et a1. (1993) report them from the
Zacapu Lake Basin, Michoacan; Macias Goytia (1990) reports them from Cuitzeo,
Michoacan; Jeffrey Parsons (personal communication) reports them from the Basin of
Mexico; and Yoko Sugiura-Nancarrow (personal communication) from the Toluca

Valley and Lake Chapala. Some of these have already been discussed above.

Similar Stone Artifacts from North American Sites

In addition, notched and grooved stones of various sorts have been reported from
all over native North America, including Michigan (Cleland 1966), Alaska (Weyer 1930;
De Laguna 1934), Pennsylvania (Meehan 1893), New York (Ritchie 1944), California

(Wallace and Lathrap 1975, Pohorecky 1976), and Texas (Watt 1938).

34

Evidence for the use of net sinkers in native North America is mixed and
somewhat less conclusive than the evidence for fishing and fish consumption in general.
Rostlund (1952:291) reports widespread distribution of grooved or notched stones
throughout the eastern half of the United States. Although widely presumed by their
excavators to be aboriginal net sinkers, Rostlund is critical of some, but by no means all,
of the claims.

Clear evidence of their use in Pennsylvania is discussed in Meehan (1893), who
documents the Native Americans of this region as expert fishermen whose diet was
largely dependent on the abundant fish of Pennsylvania’s rivers. As he reports:

Sieves and gill nets had lead lines made of small circular flat stones

having two deep notches to keep the lines from slipping. These “leads”

have been found by the thousands in the Delaware and Susquehanna
rivers...Dr. C.C. Abbott, curator of the Archaeological Department of the

University of Pennsylvania... came upon a series of about thirty seine

“leads” stretched in an irregular line about eighteen inches apart. From

their position and other evidence found in connection therewith, it was

apparent that they formed part of a net which had been set and then

abandoned (1893:8-9) (italics mine).
Meehan also points out that the use of fish nets in this region is well documented both
archaeologically and ethnographically. Not only do many pottery sherds bear net
imprints, but early writers in the region bear witness to having seen them used (189328).
In the case of aboriginal Pennsylvania, then, support for the use of net sinkers is multi-
fold: a strong tradition of fishing, including weighted nets, which is documented both
archaeologically and ethnographically, and the fact that many sinkers were found in
actual riverine contexts. In this case, at least, it appears that the morphological

characteristics of the artifacts (small, flat, circular stones with side notches) is a good

indicator of their use as net weights.

35

Custer (1996) also documents net sinkers at numerous sites in and adjacent to
Pennsylvania ranging in age from the Early Archaic to the Late Woodland periods. Net
sinkers were found at the Harry’s Farm Site in New Jersey’s Upper Delaware Valley,
radiocarbon dated to between 73 80-7320 B.P. (Early to Middle Archaic). According to
Custer (1996:114) it is one of the “small, repeatedly occupied base camps at which varied
floral and fauna] resources were procured, processed, and consumed” which typify this
era. Three caches of stone tools, including one of net sinkers, were found at the Late
Archaic (3000-1000 BC.) F aucett Site, also in the Upper Delaware Valley (1996:201).
Several Late Woodland (AD. 1000-1550) sites have yielded numerous net sinkers,
including 29 at the Williamson Site and 95 at the Overpeck Site. Both are described by
Custer (1996:293-4) as basic hunting and gathering camps which were characterized at
this time by, among other things, “an economic adaptation involving the intensive
exploitation of anadromous fish in conjunction with hunting and gathering.” Moeller (as
reported in Custer 1996:295-6) has reported dense accumulations of faunal remains,
including fish, in refuse pits from the Late Woodland period sites on the Pennsylvania

side of the Delaware River, testifying to their importance as a food resource.

The net sinkers discussed in Custer (1996: 185) appear to be identical to those
reported by Meehan: they are small (ca. 6-7 cm in diameter), round cobbles with peeked
bilateral notches. Significantly, they are noticeably distinct in form from atlatl weights

from the same area (1996: 193).

Evidence for the use of net sinkers in prehistoric New York is also strong (Ritchie
1944). Ritchie catalogues over 9,800 net sinkers from seven pre-Iroquoian sites in New

York state. Occupation levels yielding net sinkers range from roughly 3000 BC. - AD.

36

1600. The 5,000 year-old Lamoka site alone yielded some 8,000 of them (Rostlund
1952:96-7). Ritchie does not explain the criteria be used to categorize the net sinkers, but
their sheer number and their morphological consistency support their classification as a
unique artifact class. With only one or two exceptions, all of the net sinkers are flat, ovate
sandstone disks with two side notches. Except for the fact that they are made of
sandstone and are more nearly oval than circular, they are nearly identical to those
artifacts analyzed in this thesis. And, as with the Tarascans, their abundance reflects a
strong reliance on fishing, a contention borne out by the presence of additional
archaeological evidence such as fish hooks, gorges, and shell middens. All of the
artifacts labeled as net sinkers were excavated from sites adjacent to water. Again, the
link between a developed fishing technology and the presence of large numbers of flat,
circular or ovate notched stones is strong, and lends credence to classifying the latter as
net sinkers. Unfortunately, Ritchie does not provide specific information about the intra-
site provenience of the net sinkers (although undoubtedly it was noted at the time of
excavation). However, like so many Tarascan settlements, this site is adjacent to lake
waters suitable for fishing.

The northeastern Native Americans’ use of net sinkers has been documented not
only archaeologically but also ethnohistorically. Samuel de Champlain left extensive
written records of his travels throughout this region, and makes numerous mentions of
fishing practices among the natives he encounters. In general he considered the natives
of this region to be excellent and skilled fishermen, and the waterways to be very

productive (see Biggar 1929). Speaking specifically of the Iroquois’ territory and the St.

37

Lawrence river, he relates

...the rivers, streams lakes, and ponds are as numerous as one can
wish, with abundance of salmon, very beautiful fine large trout of every
kind, sturgeons of three sizes, shad, very good bass, some weighing
twenty pounds, suckers of all kinds, some of them very large, and pike,
some of them five feet long, catfish without scales in two or three
varieties, large and small Whitefish a foot in length, pickeral smelts,
tench...

Of the New York area, he mentions that the inhabitants of Lake Couchiching,
adjacent to Lake Huron, “make great catches of fish which they preserve for the winter”
(1929156). Although in this instance he does not record their method of harvest, he goes
on to state that at the adjoining Lake Simcoe “the great catch of fish takes place by means
of a number of weirs which almost close the strait, leaving only small openings where
they set their nets in which the fish are caught...” (1929:56-7). Of the sedentary
Attigouautan tribe, who lived on the peninsula between the Georgian Bay and Lake
Huron, he recorded their practice of making and trading fish nets with some of their
nomadic neighbors (1929: 13 1). These nets were made by the women of the tribe, out of
spun hemp fiber (1929:136). In contrast, among other tribes in the area the men
fashioned the nets, which were employed in winter as well as summer (1929: 166-7). As
to fishing in the winter, which greatly impressed him, Champlain records:

...they make several round holes in the ice and that through which

they are to draw the seine is some five feet long and three feet wide. Then

they begin to set their net by this opening; they fasten it to a wooden pole

six or seven feet long, and place it under the ice, and pass this pole from

hole to hole, where one or two men put their hands through and take hold

of the pole to which one end of the net is tied, until they come back to the

opening five or six feet wide. Then they let the net drop to the bottom by

means of certain small stones fastened to the end of it. After it has been

on the bottom they draw it up again by main force by its two ends, and
thus they bring up the fish that are caught in it (1929:167-8; italics mine).

38

 

Champlain does not mention which species of fish were harvested in this way. It is
possible that nets (with sinkers) were used to capture the well-known Great Lakes
Whitefish (Coregonus sp.[Cleland 1966: 172]), a species entirely unrelated to the
Whitefish (“pescado blanco ”, or Chirostoma estor) of the Tarascans’ territory. This type
of winter use of nets is what Schoolcraft reports when he describes the same practice:

Another mode of taking fish in the winter, is by making a series of

orifices, through the ice, in a direct line. A gill-net is then pushed, by its

head-lines, from one orifice to another, until its entire length is displayed.

Buoys and sinkers are attached to it, and it is then let down into deep

water, where white fish, and other larger species, resort at this season. The

next morning the net is drawn up, the fisherman secures his prey, and

again sets his net as before. By this mode, which is very common

throughout the lakes where deep water abound, these species are captured

at the greatest depths, while sheltering themselves in their deepest winter

recesses (1852:51) (italics mine).

A legitimate question is why the natives of Pennsylvania and New York used
stone for their net weights when pottery was obviously abundant. One would imagine
that it would be much less labor intensive to grind pottery sherds than stone into grooved
disks, although this needs to be tested experimentally. The exception, perhaps, would be
sandstone. It is a relatively soft stone and for the natives of the New York area may have
provided the durability of stone with the ease of workability of fired ceramic. Perhaps
stone is more durable; it may not be more difficult to work after all (for instance, it may
not shatter as easily as pottery), but that does not then explain why the Tarascans used
pottery. Perhaps the Tarascans’ pottery, which was of superior craftsmanship, was more
durable than that of the native Pennsylvanians, and thus durable enough that stone (and
the labor involved in working it) was not necessary. Or, less likely, perhaps the natives of

the northeast used pottery too, but it has not survived in the archaeological record as the

stone did. As Ritchie notes, the pre-Iroquoian pottery “is not remarkable for quality,

39

being rather coarsely grit-tempered and friable” (1944:110). Certainly stone net sinkers
are known archaeologically from other areas of the world, such as Neolithic Finland
(Broadbent et al. 1993: 197—8). As Joseph Chartkoff (personal communication) suggests,
perhaps stone’s density makes it more suitable for use in active waters. This would
account for its use in the rivers of the northeastern United States. Likewise, pottery
sherds would have been perfectly adequate for the relatively calm shallows of Lake
Patzcuaro. The different media used for net sinkers could even reflect differences in
gender-based division of labor. The preference for one material over another when both
were available is still open to explanation. It may simply be idiosyncratic or it may have
to do with inherent properties of the chosen medium. Additional archaeological,
ethnographic, and experimental data would all be useful in clarifying this question.
Rostlund (1952) documents the fishing expertise of the Native Americans who
occupied the Great Lakes region. The cultures of this region provide a relevant point of
comparison to the Tarascans not because they harvested the same species of fish but
rather because they developed, at least in some times and places, a similar specialization
or focus on fish resources as a dietary staple (Cleland 1966). Their use of set gill nets was
noted by numerous European explorers in the area (eg. Schoolcraft 1852). Due to their
habit of feeding on small lake-bottom organisms and the fact that they are not biting
fishes, members of the Whitefish family which inhabit the Great Lakes are not readily
taken by many common techniques such as fishhooks, weirs, or spears (Rostlund
1952:28-9); although Jenness reports that the Ojibwa of Parry Island commonly took
Whitefish with spears (in Cleland 19662172). Schoolcrafi, too, notes that “The white fish,

so common to the whole line of lakes, never bites at a hook, and is captured solely by

40

9, “

nets or spears” (1852253). Therefore, wherever Whitefish were an abundant food
source, we might reasonably expect net fishing, and net sinkers, to have been employed.
As Rostlund reports, “It was the larger and fatter whitefishes more than the small lake
herrings that were sought by the native fishermen, and the problem of how to catch these
night and bottom feeders was solved by means of the set gill net (1952:29). F urtherrnore,

“. .. it (is) very clear that the gill-net fishery par excellence in native North

America was found in this region of great interior lakes inhabited by the

Whitefish family; and it may be added that this great food resource could

not have been adequately exploited had the gill net been unknown. . .I

submit that as fishermen these people from the Great Lakes toward the

Mackenzie Valley were second to none in aboriginal North America

(1952:29-30).

Schoolcrafi testifies that the natives of this region successfully employed the gill
net technique described above during the wintertime: “Fish are sometimes brought up in
the immediate vicinity of Michillimackinac, from a depth of eighty fathoms (1852:51).

Although he does not report net sinkers, Cleland finds it likely that net fishing
was practiced in some areas of the Great Lakes, especially during the Middle Woodland
period (300 BC-AD 400):

With the advent of Middle Woodland influence from the south,

fish became a more important subsistence item and a new technique was

employed to catch them, either nets or weir traps. The use of this

technique is indicated by the fact that the fish remains of this horizon more
nearly represent a random sample of both the species and sizes of fish
which were in the river. When one employs a seine or builds a fish weir,

large as well as small fish are caught; these are non-selective fishing

techniques (1966: 141).

Such a faunal “signature” was found in the Middle Woodland levels of the
Schultz site in Michigan, located on the Tittabawassee River in Saginaw County, where it

most likely represents a focused reliance on fish as a food resource in early spring when

41

other food resources were scarce (1966:141-4). Cleland reports evidence documenting a
seasonal fishing village pattern in the Saginaw, Michigan area by at least 2,000 BC.
(1966:55). By the Middle Woodland period (300 BC. — AD. 400) this fishing
specialization most likely utilized nets and was responsible for a substantial increase in
population density and summer residential stability (1966:94-5; 141; 143). Cleland’s
extensive study of this region’s archaeological record led him to concludethat the
. . large aboriginal villages of the Upper Great Lakes were sustained and attained
economic stability through an adaptation specialized in the exploitation of fish resources”
(1966: 162). No net sinkers have been found at the J untunen site in Mackinac County,
Michigan; nevertheless, Cleland argues that the sheer quantity of Whitefish remains
found there make it likely that weighted gill nets were used by its inhabitants since
Whitefish are not easily speared (19662176). Cleland might just as easily have been
speaking of the Lake Patzcuaro Tarascans as the Great Lakes natives when he states that
“fish were the staple food element in the diet of J untunen site peoples and. . . the very
location and existence of the site was a function of available fish resources in that
locality” ( 1966: 194).

Several researchers have reported net sinkers from California (see Moratto 1984).
For example, Pohorecky (1976:54-94) reports eight stone sinkers from the Willow Creek
sites in Monterey County. These were found with a variety of other fishing paraphernalia
as well as tools and other articles manufactured from shells and fish bones, including
fishbone awls, worked shell, shell spoons and scrapers, shell fishhooks, and shell beads.
Two carbon 14 dates of AD. 72 +/- 250 and AD. 112 +/- 400 were obtained from the

sites (Pohorecky 1976297 in Moratto 19841240). Much earlier examples of “notched and

42

grooved net sinkers” are reported fiom Wallace and Lathrap (1975) (see Moratto
1984:259) for a site in Alameda County (Ala-307), which the excavators dated to 1910+/-
450 BC. and 750+l-300 B.C. Again, these were found in context with other features,
such as shell middens, which clearly indicate a reliance on fish as a food resource. Other
sites and examples from this state abound and are beyond the scope of this paper to
detail. Interested readers are directed to Moratto 1984 for a more comprehensive
treatment.

The case from Lovelock Cave, Nevada (Loud and Harrington 1929) is less clear,
but nevertheless establishes a connection between fish remains, fishing paraphernalia
(nets, books, etc), and manufactured stone disks. In this case, the stone disks are not
notched but drilled (Figure 36). No specific claim is made by the excavators that the
perforated disks were used to weight nets; they classified them simply as “perforated
stones” (1929: plates 55 and 65). Remarkably well-preserved fragments of fishing nets
of several kinds were found, however, in rather large quantities (Figure 37). Some of
these have been estimated to have reached 30-35 feet in length, with one exceeding 42
feet (1929:89). Certainly the stones could have been used as net sinkers, although
evidence to support this is not as strong as in Pennsylvania or New York.

Similar evidence comes from the Aleutian village of Hot Springs at Port MOller,
Alaska (Weyer 1930). Abundant fish remains in refuse pits and occupation levels
indicate that the inhabitants were fishers as well as hunters (19302246). More than 50
artifacts of various shapes, including some which are roughly circular, were found and
specifically classified by Weyer as net sinkers. Most of these are grooved, but at least

one type is perforated. All are stone (Figure 38). Again, the evidence that the artifacts

43

 

were used as net sinkers is rather circumstantial (Weyer does not explain his
classification), but certainly consistent with the cultural context. It should be noted that
the use of stone rather than pottery would be expected this far north.

Although de Laguna (1934) also excavated in coastal Alaska and also found
numerous notched stones (Figures 39a-b), she came to a very different conclusion
regarding their function. Noting that “. . . the distribution of the notched and grooved
stones does not correspond to that of the fish hooks,” nor to areas near the water, or rich
in fish bones, de Laguna suggests that “. . . they may have had nothing to do with fish”
(1934:170-1). Instead, she cites ethnographic and archaeological evidence that they were
used as various types of bolas and throwing stones. While her evidence is strong, it does
not exclude the possibility that at least some of the stones were also used to weight
fishing nets or lines. De Laguna further remarks that “It is interesting to note that the
Aleut name for the bird bola weight is the same as that for the fish-line sinker, suggesting
that the specimens may have been actually interchangeable ” (19342171, italics mine).
The possibility exists that some artifacts originally fashioned as bola weights were reused
as fish line sinkers, further confusing the archaeologist. At any rate, they almost certainly
were not used on true netted seines or gill nets, for as Rostlund points out (1952:97), the
Aleuts did not acquire these until after contact with the Russians.

Not all side notched stones functioned as net sinkers, however. A good example
of this comes from central Texas, where a large number of such stones have been found
and claimed by several early researchers to be net sinkers (Watt 193 8) (Figures 40a-b).
The artifacts fell into four morphological categories: 1) natural pebbles unworked except

for side notches (the most common type, 1938:46); 2) flaked or chipped; 3) ground or

44

peeked; and 4) unifacially worked (usually flaked) (1938:24). Although the first category
is morphologically similar to known net sinkers (c.f. Ritchie, quoted in Watt 1938244),
the remaining three are not. The average length (many are oval or nearly rectangular) of
these artifacts is 45 mm (1938226), which is somewhat larger than the mean diameter of
30.7 mm for the Tarascan disks. Although the smallest specimen measured 30 x 20 x 12
mm, which is within the diameter range of the Tarascan disks, the largest, at 100 x 55 x
35 mm, is clearly outside the range of any known Tarascan disk. That the size ranges of
the two sets of artifacts differ so much may argue for separate functions as well.

Workrnanship on the artifacts, according to Watt, ranges from “crude” to
“skillfully shaped objects of highly specialized forms” (1938226). One might reasonably
question why net sinkers would need to be skillfirlly shaped or highly specialized. As
discussed above, they do not require excessive amounts of craftsmanship to function
adequately. Some of the stones are grooved around the entire circumference of their long
axis (1938230), which clearly entails much more work than is necessary to attach a sinker
to a net. Watt concurs, stating

Anyone who has run trot lines or throw lines along Central Texas streams

has never lost much time finding suitable rocks to use without expending

any labor on fashioning them to a peculiar shape. Also the hardness of the

material requires hours of patient work to shape to perfect form (1938255).
It is possible that more specialized or decorated forms functioned as “charms,” without
which the net would be considered useless for catching fish. Such charms were
’ reportedly used by the North American Chipewyan (Oswalt 1988277). It is also possible,
of course, that they served a purpose or purposes altogether unrelated to net fishing.

Although many of the artifacts were found “across the trends of streams” (Watt

1938248), none has apparently been found in an actual stream bed (1938255), unlike those

45

in the northeastern states. Many were also found in camp sites on high banks adjacent to
rivers (1938253). Watt makes a valuable point when he says

The term net sinker is most commonly applied and accepted for these

objects and the irnpelling reason is the fact that the sites where found (sic)

are predominantly along the streams or shores of other waters. However,

the fact must not be overlooked that this is a condition that applies to

almost all camps of any nature (1938255).

Not surprisingly, then, Watt concludes that it is very unlikely that these artifacts
were used as net weights. He deems their most likely use to have been as grips and
shuttles for winding arrow shafts (1938256-7), but does not rule out other, less likely,
uses, for lack of evidence: hammers, pendants, ornaments, or charm stones.

In sum, then, although not all cases of purported net sinkers can be substantiated,
they were undoubtedly used for this purpose in some areas and periods. Referring to the
so-called Waco sinkers, Watt sums up the problem of identification:

The early writers named...them sinkers, from a definite knowledge that

notched pebbles were attached to fishing lines or nets as weights to hold

them under the water; however similarity of form does not necessarily

determine that all such similar forms were used for identical purposes

(1938254).

When evaluating the evidence for fishing net weights, several factors seem important in
lending credence to their functional classification. First and foremost would seem to be
the existence of a developed fishing technology and thus a substantial cultural reliance
upon fish as a food source. Rostlund posits that . . the degree of proficiency in the use
of fish nets is a measure of how effectively the fish resource is utilized...” (1952284). As
he points out,

Wherever nets can be used, more fish can be taken with them and with

less effort than by any other method. But fish nets are not so easily

manufactured as some other fishing gear, and the very fact that people
spend time and effort in making nets can only mean that they are seriously

46

engaged in fishing. A highly developed net fishery is a distinguishing

mark of a specialized or professional fisherfolk. (1952281)
It is among just such cultures that we would most reasonably expect to find fishing net
weights. With this criterion, the classification of discoidal stones or sherds, with or
without notches, as a functional artifact used as a net sinker, seems eminently reasonable
for not only the natives of northeast North America and the Great Lakes but also the

Tarascans, who were highly skilled fishermen.

47

CHAPTER 4
FISH AND FISHING IN THE TARASCAN EMPIRE

 

 

Fish Nomenclature

Scientific Name Spanish Name Tarascan Name

Chirostoma. ottenuatum charal

Ch patzcuaro charal khuerepu, k’uerépu, cuerepu/o
Ch. grondocule charal khuerepu (etc); chakuami

Ch. estor pescado blanco urapeti, kun’rca urapeti/urapiti
Ch jordoni pescado blanco

Ch. bartoni pescado blanco, charal (Foster) kuerepo turipiti

Ch michoacanae pescado blanco

Algansea lacustris akumara, acumarani

Goodeo Iuitpoldi thiru, t’iru

 

Table 3. (Compiled with information provided in Toledo 19912166; Foster 1948; Craine
and Reindorp 1970; and Pollard 1993).

Fish in the Tarascan Diet and Economy

Not surprisingly for a society situated in a large lake basin, fish formed a staple of
both the Tarascan diet (Gorenstein and Pollard 1983:105-7) and economy (19832103-8).
In fact, the Aztecs’ name for the Tarascans’ territory, Michoaccin, literally means “the
place of the masters of fish” in Nahuatl, their native tongue (Coe 19842142).

Fish comprised the Tarascans’ single largest export, and were widely exchanged
outside the Lake Patzcuaro Basin for essential foods such as maize, beans, and chile
peppers (Gorenstein and Pollard 19832109).

Ethnographic sources record the taking of several important species of fish in the
Prehispanic period, including Chirostoma estor (Kuruco urapeii), Ch. borloni (choral),
Ch grondocule (k ’uere'pu), Algansea lacustris (okumaro), and Goodeidoe spp. (t 'iru)

(see Gorenstein and Pollard 19832171).

48

 

Fish provide important nutrients, and helped make the Lake Patzcuaro basin an
especially productive environment. Although lakeshore comprised only 15% of the basin
land area, it supported 69% of the settlements and 74% of the population in the
protohistoric period (Gorenstein and Pollard 19832153).

Gorenstein and Pollard note that “the species used protohistorically are well
distributed in the lake, some preferring littoral niches (e. g. C. bartoni), others the open
water (C. estor), and others the open deep water (C. grondocule) ” and estimate the
annual fish yield of Lake Patzcuaro as 4,732,800 kilograms (19832176-7). They further
point out that “Since the protohistoric lake was considerably larger, it provided more
open water which is the niche of the more valued and abundant fish (uropiti [ C. estor],
pescado blanco; cuerepu) ( l 983 2 176 -7).

Fish in general provide vitamins A, D, Bl (thiamine), and 132 (riboflavin)
(Rostlund 1952 in Cleland 19662196). Gorenstein and Pollard (19832173) calculate that a
100 gram portion of fish (C. estor, or pescado blanco) would have provided 67 usable
calories and a range of important nutrients such as niacin, iron, calcium, zinc. According
to The Concise Encyclopedia of Foods and Nutrition (Ensminger et a1. 1995), most fish
contain 18-20% protein, of which 85-95% is digestible. Dried choral, which could be
consumed whole, provided even more nutrition. Ortiz de Montellano calculates that each
fish in this form is comprised of more than 61% protein and contains three times the daily
requirement of niacin per 100 grams and a substantial amount of vitamin A (19902115).
Fish comprises a high quality, “complete” protein, meaning that it contains all the amino
acids humans need for proper nutrition (Ronzio 19972369). Fish meets or exceeds the

recommended daily intake of five of these amino acids (Gorenstein and Pollard

49

19832173). With a three ounce (85 gram) portion of lean white fish providing over 20
grams of protein (Pennington 1998:332-7), a modest portion of fish could make a
substantial contribution to daily protein needs (U.S.R.D.A. is 63 grams for adult males
and 50 grams for adult females -- see Ronzio 1997).

Ethnohistoric documents, such as The Chronicles of Michoaccin, record the
techniques used by the Tarascan fishermen and the types of fish taken, including the use
of nets, fish hooks, and baskets (1970:111,114,116-17, 131-2). Fishermen may not have
always specialized in a single technique, as The Chronicles of Michoacan contains a story
of a fisherman who fished at night with a net and by day with a hook (1970:114). Nets
could be cast from canoes on the deep, open lake or set up along the shore (19702127). It
is likely that the nets set up near the shore were used to catch Ch. bartom’, since this is the
species which prefers a littoral habitat. It is also probable that these are the nets which
utilized net sinkers of the sort analyzed in this thesis.

Lumholtz observed a multitude of methods for catching fish among the
Tarahumara of northern Mexico. Although the Tarahumara are not ethnically Tarascan,
their fishing practices are useful in illustrating the variety of fishing practices among
native peoples in Mesoamerica. These included catching by hand, the use of open mats,
spears, stone corrals, and other types of traps to trap fish, as well as poisoning the water.
Lumholtz also reported that during the summer the Tarahumara sometimes improvised a
net of sorts by fastening multiple blankets together and dragging the river. The
Tarahumara weighted the net by making a hem along the bottom about four inches deep

and filling it with sand (in Bennett and Zingg 1976:121-4). No net sinkers were used.

50

At the time of the Spanish Conquest, Tarascans harvested fish using fishhooks as
well as various sorts of nets (see above and below) (The Chronicles of Michoacan 1970).
Fish were generally prepared roasted, cooked, or sun-dried (Alcala 1980, in Guzman et
a1. 2001). It was also common to present visitors to the king, or cazonci, with gifts of
fish. The Relacz'on de Michoacdn relates how Cristobal de Olid requested several types
of fish upon his arrival in the Tarascan kingdom, among them, “cuerepo ” (Chirostoma
patzcuaro), “acumarani " (Algansea lacustris), “urapeti ” (Ch. estor), and “thiru”
(Goodea Iriitpoldi) (Girzman et al. 2001).

‘ The island of La Pacanda, in Lake Patzcuaro, was an important Tarascan
settlement in‘Pre'hispanic times, and, like many other Tarascan settlements in the lake
basin, engaged in fishing as an important economic activity. Fishing, along with farming,
remains an economic staple, and modem—day Pacandefios derive the bulk of their income
from selling fish and produce at local markets outside the island (Smith 1965218, 41).

Smith (1965) mentions several fishing techniques still in use at the time of his
dissertation, including chinchorro, cherémequa, and butterfly nets. He makes no specific
mention of net sinkers. However, as the reader will recall, Foster (1948:102-4) reported
that large, “unworked stones of 4-8 kilos” were used with modern chinchorro nets to fish
bass and small, “pebble” sinkers were used with the cheremékua nets. Since recent years
have been a time of rapid economic and technological change (Smith 19652168), it is
possible that fishing technology changed in the seventeen years separating Foster’s and
Smith’s ethnographies. Alternatively, Smith may not have reported them even though
they were still being used. It is also quite possible that fishing practices and their role in

the economy may be different from pre-Conquest times. For example, Sugiura’s

51

description of modern fishing techniques in the Toluca basin makes no mention of net
weights (Sugiura 1998), despite the fact that she has recovered several examples of
worked sherds from local archaeological sites which she believes to be Prehispanic net
weights (Pollard, personal communication). Further archaeological research is needed to
clarify this. -

Like the Pacandefios, the people of Quiroga, also in the Lake Patzcuaro Basin,
practiced fishing as an important economic activity during Prehispanic times (Brand
19511162). Unlike the Pacandefios, however, the inhabitants of 'Quiroga have IOst their
lake access rights in modern times and thus cannot fish or‘hunt waterfowl as they once
did; These items must now be purchased from their‘neighbors in' other’la'ke‘side towns
-such- as Santa Fe' and Tzintzuntzan. Fish, though now purchased,-are-animportant-source
of protein, and, particularly in the form of dried charal, also provide calcium for the
Quirogans’ diet.

Gorenstein and Pollard (19832122) provide population estimates for several
Tarascan empire administrative centers. They also calculate how many people in each
area could have been supported by a maize-only diet. The estimated populations of
Urichu and Pareo are both below what could have been supported by a maize diet.
However, the population estimates for the island of Xaracuaro are well above what could
have been supported by maize alone. The inhabitants of this island would have needed
supplemental food to nourish themselves adequately. Despite The Chronicles of
lt/[ichoacan’s assertion (1970213) that “The fishermen did not eat the fish but took it' all to
the Cazonci and the lords, for they ate nothing but fish,” fish seems a likely candidate for

this role, and would account for the large numbers of net sinkers found in the area.

52

Fishing remains an important activity among the modern Tarascans, as
documented by F oster (194 8) and Toledo (1991), although its range is perhaps more
limited. Brand reports, for example, that in present-day Quiroga, also in the Lake
Patzcuaro basin,

Very little hunting or fishing is done by the inhabitants of the Quiroga

area. Since they have no lake rights they neither catch fish nor hunt ducks.

These items are purchased from peoples of Santa Fe, Tzintzuntzan, and

other lake communities...This is far different from the conditions which

obtained before the coming of the Spaniards. . . In prehistoric

times... fishing for the various fish of the lake... (was) very important

(19512162).

At the time of Foster’s ethnography (1948), all fishing on Lake ”Patzcuaro was
done by Tarascans, and fishing was the third most important industry in Tzintzuntzan,
'one'of the-lake’smajor settlements(194821-01). 'Foster'discusses net fishing technology
at length, andalthough the ceramic. disks are considerably lighter, with a mean weight of
6.21 grams, than the 4-8 kilos he observed to weight the large chinchorro seines used to
fish bass from canoes (19482102), they may be analogous to the “tiny pebble sinkers” he
reports were used with a type of gill net commonly suspended between two poles near the
lakeshore and used to catch the young of the famous Tarascan whitefish, or khuerepu
( 1948: 104).

Although reduced in scale, fishing remains, nevertheless, an important economic
and cultural activity. Artisanal fishing is still practiced in Lake Patzcuaro; as of1‘993 it
still produced 29% of the fish harvest and contributed a full 80% of the fish consumed
locally (Toledo et at 19932135).

Of the-five'native-methods still-used, cherémecuanets‘currentlydominate, with

some 15,194 reported in use in 1986. They are generally 25-50 meters in length.

53

Because of their light mesh, cherémecuas cannot be used to catch all types of fish.
Chinchorro (literally “small boat”) nets are also widely used but are now made with
synthetic monofilament rather than cotton as in earlier times. Since the synthetic nets are
stronger (and probably lighter) than the cotton ones, it is possible that they are also made
larger than they were previously. They are comprised of two “arms,” with a central
pocket. The arms of the chinchorro nets currently range from 70-300 meters in length.
Their small mesh size (5-20 mm) enables them to capture all species in almost all size
ranges. The use of butterfly nets is in sharp decline. As of 1955 it was still the dominant
fishing technique, but 'by 1989, only 38 were left in use. Fishing hooks remain in use 'but
are of only minimal inrportance (Toledo et at 19932143 -4').

'Of four-Chirostomaspecies (Ch. estor, Ch. grandocule, Ch. attenuat'um, and Ch.
patzcuaro) which are native to Lake Patzcuaro, the first, known locally as pescado
blanco, or “Whitefish,” is distinguished from the latter three, collectively referred to as
charal, because it grows to sizes greater than 20 cm in length. The greatest recorded size
for the first few months of 1991 was 39 cm, although most average fi'om 12-18 cm
(Toledo et al. 19932137). Faunal remains of these fish have not yet been studied in an
archaeological context; research in this area would be useful in determining whether
these size ranges were similar in Prehispanic periods or whether modern conditions such

as fishing practices or the introduction of carnivorous bass has reduced their size.

Fish in Mesoamerican Ideology

Having thus demonstrated that the Tarascan disks most likely functioned as

fishing net weights, it becomes appropriate to examine the larger role of fish in

54

Mesoamerican culture beyond its economic importance. Establishing the ideological or
mythical meanings of fish in the cultural framework may assist in interpreting future
discoveries of fish remains in the archaeological record, although, of course, it will not
explain every one. For example, knowing that fish are an important symbol of the
creation of life in Mesoamerican cultures may enable the archaeologist to consider
multiple layers of meanings when he or she uncovers remains interred within a tomb.
These meanings may be quite different from those gleaned from the faunal remains of a
household refirse pit.

Fish play an important part in the mythology of Mesoamerica and adjacent areas.
From the Caribbean Taino, for example, comes this myth of the sacrificed son of god, as
reported by Joseph Campbell (19892315):

There was a man called Yaya, whose name we do not know. He
had a son called Yaya-e1, who. wanted to. kill. him, and. so, was. exiled. For
four months he was exiled, after which his father slew him, packed the
bonesin amortuary um, and hung the urn from the roof, where for some
time it remained.

Then Yaya one day said to his wife, “I would like to see our son,
Yaya-e1.” She was pleased. So, taking down the urn, he turned it upside
down to have a look at the bones, but what fell out were fishes, large and
small. And seeing that the bones had turned into fish, the couple ate them.

When one day Yaya had lefi his house to go to inspect his gardens,
the four sons arrived of a woman who had died in giving birth to them.
Itaba Cahubaba, she was called, The Blood-bathed Old (Earth) Mother.
She had been cut open and the four taken out... when Yaya had departed
fiom his house, these four entered to get the mortuary um...all were
gorging themselves on the fishes it contained when Yaya was heard
returning. In haste to hang the urn up, they hung it badly. The um fell,

broke, and out poured so much water that it flooded the earth. Many ‘fish
came too. The origin of the sea was from that urn. ..

This myth illustrates a clear association of fish and water with the origins of life and as an

important source of sustenance for people. As Campbell points out, the mythic

55

archetypes of a son sacrificed to or by his deity father, as well as the association of this
son with a meal of fish should be familiar to anyone familiar with the J udeo-Christian
mythic tradition (Campbell 1989:315).

In Mesoamerica proper, water and fish play important parts in Mayan creation
myths as well. For example, First Father in his Maize God aspect is both killed and
reborn in the waters of the sea (F reidel et al. 1993:92). The primordial sea, depicted
again and again in Maya art, is also “the place of Creation, where First Father raised the
sky and where First Mother modeled the flesh of humanity from maize dough and water”
(19932284).

Fish comprise one of the thirteen constellations used in the Maya zodiac (F reidel
et a1 219932102). In Maya mythology, the Hero Twins are credited with defeating death
and banishing the Xibalbans from the world. They also make possible the Fourth
Creation of the World, the one in which humans are created (19932110-11). When the
Maya Hero Twins, Hunahpu and Xbalanke (One-Ahaw and Yax-Balam) are killed by the
Lords of Xibalba (the underworld) in a raging fire, the jealous and murdering gods crush
their bones and throw the dust into a river. Five days later, the Twins emerge as catfish
men (1993: 109,348).

The Maya ballgame was performed as a ritual reenactment of the creation of the
world, and was played in order to maintain the balance of the cosmos. The ritual
ballgame was played “by elites and involved the sacrifice of human captives. As Friedel et
a1. explain, “Ballgame sacrifice renewed the basic covenant between gods and people”
(1993362). Art from these balleomts, which are ubiquitous in the Maya cultural area, is

rich with the imagery of Maya creation myths. For example, carved panels at the

56

ballcourt of Yaxchilan depict the ruler Bird-Jaguar wearing the net skirt of the Maize God
(First Father) and with a fish on his back, “reminiscent of the Hero Twins who first
resurrected with catfish faces” (19932360). Panels at the ballcourts of La Amelia and Dos
Pilas depict similarly outfitted rulers (1993:361-2).

The so-called Maya “fish-in-hand” glyph appears widely on Maya art and
architecture in the context of blood-letting and sacred rites. It is a verb which apparently
means something like “to manifest a vision or a divinity” (Schele and Freidel 19902473).
Again, there is an association of fish with something sacred, as well as with religious rites
which were performed with the objective of maintaining a balance between the earthly
and the divine.

Whether or not fish connoted these same meanings for the Tarascans is not
known. Certainly the Tarascans shared many Mesoamerican cultural traditions, such as
ritual ballcourts and human sacrifice, with their neighbors and predecessors. Kubler
(1967, 1970), drawing on Panofsky’s (1960:84-106) “law of disjunction,” which the latter
used to describe the discontinuity of meaning associated with similar iconographic
elements between classical antiquity and medieval Europe, cautions against assuming
that similar or even identical motifs canied the same meanings across time and/or space,
even within the same general areas. Speaking of medieval Europe, Kubler asserts that
“Continuous form does not predicate continuous meaning, nor does continuity of form or
of meaning necessarily imply continuity of culture” (cited in Nicholson 19762160). He
takes a similar stance on applying a direct historical approach to Mesoamerican
iconography (1976: 160). Nicholson, however, holds a more optimistic view,“ stating that

“Whether all major Mesoamerican groups participated in an essentially similar religious-

57

ritual system or not... it seems clear that at least a core of interrelated basic concepts was
widely shared” (19762162-3). At the same time, he admits that “. ..the whole question of
Mesoamerican Preclassic-Classic-Postclassic continuities in religious iconography
requires a much more thorough, comprehensive analysis than it has yet received”
(19762163). He states that

...iconographic continuity can be best established by careful determination

of similarity of images through time. And a single motif, it would

probably be...agreed, would normally have less value than a consistently

associated cluster of iconographic elements, the more complex the better

(19762163).

Even when iconographic continuity is established, however, it must also be demonstrated
that their associated meanings were also carried through time. Nicholson realizes that

Without the aid of coeval or otherwise relevant texts it is undeniably very

difficult to determine whether representations and symbols formally

similar but temporally widely separated actually did convey basically

similar meanings ( 1976: 170).

The problem of continuity of meaning is especially difficult to solve in the case of
the Tarascans, for not only did they lack a writing system, but the examples of fish to
which we are trying to assign a cultural meaning are faunal remains, not art icons.
Nevertheless, l believe that, at least as grave offerings, the Tarascan fish remains may be
viewed as the equivalent of an artistic depiction, as both are consciously constructed by
their makers. The fact that fish are consistently associated with the origins of life in
different cultures across a significant portion of time and space in Mesoamerica as well as
adjoining areas, however limited the above examples are, is at least enough to suggest

that the Tarascan grave offerings may have represented some similar idea. Thus, I would

agree with Nicholson (1976:173) that “the direct historical approach can be profitably

58

employed in Mesoamerican research to interpret. . . the iconographic systems of cultural
traditions... ;” further, I might suggest that this approach is not only appropriate, but may

also be the most fruitful one, if carefully applied.

Fish in Tarascan Ideology

Among the Tarascans, as is so often the case with important foods among other
cultures, strong linkages exist between fish as an economic staple and as an ideological
icon.

On the island of La Pacanda, responsibility for organizing the modem-day fiesta
of Corpus Christi (“El Corpus”) is shared by three formal groups comprised of most of
the adult men of the village. The three groups are known, interestingly, as Los
Chinchorreros (“those who fish with chinchorro nets”), Los Cherémequeros (“those who
fish with cherémequa nets”), and Los T rabajadores (“the workmen,” “those who work
the land”) (Smith 1965290). Although membership in the groups does not literally reflect
occupation (1965291), the groups do represent central economic activities of the islanders
(1965296). During part of the ceremony, for example, the Chinchorreros and
Chere'mequeros dance in a line with their namesake fishing net, trying to capture a large,
cloth fish effigy which dances in the middle of them. Although these activities are
obviously important to the economy of the islanders, they serve to illustrate that the line
between economic and symbolic activities is not clear-cut, and in fact may be
indistinguishable. Whether the symbolic import of fish and fishing among the modem-
day Pacandenos is due to the influence of Christianity introduced by the Spaniards or has

roots that extend deep into their Tarascan heritage is difficult to say, since both provide

59

viable precedents. Certainly fish and fishing play a part in both Christian and Prehispanic
Tarascan mythology. For this reason, perhaps the source of the modem-day practices is
irrelevant. It may be just those Christian ideas and symbols which had close parallels in
their own culture that the Prehispanic Tarascans were most likely to identify with and
incorporate into their new, changing culture. Certainly a similar process occurred when

the Maya became Christianized (see Friedel et al. 1993).

Fish in Tarascan mythology

In comparison with other Prehispanic civilizations, relatively little is known about
Tarascan religion and mythology. Much of what is known to us has come from The
Relacz’on of Michoacan. Modem-day ethnographies and archaeological work have given
us additional insights. Nevertheless, significant bits and pieces can be gleaned from these
various sources and are sufficient to prove that fish did play an important role in Tarascan

mythology and ideology.

We do know that the island of Xaracuaro (one of the three sites discussed in this
paper) was one of eight religious centers of the Tarascan empire (Chronicles of
Michoacan). According to Gorenstein and Pollard, “All were clustered around the lake
(Patzcuaro) at sites which were highly visible to the settlements at higher elevations of
the Basin slopes,” with Xaracuaro serving the west (1983: 123-4). On it was a “great
temple” (Chronicles of Michoacdn:l 13). The island’s inhabitants were referred to as
“hurendetiechan, " (Joaquin 2000:711) from “hurendi, ” (sage or knowing); thus
hurendetiechan = “the wise ones from the island” or “the people who were here first”

(Pollard, personal communication 2001). Xaracuaro itself was known as “Varucaten

6O

hazicurin, " from varun, “island,” varucata, “a fish thing,” and hazicurin, “on a high
place.” Joaquin (20002721) thus translates “Varucaten hazicurin” as ‘On the place of the
fish.’ We also know from the archaeological evidence (abundance of net sinkers) that

fishing was a central activity of the island’s inhabitants.

Xaratanga was the Tarascan moon goddess (Pollard 1993:134,l37); she was also
associated with childbirth and fertility as well as with the sea (19932134), lakes, and
marshes (19932151). Conch shells (a symbol of the primordial, life-giving sea?) were
used in worshipping her (19932134). The Tarascans believed that she had the power to
conceal fish from the fishermen (19932137). Significantly, among the Tarascans the
worship of Xaratanga was associated with the ritual ball game, which, among the
Mayans, had clear and documented associations with the sea and with creation.
Depictions of vanquishing nrlers with fish on their backs occur at several Maya
ballcourts, including Dos Pilas, La Amelia, and Yaxchilan (Freidel et al. 1993:360-2).
Xaratanga shared several of these associations, then: water, fish, creation, sacrifice, and
life and death. As documented above, this grouping of symbols and meanings occurred

elsewhere in Mesoamerica and the Caribbean.

Xaratanga’s major temple was at Xaracuaro, although it was moved in the
Postclassic, most likely as a result of the merging at this time of earlier, native religious
beliefs with those of the ethnic Chichimecs, who founded the Postclassic Tarascan royal
dynasty. The mythology chronicling this process yields interesting clues to the Tarascan
culture, economy, and political process of state formation. According to The Chronicles
of Michoacan (Craine and Reindorp 19702113-20), one evening some lords of Curicaveri

encounter a fisherman named Curiparaxan from xaracuaro who is fishing from his canoe.

61

The lords confront the fisherman, who in turn builds a fire and feasts them on his catch,
which the priests find very tasty. In talking with the fishermen, the lords discover that
they, too, are related to the gods worshipped on Xaracuaro and Pacandan. They ask the
fisherman if he has a daughter and arrange to meet her. They subsequently take the
daughter away to raise her, predicting to the fisherman that she “will conquer this land,
and you for your part will stand with one foot on the land and one on the water”
(19702116). This prophecy echoes not only the dual importance of agriculture and
fishing to the Tarascan economy, but also the merging of two separate cultural and
religious systems, the indigenous Tarascan and the ethnic Chichirnec. The myth records
the Chichimecs’ founding of the royal dynasty and the formation of the Tarascan state:
The fisherman’s daughter grows up and marries one of the Chichimecs, Pavacume, and
then gives birth to the legendary T ariacuri, who becomes the first ruler of the Tarascan

state. It also describes the merging of two separate religious systems. As Pollard relates:

The forging of the centralized state after AD. 1300 required the creation
of a common language of cosmic forces. The patron gods of the dominant
ethnic elite were elevated to celestial power, while the various regional
deities and worldviews, which were themselves products of generations of
change, were elevated, incorporated, or marginalized (19702134).

The myth is central to understanding the history and culture of the Tarascan empire.

Other mythic references to fish in The Chronicles of Michoacan include a story
about some priests of Xaratanga who drink too much bad wine and send some serving
women to catch some fish to cure their hangover (Craine and Reindorp 1970:111-13).
The women try in vain but “could catch no fish because Xaratanga, who was such a great
goddess, had already concealed them” (1970:112). Not wanting to fail their masters, the

women substitute chopped up snake and make a stew for the priests. Not surprisingly,

62

the snake stew does not produce the anticipated cure. This myth seems to suggest that
the Tarascans believed that fish possessed either some real or symbolic healing qualities.
In addition, it documents Tarascan elites’ consumption of fish. The Chronicles of
Michoacan (Craine and Riendorp 1970:12-13) states that the Varuri was the government
official in charge of all the net fishermen who supplied fish to the Cazonci and all the
lords, who ate “nothing but fish.” Fish would have been an obvious nutritional
supplement to an elite diet, but could this consumption also have had religio—symbolic
value which reinforced their elite status? Given fish’s religious associations, it is a
possibility which cannot be overlooked and which deserves further investigation.

Fish in Tarascan burials ’

Fish remains have been found in the tombs of Epiclassic-Early Postclassic elites
at the site of Urichu (Pollard and Cahue 1999; Guzman et al. 2001), attesting to their
cultural importance. Whether this importance was confined to their nutritional aspects or
also encompassed ideological meanings remains to be explored. Thus, the ethnographic
present can be compared with ethnohistoric records and the archaeological record so that
similarities and differences between Prehispanic and modern Tarascan fishing practices
- can be explored.

One Classic-Epiclassic period (A.D. 350-1000) tomb of Urichu elites has yielded
fish remains (Pollard and Cahue 1999; Guzman et al. 2001). One internment was an
adolescent of undetermined sex; the other contained two adult males. Tripod bowls
containing fish bones and scales were found in both burials. Although the adolescent is
of undetermined sex, it should be noted that this burial lacked grave goods typically

associated with women, such as spindle whorls and shell beads (Pollard and Cahue227 5).

63

 

Thus, it is possible (but as yet undetermined) that all three individuals associated with the
fish remains were male. Whether or not this is significant or merely coincidental is
unknown at this point. Archaeological evidence of the Tarascans’ use of fish is
extremely rare; these tombs represent the only example of fish remains found with human
burials and only the second example of interred fish remains at all (Guzman et al. 2001).
The other known example is the discovery of Whitefish (Chirostoma sp.) buried at the
yacatas in Tzintuntzan dating to the Late Postclassic (Martin del Campo 1946; Polaco y
Guzman 1997, in Guzman et al. 2001:152). The sample size is extremely small and
would have to be enlarged significantly before any reliable patterns could emerge.
Pollard and Cahue state that elites of this period derived their status partially
“from their participation in the macroregional exchange system by which (elite) goods
were obtained; their importation as finished goods reveals the dependence of these elites
on social/economic power generated in other parts of western Mesoamerica” (19992275).
Perhaps the fish remains represent a locally harvested commodity which was traded
outside the immediate area in exchange for symbolically valuable goods from other
regional centers which in turn reified and enhanced the status of the Urichu elites.
Alternatively, the elites’ burial association with a locally prized fish could foreshadow a
trend in which their identity and status became “primarily associated with locally
produced, distinctively Tarascan, goods and control of tributary, military, political, and
ideological networks” associated with the emergence of the Tarascan state in the
Postclassic (A.D. 900-1525) period (Pollard and Cahue 19992259). If so, it is, indeed, an
early example of this trend. Since the burials contain other elite status items, such as

polychrome ceramics, it is highly unlikely that they reflect an individual’s occupations as

64

a fisherman. According to Pollard and Cahue, the tomb belongs to one local elite with
local heads of high rank.

The fish remains in both burials are of the locally abundant fish known as “t’iru”
in Tarascan (Pollard and Cahue 1999:270; Guzman et a1. 2001:155-6), or scientifically as
Goodea luitpoldi. This fish is also referred to as thiron in the Relacr'on de Michoacan.
This is rather curious in light of the fact that it is the members of the Whitefish family
(Atherinidae: Chirostoma bartoni, C. estor, C. grandoculae, and C. patzcuaro), or
pescado blanco as they are referred to collectively, which are well-documented as
economic and cultural staples of Tarascan society, both Prehispanic and modern
(Guzman et a1 2001). The tomb remains thus represent the first archaeological evidence
of the use of t 'iru by Classic and Epiclassic period elites, as well as attesting to the
importance of fish during the Classic period. F urtherrnore, this tradition spanned at least
400 years, as the remains associated with the two adult males date to between AD. 405-
505, and the fish remains associated with the adolescent burial yielded radiocarbon dates
between A.D. 800-1000 (20012159). Guzman et al. (20012160) suggest the possibility
that the Classic and Epiclassic Tarascan preference for t ’iru was replaced by one for
Whitefish (Chirostoma spp.) by the Postclassic period. Future research will be needed to
substantiate this as well as explore the reasons for the change, if it indeed exists.

Although it was not possible to determine the method of preparation of the fish
remains (e. g. cooked, dried, etc), it is known that they were deposited in the tomb whole,
since their stomach contents were found. This can be contrasted with the usual culinary
preparation of fish in the town of Ucasanastacua, where the t’iru are eviscerated before

being dried in the sun (20012160). Guzman et al. suggest several possibilities for the t’iru

65

remains uncovered in the tomb: perhaps in Prehispanic times these fish were consumed
whole, or the fish were placed in the tomb soon afier being captured, without time taken
for the usual preparations. Alternatively, perhaps the fish were not meant to be food
offerings but instead represent something symbolic (2001:160). The fact that the remains
are of t ’iru, and not of the Whitefish which were so important as a food resource, may
support this hypothesis. Alternatively, t ’iru may have been consumed only by elites and
not commoners; although the fact that they were consumed by elites is not inconsistent
with the hypothesis that they had some symbolic value. Apparently, great care was taken
in selecting the offerings, as the fish remains in each tripod bowl were remarkably
homogeneous in size (2001:160).

Future discoveries of this kind should be extremely useful in reconstructing
temporal and/or spatial trends in fish remains associated with elite burials, as well as in
shedding light on the economic and symbolic importance of fish in the Tarascan

civilization.

66

SUMMARY AND CONCLUSIONS

This is the first systematic study of the unique class of artifacts which were used
as net sinkers in Prehispanic Tarascan society.

The statistical as well as the available ethnographic evidence strongly suggest that
the most likely function of the ceramic disks examined in this paper was sinkers for
fishing nets. Any single line of evidence considered individually is not sufficient to
support this hypothesis, but taken as a whole, the totality of available evidence makes it
overwhelmingly likely.

As discussed above, the disks are distributed in highest concentration along lake
areas, at least in the survey zones of Pareo and Xaracuaro. Urichu breaks with this
pattern, for reasons unknown. Perhaps the ones found in the environs of Pareo and
Xaracuaro were lost while in active use during fishing, while those found at Urichu were
discarded or lost at a different stage in their life cycle of use.

Double side notches are clearly the norm, a morphology which makes eminent
sense for net sinkers, as it would enable them to easily be tied to a fishing net. So well-
suited are they to this task that modern fishermen sometimes (re-)use them in this
capacity when they come across them. This paper establishes the fact that there are major
morphological differences among malacates, gaming pieces, and. net sinkers, and
documents these differences systematically.

Since this is the first systematic study of Tarascan disks, it is difficult to say what
constitutes a large concentration. However, as a ratio to other ceramic sherds, the disks

are found in substantially higher concentrations at Xaracuaro and Pareo than they are at

67

Urichu. They are also found in larger overall quantities at Xaracuaro and Pareo than at
Urichu (see Figure 13). Xaracuaro yielded by far the most disks (324). Quite likely this
reflects the fact that all areas of this zone are in close proximity to the lakeshore (some
were even under water during the Tariacuri phase) and thus all parts of this zone would
have been adjacent to fishing areas. It is also likely that during the Tarascan period the
island was home to a substantial number of fishermen, who would have had sheds for
storing their nets and/or other areas for manufacturing and repairing them. The quantity
of disks is also relatively high at Pareo (125), whose northern portion bordered the lake
and would also have been eminently suitable for fishing. In fact, at Pareo the disks are
concentrated almost exclusively along the northern part of the zone. Less clear is why
Urichu, which also has a substantial area bordering the lakeshore, yielded relatively few
disks (73). Several portions of the zone, including the location of modern towns, were
not excavated or surveyed, and perhaps this accounts for some of this pattern. However,
large portions of the zone which directly border the lake were investigated and yielded
few or no disks. It is possible that fishing was not an important activity at Urichu, for
reasons which have yet to be determined.

The disks’ size and weight are also suitable for functioning in the capacity of net
sinkers, and are likely to be analagous the “tiny pebble sinkers” Foster documented used
with the cherémekua nets used to catch khuerepu and pescado blanco (1948:102—4). As
posited above,_it is likely that the fish harvested in this manner were of the species Ch.

barton i.

68

The unimodal distribution of the disks’ diameter, weight, and thickness suggests
that there was a large degree of uniformity in either net size and/or the size of the meshes
within the nets.

T-tests performed in order to compare the Tarascan disks with known populations
of malacates were somewhat indeterminate in their results. T-tests comparing the
diameter means of the Tarascan disks with those of malacates found during the same
investigation (the Proyecto malacates), as well as those from Tzintzuntzan, determined
that the disks represent a separate population, and thus in all likelihood a separate artifact
class. However, t-tests for the variable of thickness found no significant differences in
means among these three samples. This may contradict the separate artifact distinction
but it may also just show that the two artifact classes are very similar in this respect and
thus identification must rely on other factors and measurements. Further complicating
the picture, a t-test comparing weight means for the Tarascan disks and the Proyecto
malacates found in the same investigation rejects the null hypothesis with a one-tailed
test and fails to reject it with a two-tailed test. Again, this may argue against dividing the
artifacts into two separate classes or it may demonstrate that the Tarascan disks are very
similar in weight to malacates used to spin cotton. The t-test results demonstrate the
importance of using as many lines of evidence as possible, for if size and weight are the
only criteria used it would be very difficult indeed to separate the disks from the
malacates.

Although the statistical and ethnographic evidence strongly suggest that the
ceramic disks recovered during Pollard’s archaeological research were used to weight

Tarascan fishing nets, several questions and directions for future research remain.

69

As touched upon above, considerable confusion exists in the literature when it
comes to recognizing net sinkers in the archaeological record. I suggest that the
following criteria be met in order to uniformly classify artifacts as net sinkers in future
archaeological work:

1. Artifacts are constructed of a suitably weighty material (stone in aboriginal

United States; ceramic in Mesoamerica).

a. Ceramic artifacts are fashioned from re-worked pottery sherds, not fired
from wet clay (i.e., they are a “secondary use” artifact).

2. Artifacts exhibit opposing, double-sided notching, a morphological

characteristic which facilitates their attachment to the fishing net.

3. Artifacts are roughly circular, or, less commonly, rectangular with rounded

comers, in shape.

4. No central perforation (diagnostic of Mesoamerican spindle whorls) is present.

5. Artifacts fall within baseline size parameters established for the collection of

net sinkers analyzed in this paper (see Table 2 above) (at least for Mesoamerican

examples):

6. Artifacts are found within or in close proximity to lacustrine or riverine areas.

7. Artifacts were made and used by a society which had a focused (in Cleland’s

sense — see Cleland 1966) or heavy reliance on fish as a food resource and which

had a highly developed fishing technology.

This should help distinguish net sinkers from artifacts which have physical
similarities but which functioned differently (e.g. as bola weights or malacates) and thus
rightfully belong to other classes. For example, the size, weight, and morphology of the
ceramic disks analyzed in this paper can provide baseline measurements against which
future discoveries can be compared. Those falling within the range of the Tarascan disks,
at least in the Tarascan cultural area, should be considered likely net sinkers, especially if
they are found in lacustrine or riverine contexts.

It is most ironic that, although it is well-known that the Prehispanic T arascans

were skilled fishermen and that they used vertical nets, and that fish comprised a staple of

their economy, net sinkers have never been officially identified in the archaeological

7O

 

record of this area or studied systematically. This thesis firmly identifies and documents
them for the first time as a unique and important artifact class.

Since it is not possible to observe the behaviors of past cultures, it is the-lot of the
archaeologist to infer behavior from material remains. With net sinkers firmly identified
in the archaeological record for the first time one can now use” this artifact class as a
proxy for the Tarascan economy, at least for a part of it. Although fishing comprised a
major portion of the Tarascan economy, it appears that the net sinkers analyzed in this
thesis most likely reflect the harvest of only one species of fish, Ch. bartoni.
Nevertheless, this is an important contribution to the study of the Tarascan economy. The
fishing of other species may not be possible to document directly through archaeological
remains. It is now up to future research to document the patterns and fluctuations of this
fishing economy.

One area which needs clarification is the channels of fish distribution within the
Tarascan empire. Gorenstein and Pollard (1983) have documented that fish were
distributed through the market system but what is not entirely clear is to what extent the
government controlled this flow. It is also quite likely that fish comprised a major tribute .
commodity, but again, the quantities involved remain to be determined.

How much fish and of which species were consumed by elites versus commoners
is also a question for future research. As noted above, The Chronicles of Michoacan
(Craine and Riendorp 1970:12-13) state that the Varuri was the government official in
charge of all the net fishermen. Were these net fishermen required to surrender the
totality of their catch to support the ruling class or just a portion of it? Were they free to

exchange some of their catch within the market system? Were there government-

71

supported fishermen and “free enterprise” fishermen or did all fishermen split their
harvest between tribute and the market system? These questions beg clarification.

Many other sites remain to be explored and could yield fruitful information. For
example, the island of La Pacanda, which has a fishing tradition dating back to
Prehispanic times, should have an equally rich archaeological record. It is at just such a
site that one could expect to find net sinkers similar to those found by Pollard et al. and
analyzed herein. In addition, La Pacanda and similar sites can serve as ethnographic
models for what one might expect to find archaeologically regarding fishing practices.
For example, at least some of the households on this island include small sheds adjacent
to the lakeshore, in which fishing nets and other equipment are lefi overnight during the
busiest part of the fishing season (Smith 1965:23). Are remains of these sheds found
elsewhere in the Lake Patzcuaro Basin? It is likely that they could be, and that, if they
are found, they might be a rich source of fishing artifacts, including fishing net weights.
Although fishing is no longer practiced in Quiroga, it too is just one of the many
communities who engaged in it heavily during Prehispanic times and where one could
expect to find archaeological evidence, including net sinkers, of its practice.

The identification of net sinkers in the Tarascan archaeological record is an
important piece of the puzzle of Prehispanic Tarascan life, but it is only a starting point.
It is my sincere hope that the research contained herein will enable future investigators to
document the Tarascans’ economy and culture more fully and in greater detail than ever

before.

72

APPENDIX

73

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0m(DO(m(w Nd we MN 22 >> meN
Om(30(m(w v._. m war 2.. m NNNV
0m(30(m(_. wN m6. m.om 22 m NNNe
DICE: .w oé ...m NwN 22 m mva
_m30(_m(._. YE. Nd man 22 >> EON
.m30(_m(._. F 5m mar 22 >> mmoe
(>0..(i_ - mam: NN Em QQN 22 >> Nwmw
(>021. - Nani. m6 v.0 ©.mN 22 >> nmmr
.m30(_m(._. (m 0w ©.mN 22 >> mmVN
.mDO(_m(._. mw m adv ZN >> mQVN
aux—S— Ymr N.» Nov ZN >> 33
DmX=2 md 9N Yum UQEU m vav
(>0..(i_ - mug: MN m.m VN Zr m wNNF
OmX=2 mK m.m m.mv 22 m FQVN
DIG—m3 .m Qm m6 v.0N 22 >> 92
_m30(_m(._. N6 90 NN 22 >> 5 E.
.mDO(_m(.r v m m.mN 22 >> owe
_mDO(_m(._. new m mNm ZN >> 05F
.m30(_m(.r mg...“ N.m ®.oN 22 m noov
.m30(_m(i_. m we @.mN 22 >> 500—.
(>033 - wan; 0d 90 m.oN 22 >> mNVN
_m30(_m(.r ad Tm ohm ZN >> wwVN
(>055 - mam... Nd m6. em 2? m O©VN
(>0_..(i_ - mini. 0 0% 5mm 22 m NomN
(>0q.(4 - No.34 ®.NF Nb wNm ZN >> NomN
310E: .w mN ad 0.3 22 >> NQVN
mwmcm um“ ”Eggs NEE“ wwwcxnxr: NEE“ LmaQEEU mocOuOC COEUCOO aaumo

 

 

Sun in: .35 532a... .v 2%...

74

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

DICE: .m m... we we 22 >> mt;
DICE: .w v 59 0.5 22 m mt;
:ICED .m N m? mdm 2N >> mwvm
EDC<E<F To me awn 22 m mot
EDC<E<F 7: Ya m. 5 ZN >> 8:
EDC<E<h ms Na 08 z_. m Nu:
OE<DC<E<w we Iv Q? 22 >> va
OE<DC<E<w we fim :3 22 >> 99
DICE: .m 99 ad 95 2F >> 38
DICE: .m ma ma mdm 2N >> mmvv
E:C<E<._. Om GNP m: 22 >> $2.
EDC<E<._. me o 5mm 22 >> om:
(>03) - mED: v Wm 5‘ 22 >> om?
E:C<EE. me In men 22 m NF 3
EDC<E<._. ad Em v.5 2F m NF:
OE<DC<E<5 vN we v.8 22 m Rmv
OE<DC<E<w we Em 5mm 2.. m ”.me
OE<DCE<w 9m 9v m.mm 22 m 53.
OE<DC<E<= we we 6mm 2N >> 83.
DICE: .m vN v Yam z_. m mmvm
DICE: .m me? Go 9 22 >> vwvm
DICE: .m ms v: 5mm 22 m vovm
EDC<E<._. md on w.mm 22 m 53.
EDC<E<k N v vamm 22 m 83.
EDC<E<H me me mew 22 m move
DICE: .w v.5 Na we: 22 >> 59
$ch By Ego; AEEV mmocxo_£ AEEV negoEmfi 8:80: 5:650 mofifio

 

 

 

 

 

 

 

 

43:8. 4 29:.

75

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

EDC<E<P v.m Im m. N 2N >> 8mm
EDC<E<.F Iw 9m Ivm zN >> vmmm
EDC<E<._. N.v md mdN zN >> mmmm
EDC<E<F mN On 08 zN >> mmmm
EDC<E<._. m. E o N.v zN >> 8mm
.EDC<E<._. NF Iv I3 22 >> 53
EDC<E<F me v.m a? 22 >> 38
EDC<E<H me man me 22 >> Nmmm
OE<DC<E<w 3 Im mm 22 >> vm Io,
EDC<E<h oN Iv o? 22 >> ommm
EDC<EE. IF Im we 22 >> vmwm
EDC<E<P Nd? Iw Inn 22 m omwm
EDC<E<P we N.m N9 22 >> Ema
EDC<E<h N wd N 22 >> vam
EDC<E<H md m Ian 22 >> mmwm
DICE: .m m mv odm 22 >> at
DICE: .m mN an N 22 >> NR
DICED .m Em 0m wan 2F m mt
DICE: .m or.” m new 2F m vwmr
DICE: .m v.o In mm: mam
EDC<E<F o9 \. wv 22 m wvwm
EDC<E<H me 3. N9 22 >> 88
mEDDES. ve 0... mg 22 >> :3
mmDDEi. 50.. v.2 5mm 22 >> K3
(>033 - mnD: v.0 mv mv 2F m mer
DICED .w N.v ms Nm 22 m vat.
omega 8v Ego; AEEV mmocxoé AEEV $6Em6 9220: 52950 @058

 

 

43:8. v 2%...

76

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

DICE: .w nN n.n NnN 2? m NNnn
DICE: .m In In n.nn 2N >> NNnn
DICE: .m 3: n N.Nv 2N >> NNnn
DICE: i. Iv N.m N.oN 2N >> vnnN
(>02... - MED: an n.n nN 2N >> nnnv
(>021. - $5.. on On nen 22 >> NnnN
DICE: .._ n.n on v.Nn 2N m ONnN

N N n.v vn 2N >> nnnN
DICE: .._ v.n on N.NN 22 >> ann
DICE: .._ N.v N v.vn 2N m anN
DICE: .._ N.N n.n N.oN 2N m nnnn
DICED ... In En I5 22 >> nnnn
DICE: i_ nv N.n ON 22 m nnnn
DICE: ... In nv Ion 2N >> nnnn
DICE: ... n.N Nn Inn 2N >> nnnn
DICE: .._ 5 IN v.nn zN >> nnnn
DICE: i_ n.:. IN nn 2N >> nnnn
DICE: i. EN En N.nN 22 m Nnnn
DICE: a. m an v.NN 22 >> 58
EDC<E<N we on ndN zN >> nNnn
EDC<E<N on N n.NN 2? m Nnnn
EDC<E<N Iv m.n nvN 2N 2, Nnmn

, EDC<E<N IN an v.oN 2N >> nvnn
EDC<E<._. nn v vdn 2? m nNnv
EDC<E<N nN n.n nNN 2F >> Nnnn
EDC<E<N v.N vv nN zN >> Nnnn

82¢ a: £995 AEEV nnmconD AEEV 56:66 8:20: 8238 @298

 

 

 

 

 

 

 

 

$.28. v 29:.

77

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

DICE: ... En n.n nvn ZN m nnnn
DICE: i_ n.v N.v v.nn 2F m nnnn
0E<DCE<n ...N N.v m. N ZN >> rvnn
0E(DC(E<_. ...N on neN ZN >> Fvnn
0E<D 0692.. In v v.vN ZN >> Fvnn
0E(DC(E(.. n.n n.n NN ZN >> Fvon
0E<DCE<n n.n n.n on ZN m Pvnn
DICE: i. Iv N.n m.mN 22 m oNNN
DICE: i. on n.n _.v 2.. m oFNN
DICE: ... I? an ndn ZN >> oFNN
DICE: ... v.v an. EmN ZN >> onNN
DICE: .m n.n av n.nn 22 m mmnN
DICE: i_ n.n N.n InN ZN >> nnvn
DICED ... N.nP En v.Nn ZN >> worn
DICE: ..D N.v n.n nNN ZN >> nnnN
DICE: ..D n NC N.NN ZN >> nnnN
DICE: ... N.N_. NC nvv ZN >> vvnv
DICE: ... En In N.nN ZN >> mvNN
DICE: .4 En n 5 22 m oFNN
DICE: ..D n.v an n.mN ZN >> m_.NN
DICE: ... v.n n n.nN ZN >> nNNN
DICE: i. an In va ZN >> nNNN
DICE: .w vN av mvN ZN >> NNnn
DICE: .w 0N n.n n.vN ZN >> NNnn
DICE: .m. n.v v v.nn 22 m NNnn
DICE: .m n.n N.v n.mN ZN >> NNnn
893 EC £995 468: $0525 NEED coDmefi $56: 8228 9.38

 

 

 

 

 

 

 

 

.828. v 2%»

78

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

DICE: ... n.N N.v N.NN ZN >> oPnn
DICE: ... n.N N.v vN Zr m ann
DICE: ... n.v n.n n.vN ZN >> opnn
DICE: .._ n.n v.n n.nn ZN >> opnn
DICE: .._ n.n n ndv 22 m ownn
DICE: ... n.N N.n EnN ZN m nnmN
DICE: ... v.N n NN 22 m nnmN
DICE: ... n v nN ZN >> nnnN
DICE: .._ n.v nv v.nN ZN >> nnnN
DICED ... n.n Ev nén ZN >> nnmN
DICE: ... n.n? n.n n.vn ZN >> nnmN
DICE: ... N.nF n.n INn ZN >> nnnN
DICE: ... n.N N.n ov 22 m nnmN
DICE: ... Q: v.n Eov ZN >> nnmN
DICE: .._ v.v Iv N.NN ZN m ovnn
DICE: .D n.N n.v n.NN ZN >> ovnn
DICE: 5. n.N n.n n.nN ZN >> ovnn
DICE: ... N.v N.v n.nN ZN >> ovnn
DICE: ... v.n N.n InN ZN >> ovnn
DICE: ... N N.n n.Nn ZN >> ovnn
DICE: ... n.NN N.n n.nn ZN >> ovnn
DICE: ... n.nN n.n n.nv ZN >> ovnn
DICE: ... 3. In N.vn 22 m ann
DICE: 5. v.2 In n.nn 22 >> ann
DICE: ... IN v.v n.nN ZN >> nnnn
DICED .._ n.N n.v n.nN ZN >> nnnn
8ch a: £995 3.330525 NEE: .3956 3220: 529.8 0.98

 

 

 

 

 

 

 

 

43:8. v 2%;

79

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

DICE: ..D N.NN n.n n.nv ZN m nnnn
DICE: ... I? n v.vn E. m nnnn
DICE: n.nN In N.Nn ZN >> onnn
DICE: n.n N.v va ZN m onnn
DICE: n.N N.n n.nv ZN m onnn
DICE: n.n In n.nN ZN 2, Oan
DICED N.n v.n n.Nv ZN m Oan
DICE: ... n.n n.n v.nn E. m mvnv
DICE: .._ n.n v.n n.nN ZN >> 85
DICED ... v n n.nN ZN >> nnFn
DICE: .: n n.n N.vn ZN >> nnnn
N N.n N.n v.mN 22 m Nnnn

N v.9 Iv N.Nv ZN >> mnnn
DICE: n.N v.n N.NN ZN >> van
DICED n.N N.n N.NN ZN >> van
DICE: n.N v n.NN ZN >> van
DICE: N.N N.n n.nN ZN >> van
DICE: N.N N.v n.vN ZN m van
DICE: n.n n.n nN ZN >> van
DICE: v.n n.n n.NN ZN m van
DICED 2. v.N n.n n.NN ZN >> nnnn
DICE: ... n.N n.n EnN ZN m van
DICE: ... n.n v.n v.NN ZN >> vwvn
DICED 3. n.n nv InN ZN >> vvvn
DICE: ... n.n n.v N.nn ZN >> vrvn
DICE: ... IN nv ION ZN >> ann

8ch 8: £995 NEEV mmmcxog AEEV .mfiEmfi 3:20: 5:650 @288

 

 

 

 

 

 

 

 

43:8. w an:

80

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

DICE: i. v.N n nN ZN >> Nnnn
DICE: ... N.N N.n n.nN ZN >> Nnnn
DICE: ... v.N n.v n.NN ZN m Nnnn
DICE: ... IN n.n N.nN 22 m Nnnn
DICE: i_ v.N n.n n.vN ZN >> Nnon
DICE: ... n.n v.v v.nn ZN >> Nnnn
DICE: ..D v.N N.v n.NN ZN >> 38
DICE: ... n.n n.v N.nN ZN >> nnnN
DICE: ... Iv n n.nN ZN >> nmnN
DICE: ... n.v n.n v.nN ZN m nnnN

N n.N n.v n.NN ZN >> nNnN
DICE: ... n.v n.v IN_. 2? m nnnn
DICE: i_ IN n.n mw ZN m nnnn
DICE: ... IF n.N n.n? ZN m nnnn
DICE: ... N.N n.n n.NN ZN >> nnnn
DICE: i_ n.v N.n v.nN ZN >> nnnn
DICE: ... v.N N.v n.NN ZN >> nnnn
DICE: .._ N.N n.n n.NN ZN >> nnnn
DICE: ... n.N In ndN ZN >> nnnn
DICE: J N N.v mdN ZN m nnnn
DICE: .D N.n n.n IvN ZN >> nnnn
DICE: ... v Iv vn 22 m nnnn
DICE: ..D n.n n.n n.nn ZN >> nnnn
DICE: ... n.NN v.2 N.Nv ZN >> nnnn
DICE: ... N.NN N.n Nv 2.. m nnnn
DICE: ... n.n? v.N v.nv ZN >> nnnn

$93 a. .fiE; NEE. nnmcon. €35 EXCEED 8220: 859.8 W058

 

 

 

 

 

 

 

 

.328. v 2.3.

81

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

82

 

DICE: .... n.N N.n n.nn ZN >> vmnn
DICE: ... new v.N n.nn 22 m vmnn
DICE: .._ n N.n n.nv ZN m vnnn
DICE: .._ N.n n.v n.Nn ZN m nnnn
DICE: n.n v.n InN ZN >> mnnn
DICE: v v.n N.nN ZN >> mnnn
DICE: n.n Iv n.nN ZN >> mnnn
DICE: n.n n.v InN ZN >> mnnn
DICE: n.N Iv n.NN ZN >> Pnnn
DICE: n... v n.nN ZN >> ..nnn
DICE: n.N N.n NN ZN >> Nnnn
DICE: n En v.vN ZN >> Pnnn
DICE: n.N n.n v.nN ZN >> 53
DICE: N.N v n.nN ZN >> Nnnn
DICE: v v.n n.nN ZN >> 53
DICE: n.v n.n N.vN ZN >> Nnnn
DICE: n n.n n.NN ZN >> 53
DICE: N.n N.n v.NN ZN >> ann
DICE: N.n N.n ndn ZN m ann
DICE: n.v n n.nN ZN >> Nnnn
DICE: N.NN n.v_. Eov ZN >> Pnnn
DICE: n.vw n N.nv . ZN >> ann
DICE: i_ N n.n n.NN z. m Nnnn
DICE: ... N.N n.v N.NN ZN m Nnnn
DICE: ... n.N n.n n.vN ZN >> Nnnn
DICE: ... n.N n.v N.NN ZN >> Nnnn
omega Damage; AEE. mnoconD ASE. .mEEmfi 3:20: 8:650 @958

 

 

 

 

 

 

 

.528. v 28..

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

DICE: .._ n.n n n.NN ZN >> vvmn
DICE: .._ N.n_. . ndw n.nn ZN >> vvnn
DICE: .._ N.n? n.n n.nv 22 >> nvnn
DICE: .._ n.n n.n n.nN 22 >> Onnn
DICE: .._ n.N N.n EON ZN >> nnnn
DICE: .._ In n.v N.nN ZN >> nnnn
DICE: .._ N.n Iv N.nN ZN >> nOvn
DICE: ... n.n n.n n.nN ZN >> nOvn
DICE: .._ v.n n.v nN ZN >> nOvn
DICE: .._ N.v n.n n.n? 22 m vmnn
DICE: .D n... N.n ION ZN m vmnn
DICE: .._ n.N n.v n._.N ZN >> vmnn
DICE: .._ n... n.v nN ZF m vmnn
DICE: .._ n.N In N ZN >> vmnn
DICE: .._ n.N n.v nN ZN >> vmnn
DICE: ... N.n Iv n.NN ZN >> vmnn
DICE: .._ In N.v n.vN ZN >> vmnn
DICE: .._ n.v N.n N.nN ZN >> vmnn
DICE: ... n.n n n.nN ZN >> vmnn
DICE: ... n.N N.v n.vN ZN >> vmnn
DICE: ... n v N ZN >> vmnn
DICE: .D n.n v n.nN ZN >> vmnn
DICE: .._ v.n n.n n.nN ZN >> vmnn
DICE: .._ N.v Iv n.nN ZN >> vmnn
DICE: .D n.n n ndn 2? >> vmnn
DICE: ... n.N In nn ZN m vmnn
8ch 13. £953 ASE. 95525 ASE. 55:56 5:80: 5:650 9650

 

 

 

 

 

 

 

 

.628. v 2.2..

83

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

DICE: .D n.n N Inn 22 m nvnn
DICE: In 0.0 n.vN ZN >> nvnn
DICE: N.N In INN ZN >> nvnn
DICE: n.N n N.NN ZN >> nvnn
DICED N.n n.v N.nN ZN >> nvnn
DICE: N.n n.v n.nN ZN >> nvnn
DICE: n.n N.n InN ZN >> nvnn
DICE: n.n In v.nN ZN >> nvnn
DICE: N.N v N.NN ZN >> nnnn

DICE: .._ n.v N.n NON ZN >> vnnn

DICE: ... n.N N.n n.NN ZN >> vnnn

DICE: .._ In N.v n.nN ZN >> vnnn

DICE: ... n.v In n.NN ZN >> vnnn

DICE: .._ Iv n.v IN 22 >> onnn

DICE: .._ n.. n nOn 22 m Onnn

DICE: ... n.v n.n INN ZN >> Nan

DICE: .._ n.N N.n n.vN ZN .5 Nan

DICE: .._ n.v n n.NN 22 m Nan

DICE: ... n.n n.n nOn ZN >> Nan

DICE: i. In v.v N.On ZN >> Nan

DICE: .._ n.n N.n N.nn ZN >> Nan

DICE: ... IN n.v v.N ZN >> vvnn

DICED .._ n.N n.v n.NN ZN >> vvnn

DICE: i. ON N.n n.vN ZN >> vvnn

DICE: .D n.n N.v N.nN ZN >> vvnn

DICE: ... N.N v n.nN ZN >> vvnn

owns: a. Ego; NEE. $55.25 AEE. 55:56 5:80: 5:650 .3550

 

 

 

 

 

 

 

 

6.28. v 2.2»

84

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

DICE: i. In n.n n.Nn ZN >> ann
DICE: ... n.N N.v INN ZN >> ONnn
DICED .: n.n n.n n.nn ZN >> nNnn
DICE: ..D In N.n n.nn 22 n mNnn
DICE: .._ n.n n.n v.nN ZN >> vnNn
DICE: .._ n.n n.n n.vn ZN >> vnNn
DICE: ... N n.n ON ZN >> nnnn
DICE: .._ N.N n.n n.NN ZN >> nnnn
DICE: ..D In n N.NN ZN >> nnnn
DICE: .._ n.v n.n N.NN ZN >> nnnn
DICE: .._ n v.v InN ZN >> nnnn
DICE: ... n.v IN Ion ZN >> nnnn
DICE: .._ n.N n.n v.Nn ZN n nnnn
DICED ... NON n.n N.nn 22 >> nnnn
DICE: .._ N.n n.n n.nN ZN >> vnnn
DICE: ... n.N Iv n.nN ZN >> vnnn
DICE: ... n n.n NN ZN >> vnnn
DICE: .._ n.N n.n n.nN ZN n vnnn
DICE: .._ n.n N.n n.nN ZN >> vnnn
DICE: .: N N.n Fn ZN >> vnnn
DICE: .._ n.n N.n N.Nv Z? n onnv
DICE: ..D n.N N.v v.nN ZN >> nnnn
DICE: .: v.N N.n N.nN ZN >> nnnn
DICE: .._ n.n n.n n.nN ZN >> nnnn
DICE: .._ n.v n N.nN ZN >> nnnn
DICE: ... v.N In v.nN 22 n nvnn
55:: A E 53 ~82. $556.5 ~82. 55:56 5.55: 536:8 3550

 

 

 

 

 

 

 

 

6.28. v 2.5

85

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

DICE: .m\<>0n<:-mn_D: N.n n n.nN ZN >> nnnn
DICED .m\<>o_.<:-maD: v.n n.v InN ZN >> nnnn
DICE: .m.<>0_.<:-mn_:: N.N In n.nN Z_. n nnnn
DICE: .m.<>0_.<:-wn_D: n.v n.n n.on 22 n nnnn
DICE: .m.<>0_.<:-mdD: n.n v N.Nn ZN >> nnnn
DICE: .m_\(>0q.<:-maD: n.v n.v N.nn 2.. m nnnn
DICED .m2>on<:-mn.D: n.n? v.N N.nn ZN >> nnnn
(>02: .. me: v.N N.n n. N ZN >> nnon
(>02: - mED: In In n.NN ZN >> nnon
(>0q.(: . mED: v.n N.n v.nN ZN >> nnnn
(>033 .. MED: N.n n.n IvN ZN >> nnnn
<>On<: .. MED: n.n N.n n.NN ZN >> nnnn
(>033 - mED: n.n N.n N.NN ZN >> nnon
«>02: - MED: n.n n.n INN ZN >> nnnn
(>02: - me: n.N n.n InN ZN >> nnnn
(>02: - mED: N.v n.n n.nN ZN >> nnnn
(>02: - mmD: N.v n.n Ion ZN >> nnnn
(>0_..(: . udD: n.v n ndn ZN >> nnon
(>02: - mED: n.N n.n n.vn ZN >> nnon
<>OO<: .. MED: n.nN n.N nv ZN >> nnnn
DICE: .: n.nF n.nN nv ZN >> ann
DICE: .: N.n_. N.n N.nn ZN m ann
DICE: In N.n v.nN ZN >> nnnn
DICE: N.n N.N on 2.. m nnvn
DICE: .: n.N n.n N.nN ZN >> vnvn
DICE: .._ In n.n n.nN ZN >> ann

55:: a. :55; ASE: $55.25 ASE. 55:56 5.55: 5:650 9550

 

 

.628. v 28»

86

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

EDC<E<N \ DICE: .._ v n.v n.nN ZN >> nnnn
EDC<E<N n.n n. E n.vN 22 >> ann
EDC<E<N n.n N.v N.nN ZN >> ann
EDC<E<N n.n In v.nN ZN >> ann
EDC<E<N n.v n.v n.nN ZN >> ann
EDC<E<._. N. E n.n n. Nn ZN >> ann
EDC<E<._. n.n N.n vn ZN n ann
EDC<E<N N.n_. n.nN N.nn ZN >> 3%

EDC<E<N . DICE: .: n.n In n.vN ZN >> Nonn

EDC<E<N \ DICED .: n.n v.n InN ZN n Nonn

EDC<E<._. \ DICE: .: N.n n.N N.vn ZN >> Nonn

EDC<E<N . DICE: .._ :. N.N n.nn 22 .5 Ban

EDC<E<N \ DICE: .: nN n.NN Nn ZN >> NOOn

EDC<E<N . DICE: .._ v.vF n.NN v.nn ZN m Nonn

EDC<E<N . DICED .._ n.vN In. n.nv ZN n Nonn

EDC<E<N \ DICE: .: N.vN n.nN n.Nn n>ooLO >> Nonn
EDC<E<N n: IN v.nn ZN n vnnn
EDC<E<N n.n N n.nN 22 >> nvnn
EDC<E<N n.n. NON nev ZN >> nvnn
DICE: .: n. n.N Nev 22 >> nvnn
EDC<E<N N.n N n.nN 22 >> nnnn
EDC<E<N N.n. NN n.vn ZN n nnnn

DICED .m >902: - MED: n.N N.n v.NN ZN >> nnvn

DICE: .m\<>on<:-m_n_D: v.nN n.N v.Nv ZN >> nnvn

N In N.n n.nn Z. n nnNn

DICE: .m\<>o_.<:-wn_D: n.N In n.NN ZN >> nnnn
55:: 6. E952, AEE. 85525 AEE. 55:56 5:05: 5:650 3550

 

 

 

 

 

 

 

 

Ava—.00. v Baa...

87

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

EDC<E<._. \ DICED .._ v IN nN Z? n nnnn
EDC<E<._. \ DICED .: v.n v.N Nn ZN >> nnnn
DICE: n.N n.v InN ZN >> nvnv
DICE: n.N In n.NN ZN >> ONnn
DICED In NON N.nN 22 >> ONnn
DICE: N.n v.v N.NN ZN >> 88
DICE: n.n In n.vN 22 >> ONnn
DICE: N.n N.v N.nN ZN >> ONmn
DICE: n.N n.OF N.nN ZN >> ONnn
DICED n.N N.N n.nN ZN >> ONnn
DICE: n.n n.n n.nN ZN >> ONnn
DICED n.Or n.n N.nN ZN >> ONOn
DICE: n.n n.n N.On ZN >> ONnn
DICE: n.n IN N n.on ZN >> 88
DICE: n.n n.N n.Nn ZN >> 88
DICE: NON n.n nn ZN >> ONnn
DICE: 5.9 or n.vn ZN >> ONnn
DICE: N.n N.N n.vn ZN >> 88
DICE: N. S v.n n.nn ZN >> 88
DICE: n. I n.n n.nn ZN >> ONnn
DICED n n n.Nv Z_. n ONnn
DICE: n.NN n.n Nn ZN >> ONnn
DICE: .: v.nN n.N v.vv Z? n anv
DICE: .: In n.n n.nN ZN >> :nn
EDC<E<N . DICE: .: n.n v.n N.nn ZN >> :mn
EDC<E<._. \ DICED .: n.n v.n n.nN ZN >> nnnn
55:: E 5.5 NEE. $55.25 NEE. 55:56 5.55: 5:650 3550

 

 

 

 

 

 

 

 

.628. v 2...:

88

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

EDC<E<N v.n n.v IvN ZN >> nnnn
EDC(E(._. n. v n.n N.nF ZN >> Nnnn
EDC<E<._. n.n n N.nN ZN >> Nnnn
EDC<E<._. IN v n.vN ZN n vnnn
EDC<E<._. n N.v N.nN ZN >> nnvn
EDC<E<N n. F n N .NN ZN n nnNn
EDC<E<N n.N N.v n.nN ZN >> nnNn
EDC<E<N v.N In n.n. ZN n nnnn
EDC<E<N v.n n.v n.nN ZN >> nnnn
EDC<E<N n.ON n.n n.nv ZN >> nnnn
DICE: n.N N.v n.nN ZN >> nnnn
DICE: N.nN n N.nn ZN >> nnnn
DICE: .: N n.n n.vn 22 m nnvn
DICE: .: Or N.N N.Ov ZN >> nnvn
DICED .._ n? n n.nv ZN >> nnvn
DICE: .: n.N Iv N.vN ZN >> nnvn
DICE: .._ n.ON n.n Inn ZN >> nnvn
DICE: .m IN n.n N.NN ZN >> nvnn
DICE: .m n.nN n.v n.vn ZN >> nvnn
DICE: .m n.N n.n INv ZN >> nvnn
DICE: .m v.3 n.n N.nn ZN >> n.nn
DICE: .w n In N.vn ZN n Ova
EDC<E<N n.n v.N NN ZN >> Nan
EDC<E<N . DICE: .: IvN n.n n.nn ZN >> nNNv
EDC<E<._. . DICE: .._ N.N n N ZN >> nnnn
EDC<E<._. \ DICE: .: v.N v.v n.NN ZN n nnnn
555:: a. :95; ES. 55556.5 :22. 55:56 5:55: 5:655 9555

 

 

 

 

 

 

 

 

6.28. v 32¢

89

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

DICE: .: n_. n.n Inn ZN >> van
DICE: .: n.N n.N n.nN 2? m nvnv
DICE: .: n.n n N.vN ZN >> nNNn
DICE: .: n.N n.n n.nN ZN >> Nnnn
DICE: .: v.N n.v N.vN ZN >> nvvn
DICE: .: n.N N.v n.NN ZN >> nvvn
DICE: .: n.v N.N n.nN ZN >> nvvn
DICE: .: n.N n.v n.nN ZN >> nvvn
DICE: .: v n.v N.nN ZN >> nvvn
DICE: .: NF n.N v.nv ZN m nvvn
DICE: .: nr N.nN n.nn 55695 >> NnNn
DICE: .._ n.nN Iv N.Nn ZN >> NnNn
DICE: .: n.n Iv n.nN ZN >> nnnn
DICE: .: n.n n.v n.vN ZN >> nnnn
DICE: .: n.n N.n v.nN ZN >> nnnn
DICE: .: n.NF N.n n.vv 22 m nnnn
DICE: .: n... n.n n.ON ZN >> Nnnn
DICE: .: n.n n.n N.nN ZN >> Nnnn
DICE: .: n_. n.N nv ZN .S ann
DICED .: n.v N.n n.nN 22 m nnnn
DICE: .: v.v_. n.n n.vv ZN >> nnnn
DICE: .: ..F n.n n.nn ZN >> onnn
DICE: .: In N.N n.Nv ZN >> onnn
DICE: .: v.v n.v N.ov 22 m nvnn
DICE: .: n.ON n N.vn ZN >> nvnn
EDC<E<N n.v n.n n.vN ZN m NNnn
555:: An. 2%; ES. 5552:: ES. 55256 55:55: 5:655 9555

 

 

 

 

 

 

 

 

.628. v 2%»

9O

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

DICE: .: n.n N.v n.NN ZN >> nnnn
DICE: .: N.n In Nv ZN >> nnnn
DICE: .: IN In n.Nv ZN m nnnn
DICE: .: v N.n n.nN ZN >> vnnn
DICE: .: n n.n v.NN ZN .S vnnn
DICE: .: v.v In n.Nv ZN m vnnn
DICE: .m v.n Iv n.nn 22 n anv
DICE: .m Iv n.n INn ZN >> nnnn
DICE: .m. In_. n.n v.nv ZN >> nnnn
DICE: .w v? v.v n.vn ZN >> nnnn
DICE: .w N.n v.n n.vN ZN >> FNnn
DICE: .m n.v n.v on ZN >> ann
DICE: ...... N.NF N.n N.nv Z. n FNnn
EDC<E<._. \ DICE: .._ n.v v.nN ZN >> nnvn
DICE: .: n.N N.n n.NN ZN >> nvnv
DICE: .: n.n N.n v.Nn ZN m nvnv
DICE: .: n.v n.n N.nN 22 m onnn
DICE: .: n.n N.v N.Nn ZN >> onnn
DICE: .: n n.n n.Nn ZN m onnn
DICE: .: N.v En Inv 22 n onnn
DICE: .: n.n In n.nn ZN m onnn
DICE: .: n.O_. v n.Nv ZN >> van
DICE: .: n In N.Nv ZN m van
DICE: .: NF n.v N.nv ZN >> van
DICE: .: Inv n.n n.nv ZN >> van
DICE: .: N.n n.n n.nn 22 n van
55:6 8: 5%? EE. 5555:. ES. 55:56 55:55: 526:8 no.555

 

 

 

 

 

 

 

 

.628. v 53.:

91

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

DICE: .._ n.ON N N.Ov ZN n nNnn
EDC<E<N N.v n n.nN ZN >> nnnn
EDC<E<N n.v N.n n.nN ZN >> nnnn
EDC<E<N n.nN n.nN n.Nn ZN >> nnnn

N v.N v.v N.NN ZN n NNnn

N N n.n EON ZN >> onvn

N N.n N.n n.nn ZN m onvn

N n N.v n.Ov ZN m onvn

N n. 2 En n.nv ZN >> onvn
EDC<E<N v.N n.v n.nN ZN >> vnnn
EDC<E<N In n.v InN ZN >> vnnn
EDC<E<N N.O_. n.n Inn ZN >> vnnn
EDC<E<._. . DICE: .: N.n: N.N n.NN ZN >> Nnvn
DICE: .._ vN n.nN n.nn ZN >> Fonn
DICE: .: n.nN n.n Inv ZN >> nvnn
DICE: .: N.nN n.n n.nv ZN >> nenn
DICE: .._ nN n.n n.Nn ZN >> nvvn
DICE: .: v.vN N.n n.nn ZN n nvvn
DICE: .: In n.n Inn ZN >> nnvn
DICE: .._ v.9 In n.Nv ZN >> nnvn
DICE: .: n.N n.n N.nN ZN >> ovvn
EDC<E<N \ DICE: .: N.N n.n v.Nn 22 n nvvn
EDC<E<._. \ DICE: .: n.n In v.nn ZN >> nvvn
DICE: .._ n n.n n.nN ZN n vnnn
DICE: .: N.N n.n n.nN ZN n vnnn
DICE: ... n.v n.v on ZN >> vnnn

555:: An. :95; ES. 5555.53. NEE 55:56 55:55: 5:6:55 n5555

 

 

 

 

 

 

 

 

.628. v 2.2:.

92

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

EDC<E<._. nn 3 nn 55.695 >> n Nnv
EDC<E<N N.n n.v N.nN ZN >> nnnv
EDC<E<N . DICE: .._ N.n n.n n.nv 22 n Ovnv
EDC<E<N N.vN n.n nv ZN >> NNnv
EDC<E<N n.n n n.nn ZN >> NNnv
DICE: .m n.Nn n.ON n._.n Z_. >> NNnv
DICE: .._ In n.n nn .558 >> nvnv
DICE: .: In n.v N.NN ZN >> nNnv
EDC<E<N EON n.n N.Nn ZN n vNNv
N v.n n.n n.nn 22 n nNNv

N v.NN n.n n.nn ZN m nNNv

N n.v N.v N.Nn 22 n NN_.v
DICE: .._ n: n.n ov 22 >> Nnnn
EDC<E<N N.n n.v INv ZN >> nnnn
N n.n In v.nn ZN >> vnnn
DICE: .: v.N n.v n.ON ZN >> nnnn
DICE: .._ IN n.n n.NN ZN >> nnnn
DICE: .: v n.n InN ZN >> nnnn
EDC<E<N N.n N.v n.vN ZN >> NNnn
DICE: .._ n.N n.v n.nN ZN >> nNnn
DICE: .: n.n n.n n.Nn ZN m nNnn
DICE: .: I: In nn ZN >> nNnn
DICE: .: n.N n.n N.nn ZN n nNnn

555:: An. 5952, EE. 5555.25 «EE. 55:56 55:55: 5.655 .5555

 

 

AFN—.05. v 535...

93

ILLUSTRATIONS

94

 

 

_ a 5i

    

ciautoaucem—

lli/riu. x

 

_
. RV:

.. ......
EM:

 

 

   

  

95

 

3:2. a 33 55.3 .22... 5255 .392.

 

=— 3955

EEK...

 

N

3.5.;
513525855:

1

 

53:... 33.55

 

use an .

a a. my.

 

 

 

 

 

 

 

 

Figure 1. Lake Patzcuaro and surrounding sites

 

Figure 2. Prehispanic Tarascan disks

96

555.

 

on

Pn

Nv

EEC:
nv on nn Nn NN

nN

aw

nr

 

 

E55555 NEE. 55:55

. o
r n
- o.
-- 5. W
.- on m
.- 55 m.
... m
i- on a m
b ..m
it mm n k
«v .m 7.
i 0.? w W 9
.A m
it mV 3
m.
-- on ..m
., ...... m
e
n- on me
-- 55 r
E

 

Diameter (mm)
Mean 30.71013
Standard Error 0.397192
Median 28.2
Mode 25.2
Standard Deviation 9.083465
Sample Variance 82.50934
Kurtosis 0.253698
Skewness 0.888559
Range 44.7
Minimum 14.9
Maximum 59.6
Sum 16061.4
Count 523

 

Figure 4. Diameter summary statistics

98

 

 

ex ex. ow. ex. ex 01 e « e 2

ii
0
v

1T
0
(D

i
o
co

i 00—.

 

ONF

 

EEmoamE AEEV mmoconh

Aouanbau

Figure 5. Histogram of disk thickness (all disks)

99

Thickness (mm)

 

Mean 5.700574
Standard Error 0.098527
Median 5.1
Mode 5.2

Standard Deviation 2.253235
Sample Variance 5.077069

Kurtosis 9.199168
Skewness 2.45341 1
Range 16.8
Minimum 2.7
Maximum 19.5
Sum 2981.4
Count 523

 

Figure 6. Thickness summary statistics

100

 

 

mEEG

ea 81 ex 01 01

(

l

J

l

i ON
.. ov
- cm
i om
.. 03
1 ON?
1 o:
l o9.

 

 

29385 :3 2925

o9.

Aouenbeu

Figure 7. Histogram of disk weights (all disks)

101

Weight (9)

Mean 6.209962
Standard Error 0.217892
Median 4.2
Mode , 2.4

Standard Deviation 4.97825
Sample Variance 24.78297

Kurtosis 4.447671
Skewness 1.867175
Range 34.6
Minimum 0.4
Maximum 35
Sum 3241.6
Count 522

 

Figure 8. Weight summary statistics

102

:55 .8255

 

 

 

 

  

 

 

 

 

 

 

 

£925 a 365.02... .m> 388.20

mm vm om ow Nv mm em on mm «N we 3.
_ — . _ _ _ _ — . if _ o
O O 00. 0 I O IHIIIHI OWJI‘J“._Q‘H..» Lin-m
0 i n . ...... I. r
. .. .. . ......r33 .33.. Name! ..
O l 0.0 O O ‘ O I i O 0
w" ..." l 0 Q o o c. 0' IO? no I” o. o
8.m+§8o.onl> . . . .. . u... . ..... . . ... . . m
I. o I o I 5Q. an. o I 09 o
. -. twain” 0.. u. l NF
- . . . . . a . . . 2
. omeonam . . .
. I . I. I o C I
.303 x9331». om
. . .- 4.0.
mm
. mm
mm

 

mmmconh 50:: ll
Em_w>> .85.. I
as 5925 .
AEEV wmoconh .

Figure 9a. Regression plot of disk diameter vs. thickness and weight

The image is presented in color in the thesis.

103

 

 

 

diameter (mm) thickness (mm) wei ht

 

diameter (mm) 1
thickness (mm) 0.281014001 1
weight (g) 0.788038538 0.645722262 1

 

Figure 9b. Pearson correlations for variables of diameter, thickness, and weight

104

 

 

ZN
._._<D

Zr
.34..

22
.._._<

 

 

 

 

.935 :5 3229...: 582

Figure 10. Notch morphology (all disks)

105

 

Axemm

£0:

Ema 225 322922 862

   

Figure 11. Notch morphology (whole disks)

106

 

A965 cmxemv 3205922 5202

Figure 12. Notch morphology (broken disks)

107

 

ocean.

 

Rum»... ocou >255
Euaoaux

 

 

 

 

 

352.5

 

 

 

cow

 

 

omw

 

 

com

 

 

 

omN

 

 

vmm

coo

 

 

0:0N >023m 3a 9.2a

0mm

Figure 13. Disks per survey zone

108

8%. __m

 

mam

 

ocean. ocmaomumx

 

 

3

 

 

 

 

 

 

 

 

=65

 

So...

 

lllll com

.. oov

- cow

- com

, coo;

, com;

lll- Gov;

r 000;.

 

cow;

 

x30 .85 £9.26 2:530 Co 03am

Figure 14. Ratio of ceramic sherds per disk
109

 

  
 

MBUCO

Murco City
I I ‘
Sam at 1:

Human

 

 

   

Areas ol recent
alluvial / colluwal till

 

  
 

Yununn O

  
   

Jan-Ilia Q

PATZCUARO

 

  

o "In
’ .__._____.__—n——.——'
/ZIRAHUEN

 

 

 

Figure 15. Map of Lake Patzcuaro showing exposed lake bed due to lowered
water levels (from O’Hara 1993)

110

 

”3 JW .

I = modern villages (not surveyed or excavated)

I = units with high circle counts

Pa reo
0 L\’\

Figure 16. Map of Xaracuaro units with the highest disk counts

/

/. Lake Pfitzeuaro"
/
x".
Y

4 aricna ro

   

/ /- / / 7 N/
/"I ,/ . . -. fl}, ‘
I/z’ / .. . T/fi/l'

   
  

 

   
    

VII . a modern villages (notsurveyed or excavated)
a? a - units with high circle counts "

 

Figure 17. Map of Pareo units with the highest disk counts

112

 

N

/'

VJ- I\ ‘
"/’

   
 

   

   
 

 

I - modern village: (not surveyed or excavated)

E - units mth circle- c?

%
Figure 18. Map of Urichu units with the highest disk counts

113

omega .2. 935

DICE: <>Ow
_m30<_m<._. DICE: DICE: m._.<._ >._m<m (..-MnS...

     
 

Om<D O<m<w

 

 

om

 

 

00—.

 

 

one

 

 

 

CON

 

wvw

 

omN

 

 

AmoEON >c>5m =5 emu—E .3 935

com

Figure 19. Disks per phase (all survey zones)

114

 

EDO<_K<._.

DIUED

o

if

8!... ..on 83.0
DICE: m._.<._ DICE: >.E<m «$0.. 5 - mmDD om<Do<m<a

m A

 

 

 

 

 

 

or

 

0N

mN

 

 

 

 

mm

 

35.5. 82E .2. 935

Figure 20. Disks per phase (Urichu)

115

EDD/«Etc. DIOED

   

 

 

 

 

 

32E .2. 83.0

DIOED m._.<._ DIOED >Dm<m <>Ow 5 - meD Om<Do<m<w

, a

.

 

0N

 

ov

 

 

om

 

 

on?

 

OE.

 

 

cor

 

 

mt

 

 

Ago—5880 82E ..oa 9.20

CON

Figure 21. Disks per phase (Xaracuaro)

116

 

 

Figure 22. Map of Xaracuaro units with highest Late Urichu phase disk counts

117

 

Figure 23. Map of Xaracuaro units with highest Tariacuri phase disk counts

118

own—E ..on 935

EDD/DE... DIOED m._.<.._ DIOED >Dm<m

DIOED

   

 

—

(>9. <4 - mmDD

0

Li

Om<D 05$.

0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

on

 

 

 

398$ 32... ..on. 8.20

or

on

on

0v

om

00

on

on

Figure 24. Disks per phase (Pareo)

119

 

 

/
,\_‘\
w
R
x (
\\

 

 

 

 

Figure 27. Prehispanic spinning (from Florentine Codex)

122

 

Fig. 2. Shows the size of a
cotton whorl (left) as compared
to a maguey whorl (right).

 

Figure 28a. Prehispanic malacates from central Mexico (from Parsons and Parsons 1990)

 

 

 

 

Figure 28b. Prehispanic malacates from Coj umatlén (from Lister 1949)

124

 

 

415-34

   

 

 

 

Figure 28c. Prehispanic malacates from Zacapu Lake Basin, Michoacan (from Arnauld et
al. 1993)

125

t-Test: Two-Sample Assuming Unequal Variances

 

Variable 1 Variable 2

 

Mean 30.69483 19.35
Variance 82.54494 28.66944
Observations 522 1 0
Hypothesized Mean Difference 0
df 10
t Stat 6.522735
P(T<=t) two-tail 6.7E-05
t Critical two-tail 2.228139

 

t-test for diameter: Tarascan disks vs. Tzintzuntzan
malacates

reject null hypothesis

Figure 29. T-test for diameter: Tarascan disks vs. Tzintzuntzan malacates

126

t-Test: Two-Sample Assuming Unequal Variances

 

 

Variable 1 Variable 2
Mean 5.699425287 10.1315789
Variance 5.08612251 23.495614
Observations 522 1 9
Hypothesized Mean Difference 0
df 18
t Stat -3.97003161
P(T<=t) two-tail 0.000897804
t Critical two-tail 2.100923666

 

t—test for thickness: Tarascan disks vs. Tzintzuntzan

malacates

do not reject null hypothesis

Figure 30. T-test for thickness: Tarascan disks vs. Tzintzuntzan malacates

127

t-Test: Two-Sample Assuming Unequal Variances

 

Variable 1 Variable 2

 

Mean 30.694828 23.10526316
Variance 82.544944 25.0994152
Observations 522 1 9
Hypothesized Mean Difference 0
df 23
t Stat 6.2403675
P(T<=t) two-tail 2.288E-06
t Critical two-tail 2.0686548

 

t-test for diameter: Tarascan disks vs. Urichu malacates

reject null hypothesis

Figure 31. T-test for diameter: Tarascan disks vs. Urichu malacates

128

t-Test: Two-Sample Assuming UnequalVariances

 

Variable 1 Variable 2

Mean 5.6994253 10.1315789
Variance 5.0861225 23.495614
Observations 522 1 9
Hypothesized Mean Difference 0
df 18
t Stat -3.970032
P(T<=t) two-tail 0.0008978
t Critical two-tail 2.1009237

 

t-test for thickness: Tarascan disks vs. Urichu malacates

do not reject null hypothesis

Figure 32. T-test for thickness: Tarascan disks vs. Urichu malacates

129

t-Test: Two-Sample Assuming Unequal Variances

 

Variable 1 Variable 2

 

Mean 6.2017274 5.1470588
Variance 24.79517 5.2388971
' Observations 521 17
Hypothesized Mean Difference 0

df 21

tStat 1.76822

P(T<=t) two-tail 0.0915506

t Critical two-tail 2.0796142

 

t-test for weight: Tarascan disks vs. Urichu malacates

do not reject null hypothesis

Figure 33. T-test for weight: Tarascan disks vs. Urichu malacates

130

t—Test: Two-Sample Assuming Unequal Variances

 

Variable 1 Variable 2

 

Mean 23.10526 19.35
Variance 25.09942 28.669444
Observations 19 . 10
Hypothesized Mean Difference 0
df 17
t Stat 1.83501 1
P(T<=t) two-tail 0.084064
t Critical two-tail 2.109819

 

t-test for diameter: Urichu malacates vs. Tzintzuntzan
malacates

do not reject null hypothesis

Figure 34. T-test for diameter: Urichu malacates vs. Tzintzuntzan malacates

131

t-TeSt: Two-Sample Assuming Unequal Variances

 

 

Variable 1 Variable 2
Mean 10.131579 11.2
Variance 23.495614 6.0111111
Observations 19 10
Hypothesized Mean Difference 0
df 27
t Stat -0.7881384
P(T<=t) two-tail 0.437481
t Critical two-tail 2.0518291

 

t-test for thickness: Urichu malacates vs. Tzintzuntzan

malacates

do not reject null hypothesis

Figure 35 . T-test for thickness: Urichu malacates vs. Tzintzuntzan malacates

132

Figure 36. Drilled stone disks from Lovelock Cave, Nevada

(from Loud and Harrington 1929)

133

i
i
.
l
, .
1

 

 

Figure 37. Fish nets from Lovelock Cave, Nevada (from Loud and Harrington 1929)

134

F“ . .
lrizgueufl -

‘L E.

.1

 

Figure 38. Net sinkers from Port Moller, Alaska (from Weyer 1930)

135

 

Figures 39a—b. Notched stones from Cook Inlet, Alaska (from De Laguna 1934)

136

 

 
 

.--.n..- -....l- '

 

Figures 40a-b. Notched stones from central Texas (from Watt 193 8)

137

References Cited

Anbnimo
1989 . Relacio’n de Michoacan (edicion de Leoncio Cabrero). Historia 16, 1989,
Madrid. _

Amaud, Charlotte, Patricia Carot, and Marie-France Fauvet-Berthelot
1993 Arqueologia de las Lamas en la cuenca lacustre de Zacapu, Mchoacan,
Mexico. Centre d’Etudes Mexicaines et Centramericaines, Mexico.

Beals, Ralph L. .
1932 The Comparative Ethnology of Northern Mexico Before I 750. University
of California Press, Berkeley.

Beals, Ralph L. and Pedro Carrasco
1944 “Games of the Mountain Tarascans.” American Anthropologist 46:4(516-
522)

Bennett, Wendell C. and Robert M Zingg
1976 The : an Indian Tribe of Northern Mexico. Rio Grande Press, Inc., New
Mexico.

Bigger, H.P., editor
1929 The Works of Samuel de Champlain. Volume III, 1615-1618. The
Champlain Society, Toronto

Brand, Donald D.
1951 Quiroga: A Mexican Municipio. United States Government Printing
Office, Washington DC.

Broadbent, Noel D, Goran Burenhult, and Moreau Maxwell
1993 “Why Only Some Became Farmers,” in People of the Stone Age, Volume
Two, H unter-gatherers and Early Farmers. American Museum of Natural
History and HarperCollins.

Campbell, Joseph
1989 Historical Atlas of World .Mythology. Volume II The Way of the Seeded
Earth. Part 3 Mythologies of the Primitive Planters: The Middle and

Southern Americas. Perennial Library Harper & Row, New York.

Cleland, Charles Edward
1966 The Prehistoric Animal Ecology and Ethnozoology of the Upper Great
Lakes Region. Anthropological Papers No. 29, Museum of Anthropology,
University of Michigan, Ann Arbor.

138

Coe, Michael D.
1984 Mexico (third edition). Thames and Hudson Ltd., London.

Craine, Eugene R. and Reginald C. Reindorp, translators and editors
1970 The Chronicles of Michoacdn. University of Oklahoma Press, Norman.

Custer, Jay F.
1996 Prehistoric Cultures of Eastern Pennsylvania. Pennsylvania Historical
and Museum Commission Anthropological Series Number 7, Harrisburg.

De Laguna, Frederica
1934 The Archaeology of Cook Inlet, Alaska. University of Pennsylvania Press,
Philadelphia.

De Montellano, Bernard R. Ortiz
I990 Aztec Medicine, Health, and Nutrition. Rutgers University Press, New
Brunswick and London.

Ensminger, Audrey H., M. E. Ensminger, James E. Konlande, and John R. K. Robson
1995 The Concise Encyclopedia of Foods and Nutrition. CRC Press, Boca
Raton.

Foster, George M.
1948 Empire 19 Children: The People of T zintzuntzan. Smithsonian Institution
Institute of Social Anthropology, Publication No. 6.

Freidel, David, Linda Schele, and Joy Parker
1993 Maya Cosmos: Three Thousand Years on the Shaman 's Path. William
Morrow and Company, Inc. New York.

Gorenstein, Shirley and Helen Perlstein Pollard
1983 The T arascan Civilization: A Late Prehispanic Cultural System.
Vanderbilt University Publications in Anthropology, No. 28., Nashville,
Tennessee.

Guzman, Ana Fabiola, Oscar J. Polaco, and Helen P. Pollard
2001 Ofrendas de Peces Asociadas a Entierros del Clasico-Epiclasico en
Urichu, Michoacan, Mexico. Archaeofauna 10(2001):149-162.

J enness, D.
1935 The Ojibwa Indians of Party Island, Their Social and Religious Life.
National Museum of Canada, Bulletin Number 78, Anthropological Series
Number 17. Ottawa.

Joaquin, Pedro Marquez
2000 Relacz'on de Michoacan. Colegio de Michoacan.

139

Kelly, Isabel
1947 Excavations at Apatzingan, Michoacan. Viking Fund Publications in
Anthropology, number seven. New York.

Kubler, George
1967 “The Iconography of the Art of Teotihuancan,” in Studies in Pre-
Columbian Art and Archaeology 4. Dumbarton Oaks, Washington.

Lister, Robert H.
1949 Excavations at Cojumatla’n, Michoacan, Mexico. University of New
Mexico Publications in Anthropology, University of New Mexico,
Albuquerque.

Loud, Llewellyn L. and M. R. Harrington
1929 Lovelock Cave. University of California Publications in American
Archaeology and Ethnology. University of California Press, Berkeley,
Califomia/ Cambridge University Press, London.

Macias Goytia, Angelina
1990 Huandacareo: Lugar de Juicios, Tribunal. Coleccion Cientifica, Instituto
Nacional de Antropologia e Historia, Mexico.

Meehan, William E.
1893 F ish, Fishing and Fisheries of Pennsylvania. E.K. Meyers, Harrisburg,
Pennsylvania.

Moratto, Michael
1984 California Archaeology. Academic Press, Orlando, Florida.

Nicholson, H.B.
1976 “Preclassic Mesoamerican Iconography,” in Origins of Religious Art and
Iconography in Preclassic Mesoamerica. H.B. Nicholson, editor. UCLA
Latin American Publications and Ethnic Arts Council of Los Angeles.

Niederberger, Christine
1987 Paleopysages et Archeologie Pre- Urbaine du Bassin de Mexico
(Mexique). Etudes Mesoamericaines, volume XI, tome 1. Centre d’études
Mexicaines et Centramericaines, Mexico.

O’Hara, Sarah L.
1993 Historical Evidence of Fluctuations in the Level of Lake Patzcuaro,
Michoacan, Mexico Over the Last 600 Years. The Geographical Journal
lS9(l):Sl-62.

140

Ortiz de Montellano, Bernard R.
1990 Aztec Medicine, Health, and Nutrition. Rutgers University Press, New
Brunswick.

Oswalt, Wendell H.
1988 This Land Was Theirs: A Stuay of North American Indians. Fourth
edition. University of California, Los Angeles.

Panofsky, Erwin
1960 Renaissance and Renascences in Western Art. Almqvist and Wiksells,
Stockholm.

Parsons, Jeffrey R. and Mary H. Parsons
1990 Maguey Utilization in Highland Central Mexico: An Archaeological
Ethnography. Anthropological Papers, Museum of Anthmpolo‘gy,
University of Michigan. No. 82. Ann Arbor.

Parsons, M. H.
1972 Spindle Whorls from the T eotihuacan Valley, Mexico. Anthropological
Papers No. 45, University of Michigan Museum of Anthropology, Ann
Arbor.

1975 The Distribution of Late Postclassic Spindle Whorls in
the Valley of Mexico. American Antiquity 40(2)207-15.

Pennington, Jean A. T.
1998 Bowes & Church ’3 Food Values of Portions Commonly Used, 1 7'”
Edition. Lippincott Williams and Wilkins, Philadelphia

Pohorecky, Z. S.
1976 Archaeology of the South Coast Ranges of California. Berkeley:
Contributions of the University of California Archaeological Research
Facility 34:1 -235.

Pollard, Helen Perlstein
1993 T ariacuri ’s Legacy: The Prehispanic Tarascan State. University of
Oklahoma Press, Norman and London.

Pollard, Helen Perlstein and Laura Cahue
1999 Mortuary Patterns of Regional Elites in the Lake Patzcuaro Basin of
Western Mexico. Latin American Antiquity 10(3):259-280.

Ritchie, William A.

1944 The Pre-Iroquoian Occupations of New York State. Rochester Museum of
Arts and Sciences, Rochester, New York.

141

Ronzio, Robert A.
1997 The Encyclopedia of Nutrition and Good Health. Facts on File, New
York.

Rostlund, Erhard
1952 Freshwater Fish and Fishing in Native North America.- University of
California Press.

Schele, Linda and David Freidel
1990 A Forest of Kings: The Untold Story of the Ancient Maya. William
Morrow and Company, Inc. New York. '

Schoolcrafi, Henry R.
1852 Information Respecting the History, Condition and Prospects of the Indian

Tribes of the United States, Part 11. Lippincott, Grambo & Company,
Philadelphia.

Sugiura, Yoko
1998 La Caza, la Pesca y la Recoleccion: etnoarqueologia del modo de
subsistencia lacustre en las cienegas del Alto Lerma. UNAM.

Smith, William Charles
1965 The People of La Pacanda: Social Organization and Social Change in a
T arascan Village. PhD dissertation, University of California, Berkeley.

Toledo, Victor M.

1991 “Patzcuaro’s Lesson: Nature, Production, and Culture in an Indigenous
Region of Mexico,” pp. 147-171 in Biodiversity: Culture, Conservation
and Ecodevelopment, M. Oldfield and J. Alcom, editors. Westview Press,
Boulder.

Toledo, Victor M, Pedro Alvarez-lcaza, and Patricia Avila, editors
1993 Plan Patzcuaro 2000: Investigacion multdiscrplinaria para el desarrollo
sostenido. Fundacion Friedrich Ebert, Stiftung, Mexico.

Wallace, W.J., and D.W. Lathrap
1975 West Berkeley (CA-A1a-307): A culturally stratified shellmound on the
east shore of Dan Francisco Bay. Berkeley: Contributions of the
University of California Archaeological Research Facility 29.

Watt, Frank H.
1938 The Waco Sinker. Central Texas Archaeologist no. 4:21-70.

Webb, W.S. and W.G. Haag

1939 The Chiggerville Site. University of Kentucky Reports in Anthropology,
Volume IV, Number I.

142

Wenke, Robert J.
1990 Patterns in Prehistory: H umankind ’s First Three Million Years. Third
edition. Oxford University Press.

Weyer, Edward Moffat
1930 Archaeological Material from the Village Site at Hot Springs, Port Mailer,
Alaska. Anthropological Papers of the American Museum of Natural
History, Volume XXXI, Part IV. New York City.

Whallon, Robert
1987, “Simple Statistics.” In Quantitative Research in Archaeology: Progress
and Prospects. M.S. Aldenderfer, editor. Newbury Park, California.
Sage.

143

ll1111111111111111

3O